Page 1

J149 Achieve Advert 2

24/6/09

15:48

Page 1

tion a c u Ed k Expertise r o tw ork e w t N Ne

em va en Innova p t tio x n o i n E cat u Ed

unique Europe wide educational program to promote long-term excellence in hemophilia training

t

ion

is t r e

achieve:

Comm ent m it m t i m m I o Achi nn en C ev o e

t

Achievement

The achieve program has two key objectives: To offer a comprehensive educational program for young physicians which combines a theory course with an individualized practical training To build a pan-European network with complementary interests and expertise A prestigious panel of key European experts in hemophilia treatment will choose a group of physicians each year Chaired by Pier Mannucci and Elena Santagostino, the achieve scientific committee includes: Johannes Oldenburg (Bonn), László Nemes (Budapest), Claude Négrier (Lyon), Erik Berntorp and Rolf Ljung (Malmö), Luigi Solimeno (Milan), Paul Giangrande (Oxford), and Johanna van der Bom and Kathelijn Fischer (Utrecht).

Applic ati

Achievem

ent Education rk Expert two i en s e e N twork Ne n

uca Ed

tion

ev n em I n no va tio

C Commitment ommi Ach I ise n t i

nt n meovattio

pe r t Ex

To learn more about the achieve program, and how to apply, please contact your local Bayer Schering Pharma representative or send us an email at the following address: achieveapplications@atriumhealth.com.

on Guid

e 2009/201

0


JOURNAL OF COAGULATION DISORDERS (JCD)

October 2009

Vol 1, No 1

THE

THE EDITORS AND PUBLISHERS OF JOURNAL OF COAGULATION DISORDERS (JCD)

Owned and Published by San Lucas Medical Limited John Gault, MD, Editor-in-Chief

EDITORIAL BOARD Alessandro Gringeri, Italy

Georges E. Rivard, Canada

Michael D. Tarantino, USA

Brigit Brand, MD, Switzerland Bryce A. Kerlin, USA

Henri M. Bounameaux, Switzerland

Peter Collins, United Kingdom

Charles R Hay, United Kingdom Christopher Ludlam, United

Hugo ten Cate, the Netherlands Johannes Oldenburg, Germany

Prasad Mathew, USA

Kingdom

John Pasi, United Kingdom

Ralph A. Gruppo, USA

David Bergqvist, Sweden Deborah Brown, USA

Keith R. McCrae, USA Kim L. Spencer, USA

Sam Schulman, Canada

Eduardo Remor, Spain Frank E. Shafer, USA

Louis M Aledort, USA Mark Makris, Sheffield

Victor Jime´nez-Yuste, Spain

Journal of Coagulation Disorders editor@slm-journals.com

Pier Mannucci, Italy Raed Al Dieri, the Netherlands

Thynn Thynn Yee, UK

(JCD)

Copyright g 2009 San Lucas Medical Limited

www.slm-hematology.com

Print ISSN 2041-7969


JOURNAL OF COAGULATION DISORDERS (JCD) PUBLISHING INFORMATION Journal of Coagulation Disorders (JCD) [Print ISSN 20417969] is published semiannually (February/March and October), by San Lucas Medical, Ltd. Copyright 2009 by San Lucas Medical, Ltd.

Scientific Review and Acceptance The JCD publishes original research manuscripts, reviews, editorials, correspondences and special articles on topics such as treatment guidelines. The focus of The Journal of Coagulation Disorders is on original reports, but other highquality work that relates to haematology may be considered. The Journal of Coagulation Disorders is distributed in print and electronic form across Europe to thousands of physicians, researchers, academics, nurses and related care practitioners with an interest in coagulation disorders. Both subscription and access are free and there are no author fees for publication. Submitted manuscripts are reviewed by members of the JCD editorial board by a double-blind peer review process with the understanding that manuscripts have not been submitted for publication elsewhere. Papers are accepted for publication based on originality, quality and relevance to the audience.

Manuscript Submission All prospective authors must submit their manuscripts electronically; a manuscript may be e-mailed directly to the editorial team (editor@slm-journals.com) or submitted online (http://www.slm-hematology.com). An acknowledgement of receipt will be sent to the author within 48 hours. Prior to submitting a manuscript, authors are encouraged to subscribe to the journal, as having a subscription will allow us to better track manuscripts. This will also allow us, once your manuscript is published, to quickly send you a copy of the journal and proofs.

Manuscript Lead Times Authors should expect to receive a first decision with reviewers’ comments approximately four weeks after submission.

Instructions for Authors Manuscripts should be prepared in concise, clear, and grammatically correct English. They should contain the following: a title page; a short letter transferring copyright to San Lucas Medical Limited if the article is accepted for publication; an abstract of up to 250 words; and up to 7 keywords for indexing purposes. Abstracts are also required for review articles. Text should contain the following: introduction, results, discussion, materials and methods, acknowledgments and references, figure legends, footnotes, tables (with descriptive titles and legends), conflicts of interest, and a disclosure of funding.

Style & Format MS Word (version 2.0 and above), Word Perfect (version 5.1 and above) and RTF are the preferred formats to receive submissions.

Referencing Manuscripts must use the Vancouver (author-number) style of referencing (i.e., a number series to indicate references). The bibliography must then list these references in the numerical order in which they appear in the text.

Images & Tables Tables and charts should complement, rather than duplicate, the text. Photographs and illustrations must be the following: supplied in high resolution, at least 300 dpi, JPEG/Tiff format, and CMYK or grayscale.

Fees There are no author fees.

Free Subscriptions Access to The Journal of Coagulation Disorders is free. To obtain a free subscription, please send your contact details to editor@slm-journals.com. Please remember to include your full name, title, medical speciality, postal address and email address. Alternatively, you can subscribe online at www.slmhematology.com

Correspondence The Journal of Coagulation Disorders is committed to allowing a range of opinions to be expressed and publishes appropriate reader correspondence. To comment on a published article, please send a letter to the editor to editor@slmjournals.com

Copyright and Permissions Any use of this journal must include the customary bibliographic citation that includes author attribution, date and article title. All articles are made available and publicly accessible via the Internet and are free to use for noncommercial purposes. San Lucas Medical offers a reprint service for individuals who require professional quality reproductions of articles.

Advertising For price quotes and to purchase display advertising, please contact Jayden Rai (jayden.rai@slm-journals.com). Artwork must be submitted as high-resolution digital media with a minimum of 300 dpi. Preferred formats are as follows: PDF/Acrobat 4.05 or later; Adobe Illustrator (with all fonts outlined); Adobe Photoshop EPS, TIFF or JPEG (CMYK and 300 dpi): and Adobe InDesign (packaged with all links). Please send all artwork electronically to editor@slmjournals.com.

Disclaimer Every precaution has been taken by the authors, editors and the board to verify the results and data published in The Journal of Coagulation Disorders (JCD). However, the opinions expressed are the responsibility of the contributors, and not necessarily those of the JCD editorial team or board. As a result, JCD accepts no liability for the consequences of any inaccurate or misleading data, statement or opinion. Please write to the editor if you identify an error.


JOURNAL OF COAGULATION DISORDERS (JCD) CONTENTS

October 2009 – Vol 1, No 1

ORIGINAL ARTICLES Traffic of rFVIIa through Endothelial Cells and Redistribution into Subendothelium: Implications for a Prolonged Hemostatic Effect I. Lopez-Vilchez, J. Tusell, U. Hedner, C. Altisent, G. Escolar and A.M. Galan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 Dental and Periodontal Health in Children with Hemophilia E. Alpkılıc¸ Baskirt, H. Albayrak, G. Ak, A. Pınar Erdem, E. Sepet and B. Zulfikar . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7

REVIEW ARTICLES Gastrointestinal Angiodysplasia and Acquired Von Willebrand Syndrome: A Review of an Enigmatic Association Renu Saxena and Prashant Sharma. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 Renal Hemophilic Pseudotumors Emel Gu¨rkan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 Prophylaxis in Patients with Severe Hemophilia and Inhibitor Victor Jimenez-Yuste, Maria Teresa Alvarez, Monica Martin-Salces, E. Carlos Rodriguez-Merchan, Nora Butta, Ihosvany Fernandez-Bello, Isabel Rivas and Ana Rodriguez de la Rua . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 Safety of Radiosynovectomy in Hemophilic Synovitis: it is Time to Re-evaluate! Cuneyt Turkmen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 Immunotolerance Induction Treatments in Hemophilia Saturnino Haya, Pilar Casan ˜ a, Andre ´ s Moret, Ana R. Cid, Noelia Cabrera, Lydia Abad and Jose ´ A. Aznar . . . . . . . . . 37 An Overview of Blastocystis hominis Infection and Published Experience in Hemophilic Population Carlos Aguilar and Jose ´ F. Lucı´a . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 Rituximab in the Treatment of Postpartum Acquired Hemophilia A Maria Gabriella Mazzucconi, Francesca Biondo and Cristina Santoro . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47 Acquired von Willebrand Syndrome in Childhood and Adolescence Andrew M. Will . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55 Cardiac Surgery and Percutaneous Coronary Interventions in Patients with Hemophilia B: an Overview of Published Reported Cases Elizabeth F. Krakow, Ruslan Ganchev and Julia A.M. Anderson . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65 von Willebrand Factor Proteolysis by ADAMTS13 Derrick J. Bowen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75 Central Venous Access Devices (CVAD) for Pediatric Patients with Hemophilia: A Review Riten Kumar, Rajiv K. Pruthi and Vilmarie Rodriguez. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85 Pharmacoeconomic studies of bypassing agents in mild-to-moderate bleeding episodes in patients with hemophilia and inhibitors: a critical appraisal Erich V. De Paula and Margareth C. Ozelo . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93

Journal of Coagulation Disorders editor@slm-journals.com

(JCD)

Copyright g 2009 San Lucas Medical Limited

www.slm-hematology.com

Print ISSN 2041-7969


JOURNAL OF COAGULATION DISORDERS (JCD) SUBSCRIPTIONS To obtain your FREE subscription to the journal please complete and return the form below. First Name _____________________________________

Last Name ______________________________________

Job Title ________________________________________

Primary Specialisation __________________________

Main Hospital/Clinic ____________________________

e-Mail ___________________________________________

Postal Address Line 1 __________________________________________________________________________________ Postal Address Line 2 __________________________________________________________________________________ City _____________________________________________

County/State/Province __________________________

Postcode/Zip ____________________________________

Country _________________________________________

Subscription preference (print or electronic):

C

Print

C

Electronic

Refer a colleague (please enter their details) First Name _____________________________________

Last Name ______________________________________

Job Title ________________________________________

Primary Specialisation __________________________

Main Hospital/Clinic ____________________________

e-Mail ___________________________________________

Postal Address Line 1 __________________________________________________________________________________ Postal Address Line 2 __________________________________________________________________________________ City _____________________________________________

County/State/Province __________________________

Postcode/Zip ____________________________________

Country _________________________________________

Subscription preference (print or electronic):

C

Print

C

Electronic

Remove and fax this form to +44 (0) 20 7084 0330 or an electronic version, that can be completed can accessed from www.slm-hematology.com or e-mail the completed form to editor@slm-journals.com. All subscriptions are Free.


JOURNAL OF COAGULATION DISORDERS

ORIGINAL ARTICLE

Traffic of rFVIIa through Endothelial Cells and Redistribution into Subendothelium: Implications for a Prolonged Hemostatic Effect I Lopez-Vilchez1, J Tusell2, U Hedner3, C Altisent4, G Escolar1 and AM Galan1 Affiliations: 1Serv. Hemoterapia-Hemostasia, Hospital Clinic, CDB, IDIBAPS, UB, Barcelona, Spain; 2Novo Nordisk Pharma SA, Spain; 3Research and Development, Novo Nordisk A/S, Ma˚løv, Denmark, and University of Lund, Lund, Sweden and 4Unitat d’Hemofilia, Hospital Universitari Vall d’Hebron, Barcelona, Spain

A B S T R A C T Background Clinical evidence suggests that the hemostatic action of recombinant activated factor VII (rFVIIa) exceeds its predicted plasma life. Mechanisms involved in the long-lasting effects of rFVIIa for prophylactic treatment of patients with hemophilia and inhibitors have not been fully elucidated.

Objectives The traffic of rFVIIa through the endothelial cells (EC) and its redistribution into the subendothelial compartment was investigated. Its possible hemostatic action in experiments with flowing blood was also assessed.

Methods Cultured EC and umbilical veins were exposed to rFVIIa (6 mg/mL) for up to 2 h. Immunocytochemical techniques were combined with confocal microscopy to localize rFVIIa into EC and to trace its possible redistribution into subendothelial compartments. Vessels exposed to rFVIIa were de-endothelized and exposed to flowing blood to determine possible modifications in their hemostatic capacity.

Results Immunocytochemical studies revealed a significantly enhanced presence of FVIIa dispersedly distributed in the cytoplasm of EC previously exposed to rFVIIa. rFVIIa relocated into nuclear and peripheral areas when prolonged incubations (24 h) were performed. Immunocytochemical studies revealed that rFVIIa localizes into the endothelium and subendothelium of incubated umbilical vessels. In perfusion studies, this rFVIIa redistributed into the subendothelium improved fibrin generation and enhanced platelet thrombus formation.

Conclusions These results indicate that rFVIIa can be internalized and redistributed into endothelial and subendothelial compartments. This rFVIIa remains functional and promotes hemostatic activity when vessels are denuded. These findings may explain the prolonged prophylactic action of rFVIIa in some clinical conditions. Keywords: Pharmacodynamics, endothelial cells, hemostatic action, recombinant activated factor VII traffic, prophylaxis Correspondence: Irene Lo´pez-Vilchez, Servicio de Hemoterapia-Hemostasia, Hospital Clinic, C/Villarroel 170, 08036 Barcelona, Spain. Tel: +34-93-227-54-00, Ext. 2034; fax: +34-93-227-93-69; e-mail: ilopez1@clinic.ub.es

factor VIII (FVIII) or factor IX (FIX) [2], who cannot benefit from prophylaxis with FVIII or FIX.

INTRODUCTION It is well accepted that around 16–36% of hemophilia A and 6–8% of hemophilia B patients will develop inhibitory antibodies to the infused factor that will neutralize its hemostatic action [1]. Recombinant activated coagulation factor VII (rFVIIa) was developed for the treatment of bleeding episodes in hemophilia patients with inhibitors to coagulation JCD 2009; 1:(1). OCTOBER 2009

The introduction of rFVIIa has facilitated the clinical management of these patients. rFVIIa has the same structure and activity as the human factor, restoring hemostasis by favoring thrombin generation [3]. Interestingly, rFVII has proven useful in controlling active bleeding episodes not only in hemophilia, 1

www.slm-hematology.com


Journal of Coagulation Disorders

but also in other hemostatic deficiencies including platelet and coagulation disorders [4, 5]. An enhanced thrombin generation at damaged vessels has been proposed as the main mechanism contributing to the hemostatic action of rFVIIa in the control of active bleeding in congenital and acquired disorders of hemostasis [6, 7]. Tissue factor (TF) exposed at sites of vascular damage would help to localize the hemostatic response, favoring fibrin generation and platelet recruitment in more stable thrombi [8–10].

Reagents and Antibodies Whole blood was anticoagulated with buffered citrate/phosphate/dextrose solution (CPD) to a final concentration of citrate of 19 mM, or with low molecular weight heparin (LMWH; FragminH, Pharmacia, Madrid, Spain) at a final concentration of 20 U/mL. rFVIIa was supplied by Novo Nordisk (NovoSevenH, Novo Nordisk, Bagsvaerd, Denmark). To detect human rFVIIa, a polyclonal rabbit antihuman antibody was used (Agrisera, Va¨nna¨s, Sweden). For immunocytochemical techniques, we used as secondary antibody a goat anti-rabbit immunoglobulin (Ig)G Alexa 488 (Molecular Probes, Invitrogen, Carlsbad, CA, USA). Fluorescent properties were prolonged using the Prolong antifade kit (Molecular Probes, Invitrogen, Carlsbad, CA, USA). AuroProbe One GAR and IntenSETM M silver enhancement kit were from Amersham (Buckinghamshire, UK). Embedding kit JB-4 was from Polyscience (Warrington, USA).

The pharmacokinetic characterization of rFVIIa by different groups has determined a half-life of 2.7 h in adults and of 1.3 h for children [11–13]. Clinical experience from an exploratory phase II trial (in patients subjected to prophylaxis) suggests that the hemostatic action of recombinant activated factor VII (rFVIIa) exceeds its predicted plasma half-life [14–16]. Recent publications have highlighted the potential role of rFVIIa in the prophylaxis of hemophilic patients with inhibitors [15, 17, 18]. Although the mechanisms of action of rFVIIa in the correction of active bleeding have been widely studied, mechanisms involved in the apparent long-lasting effects of rFVIIa for prophylactic treatment remain to be elucidated. It has been speculated that a portion of the rFVIIa infused into patients could diffuse to the extravascular space and once there become available at the site of injury [19].

Studies on Human Endothelial Cells Incubation of ECs with rFVIIa. Endothelial cells were harvested from umbilical cords according to the method described initially by Jaffe et al, which is well established in our group [20, 21]. Coverslips with confluent EC monolayers were incubated in the presence or absence of rFVIIa (6 mg/mL final concentration) in Medium 199 supplemented with 2 mM glutamine, 100 U/mL penicillin, 100 mg/mL streptomycin, and 20% pooled human serum. Incubation was prolonged for 2 h according to the half-life calculated for rFVIIa, at 37 ˚C. An additional incubation of 24 h was also performed to trace the possible redistribution of rFVIIa into the EC. After incubation, coverslips were rinsed with phosphatebuffered saline (PBS) three times and fixed in 4% paraformaldehyde for 15 min at room temperature until immunolocalization was performed.

Our present studies have attempted to explain the possible mechanism involved in the prolonged hemostatic effectiveness of rFVIIa prophylaxis. Using an in vitro experimental setting, we have investigated the traffic of rFVIIa through the endothelial cells (EC) and its redistribution into the subendothelial compartments. To accomplish these objectives, we used a combination of immunocytochemical techniques and perfusion studies with circulating blood to detect the possible traffic of rFVIIa into endothelial cells and to evaluate the potential implications for its hemostatic activity.

Immunolocalization on Endothelial Cells: Confocal Microscopy.

MATERIALS AND METHODS

After the fixation procedure, EC coverslips were rinsed with PBS three times. Permeabilization and blocking for non-specific binding sites were performed with 0.1% saponin and 1% bovine serum albumin (BSA) in PBS (15 min at room temperature), followed by a washing step with PBS. Afterwards, coverslips were incubated with the primary antibody (14 mg/mL rabbit anti-human FVII/VIIa antibody in 0.1% saponin) for 1 h at 37 ˚C. After removing the excess primary antibody with PBS, a second incubation with a green fluorescent-labeled anti-rabbit antibody Alexa-488 was performed in darkness for 45 min at 37 ˚C. Coverslips were rinsed with PBS and with distilled water to remove the excess antibody, and then mounted in ProlongH Gold antifade mounting media (Molecular Probes) for confocal microscopy analysis. Non-specific

Experimental Design The present study was designed to evaluate the possible redistribution of rFVIIa in different intracellular compartments, specifically on the endothelium and subendothelial matrix. Isolated endothelial cells (EC) harvested from umbilical cords and human umbilical vessels were incubated with rFVIIa (6 mg/ mL final concentration). The presence of rFVIIa in EC and on endothelial extracellular matrices was detected by immunocytochemical studies using specific antibodies against FVIIa detected by confocal microscopy. The physiological relevance of rFVIIa redistribution to EC or subendothelial matrices was tested using flow systems. The effects on thrombus formation and fibrin generation were assessed. JCD 2009; 1:(1). OCTOBER 2009

2

www.slm-hematology.com


human umbilical vein. Vessels were incubated for 2 h with Hank’s balanced buffer (HBS) or with rFVIIa (6 mg/mL in the same buffer). After incubation, veins were thoroughly washed with PBS, everted, and subjected to a mechanical de-endothelization. Denuded umbilical veins were exposed to flowing blood anticoagulated with LMWH.

labeling was also assessed by omitting the primary antibody. Confocal images were acquired using a Leica TCS SL laser scanning confocal spectral microscope (Leica Microsystems Heidelberg GmbH, Mannheim, Germany) with argon and HeNe lasers attached to a Leica DMIRE2 inverted microscope. The images of fluorescein isothiocyanate (FITC) labeling were acquired as follows: excitation at 488 nm, triple dichroic beam-splitter (TD 488/543/633) and emission detection range of 500–535 nm. All images were obtained using a 636 oil immersion objective lens (NA 1.32) equipped with phase contrast optics and the confocal pinhole set at 1 Airy unit.

Blood (22 mL) anticoagulated with LMWH (20 U/mL) was perfused in an annular chamber using as thrombogenic substrate the damaged vascular segments [24]. Perfusions were performed at a shear rate of 600 s21 for 10 min. After perfusion, vascular segments were rinsed with PBS (0.15 M), fixed with 2.5% glutaraldehyde (in 0.15 M PBS) at 4 ˚C for 24 h and processed histologically for morphometric evaluation. Fibrin deposition and platelet interaction were evaluated by light microscopy connected to a computer, provided with special software that automatically classifies and quantifies platelet and fibrin coverage [25]. Platelet interaction was expressed globally as a percentage of the surface covered by platelets (%P). The presence of fibrin was also morphometrically quantified and expressed as a percentage of fibrin (%F) [25].

Studies on Human Umbilical Vessels Incubation of Vessels with rFVIIa. All the umbilical cords were collected within 24 h after delivery. Once rinsed with PBS, each cord was cut into two pieces of approximately 10 cm. Hank’s buffered salt solution (HBSS), with and without rFVIIa (6 mg/mL final concentration), was infused into a vein with both ends clamped and incubated for 2 h at 37 ˚C. After incubation, veins were rinsed with PBS, accurately separated from the remaining cord, and fixed overnight in glutaraldehyde (2.5% final concentration). Fixed veins were rinsed with PBS and dehydrated in an increasing gradient of ethanol. Finally, veins were embedded in glycol-methacrylate and processed histologically to obtain thin (2 mm) cross-sections [22].

Statistics Results were expressed as mean ¡ standard error of the mean (SEM). Values provided in immunolocalization experiments on endothelial cells and vascular segments correspond to four to eight experiments. Oneway ANOVA test for independent experiments was applied when multiple comparisons were required. Student’s t-test for paired data was used for comparisons in perfusion experiments. The level of statistical significance was established at P,0.01.

Immunolocalization on Vessel Cross-sections. The presence of rFVIIa in the cross-sections of the umbilical vein was assessed by immunocytochemical techniques at room temperature according to the instructions provided by the manufacturer of the AuroProbe One GAR kit. Non-specific binding sites were blocked with 0.8% BSA and 0.1% fish gelatin provided by the manufacturer, pH 7.4, for 30 min. Sections were incubated with 14 mg/mL rabbit antihuman FVII/VIIa antibody for 1 h. After removing the excess primary antibody by washing it three times with PBS, sections were incubated with a gold-conjugated goat anti-rabbit secondary antibody for 4 h. Excess secondary antibody was removed by washing three times with PBS and then three times with distilled water. Finally, samples were treated with an IntenSE silver enhancement reagent [23].

RESULTS Internalization of rFVIIa into Endothelial Cell Cultures Experiments performed with human EC cultures confirmed that rFVIIa was inside the EC. Cultures of EC monolayers exposed to rFVIIa (6 mg/mL) for 2 h showed an intense labeling to rFVIIa that appeared homogeneously distributed in the EC cytoplasm (Figure 1A,B). The analysis of control EC cultures, not incubated with rFVIIa, showed very weak signal labeling, probably related to the residual presence of FVII/VIIa in the supplemented culture medium, which contains a 20% serum pool.

In all experimental settings, non-specific binding was assessed with non-specific antibody IgG. Samples were visualized using a Reichert-Jung light microscope using epipolarization techniques.

Results with prolonged incubations of rFVIIa for 24 h (6 mg/mL) revealed a relocation of this rFVIIa in the EC cytoplasm, being more intense in the periphery and nucleus of the EC (Figure 1C).

Haemostatic Activity: Studies in Perfusion Chambers with Flowing Blood

Controls with the secondary antibody alone to assess non-specific binding showed no fluorescence, confirming that labeling in the EC cultures resulted from the presence of rFVIIa.

Hemostatic activity associated with extravascular rFVIIa was assessed using vascular segments from www.slm-hematology.com

3

JCD 2009; 1:(1). OCTOBER 2009


Journal of Coagulation Disorders

Figure 1: Representative fluorescence micrographs from confocal microscopy techniques applied to confluent endothelial cell (EC) cultures incubated in the absence (A) or in the presence (B) of rFVIIa (6 mg/mL) for 2 h at 37˚C. (C) corresponds to EC incubated with the same concentration of rFVIIa for 24 h at 37˚C. Arrows show positive labeling for FVIIa. Acquisition conditions were maintained constant for all the studied conditions. Magnification 696.

Redistribution of rFVIIa in Umbilical Veins The immunocytochemical techniques applied to the thin sections of umbilical cord veins revealed positive immunogold labeling for rFVIIa associated with the endothelium and extending to deeper areas in the veins such as the subendothelium and even the internal elastic lamina in those veins that had previously been incubated with rFVIIa for 2 h (Figure 2). Vessels not incubated with rFVIIa presented a much weaker positive labeling for FVIIa in the EC layer. One possible explanation for this labeling could be the origin of the veins, which have been exposed to blood. Controls with the secondary antibody alone to assess non-specific binding showed a mild dispersed labeling fluorescence without specific location.

Hemostatic Activity Associated with rFVIIa Associated with the Vessels Blood was perfused through denuded human umbilical veins previously incubated with HBS or rFVIIa (Figure 3). Perfusion runs over human umbilical veins incubated with HBS resulted in a percentage of platelet coverage equivalent to 11.5¡2.2 %, whereas fibrin formation was minimal (5.2¡4.9 %) (Figure 3A). In contrast, previous incubation of the vessel with rFVIIa resulted in a significant increase in fibrin deposition on the damaged areas (55.0¡10.4% vs 5.2¡4.9%; n54, P,0.01), but no significant changes in the percentage of the surface covered by platelets were detected (14.5¡3.8% vs 11.5¡2.2%) (Figure 3B).

Figure 2: Immunogold labeling on thin sections of umbilical cord vein incubated in the absence (A) or in the presence (B) of rFVIIa (6 mg/mL) for 2 h at 37˚C. (C) Non-specific interactions of the secondary antibody. Representative micrographs showing the presence of rFVIIa, especially in the endothelial layer (arrows), but also in the subendothelium and in deeper areas of the vasculature. Images were obtained from the same microscopic field. In each panel, the upper image was obtained in the light field modus, and the image below was obtained by epipolarization microscopic techniques. Magnification 4006.

All these results are summarized in a bar diagram (Figure 4).

DISCUSSION Our experimental results indicate that rFVIIa traffics through EC and redistributes into endothelial and JCD 2009; 1:(1). OCTOBER 2009

4

www.slm-hematology.com


Figure 3: Hemostatic activity associated with rFVIIa redistributed in the vascular wall. Images correspond to representative fields of perfusion studies performed with vascular vessels from human umbilical vein (A and B). Vessels were incubated for 2 h in the absence (upper image) or in the presence of 6 mg/mL rFVIIa (lower image). P, platelets; F, fibrin. Magnification 4006.

Figure 4: Bar diagram showing the percentage of subendothelium covered by platelets (%P; black) or fibrin (%F; white) in perfusion studies performed with blood anticoagulated with LMWH using de-endothelized human umbilical veins previously incubated with HBS (CON) or rFVIIa (6 mg/mL). Results are expressed as mean¡SEM; *P,0.01 vs CON.

subendothelial compartments. Interestingly, our data reveal that rFVIIa reaches the subendothelium, where it remains functional and can still promote hemostasis as shown in the flow model. Redistribution of rFVIIa into vascular compartments may explain the prolonged action of rFVIIa in the prophylaxis of bleeding disorders in some clinical conditions [14–16].

becomes internalized by the EC and reaches deeper areas of the vasculature such as the subendothelium. Although our immunocytochemical studies show an increased presence of rFVIIa in the EC cytoplasm after a short exposure (2 h), prolonged incubations (for 24 h) resulted in further relocation of rFVIIa into the EC towards the cell periphery and nucleus. It is interesting to note that this observation has not been communicated previously by others. However, there is evidence that EPCR can translocate from the plasma membrane to the nucleus, even bound to APC, where it redirects gene expression [31]. Therefore, if rFVIIa is able to bind to EPCR, it is plausible to find rFVIIa in the nucleus. Moreover, it has been suggested that the FVIIa–TF complex may play an active role in the modulation of gene expression [32]. Further studies are required to understand the physiological and clinical implications of this process during rFVIIa treatment.

Previous studies reported that the presence of rFVIIa potentiates coagulation and thrombus formation [9, 10]. This effect is probably related to an increase in thrombin generation through a bypassing activity [6, 7]. Several studies have shown that rFVIIa caused a local increase in procoagulant action on the subendothelium under several hemostatic dysfunctions [8– 10, 24, 26, 27]. All this evidence indicates that the mechanism of action of rFVIIa under active bleeding situations may mainly be driven by an increase in fibrin and an improvement in thrombus formation on vascular damaged vessels where tissue factor is exposed. Notwithstanding, the previous mechanism would not explain the beneficial action observed in prophylaxis [15, 16].

Interestingly, the rFVIIa redistributed in the subendothelial compartments remains in an active form, as demonstrated by its ability to promote fibrin generation on damaged vessels (Figure 3). These data suggest that, in contrast to the inactivation of the procoagulant action of rFVIIa into EC anticipated by earlier studies [30], a portion of the rFVIIa that reaches the subendothelium is still capable of promoting and enhancing hemostatic activity. It is not surprising that the presence of the TF–rFVIIa complex in deeper damaged areas could also facilitate the formation of a more stable fibrin structure, resistant to fibrinolytic mechanisms, as has been suggested in previous studies [24, 33]. It is known that the presence of rFVIIa normalizes fibrin clot permeability, improving network structure [34].

Recently, it has been hypothesized that rFVIIa could diffuse to the extravascular compartment, increasing the local concentration of rFVIIa available at the site of injury and favoring the formation of the TF–rFVIIa complex [19]. It has already been reported that rFVIIa is able to bind to endothelial cell protein C receptor (EPCR) with similar affinity to protein C or activated protein C (APC) [28, 29]. These authors suggested that this binding would facilitate the internalization of FVIIa into the EC. Further studies have suggested that rFVIIa bound to EPCR would reduce the procoagulant activity associated with the FVIIa–TF complex [30]. The latter findings indirectly imply that the rFVIIa present in EC would have a limited role in promoting procoagulant activities.

In summary, our present studies in a homologous human experimental system demonstrate that rFVIIa could redistribute into endothelial cells and subendothelial compartments, where it would still retain its hemostatic activity. The redistribution of rFVIIa into

The results from our present investigations prove trafficking of FVIIa to the subendothelium through endothelial cells. Our studies indicate that rFVIIa www.slm-hematology.com

5

JCD 2009; 1:(1). OCTOBER 2009


Journal of Coagulation Disorders

15. Morfini M, Auerswald G, Kobelt RA, et al. Prophylactic treatment of haemophilia patients with inhibitors: clinical experience with recombinant factor VIIa in European Haemophilia Centres. Haemophilia. 2007;13(5):502–507. 16. Konkle BA, Ebbesen LS, Erhardtsen E, et al. Randomized, prospective clinical trial of recombinant factor VIIa for secondary prophylaxis in hemophilia patients with inhibitors. J Thromb Haemost. 2007;5(9):1904–1913. 17. Yilmaz AA, Yalcin S, Serdaroglu H, Sonmezer M, Uysalel A. Prophylaxis with recombinant-activated factor VII (rFVIIa) for minimally invasive surgery in a patient with congenital factor VII deficiency: a case report with a single-low dose of rFVIIa. Blood Coagul Fibrinolysis. 2008;19(7):693–695. 18. Jimenez-Yuste V, Alvarez MT, Martin-Salces M, et al. Prophylaxis in 10 patients with severe haemophilia A and inhibitor: different approaches for different clinical situations. Haemophilia. 2009;15(1):203–209. 19. Hedner U. Potential role of recombinant factor FVIIa in prophylaxis in severe hemophilia patients with inhibitors. J Thromb Haemost. 2006;4(11):2498–2500. 20. Jaffe EA, Nachman RL, Becker CG. Culture of human endothelial cells derived from umbilical veins. J Clin Invest. 1973;52: 2745–2756. 21. Fuste B, Serradell M, Escolar G, et al. Erythropoietin triggers a signaling pathway in endothelial cells and increases the. Thromb Haemost. 2002;88:678–685. 22. Escolar G, Bastida E, Ordinas A, Castillo R. Interaction of platelets with subendothelium in humans treated with aspirin and dipyridamole alone or in combination. Thromb Res. 1985;40: 419–424. 23. Furlan M, Robles R, Lammle B, Zimmermann J, Hunziker E. Immunogold labelling of human von Willebrand factor adsorbed to collagen. Blood Coagul Fibrinolysis. 1991;2:441–446. 24. Tonda R, Galan AM, Pino M, et al. Hemostatic effect of activated recombinant factor VII (rFVIIa) in liver disease: studies in an in vitro model. J Hepatol. 2003;39(6):954–959. 25. Escolar G, Galan AM, Mazzara R, Castillo R, Ordinas A. Measurement of platelet interactions with subendothelial substrata: relevance to transfusion medicine. Transf Med Rev. 2001; 15:144–156. 26. Tonda R, Galan AM, Pino M, Lozano M, Ordinas A, Escolar G. Hemostatic effect of activated recombinant factor VIIa in Bernard–Soulier syndrome: studies in an in vitro model. Transfusion. 2004;44(12):1790–1791. 27. Tonda R, Galan AM, Mazzara R, White JG, Ordinas A, Escolar G. Platelet membrane fragments enhance the procoagulant effect of recombinant factor VIIa in studies with circulating human blood under conditions of experimental thrombocytopenia. Semin Hematol. 2004;41:157–162. 28. Mandal SK, Pendurthi UR, Rao LVM. Tissue factor internalization and trafficking in fibroblasts: involvement of protease activated receptors-mediated cell signaling. Blood. 2006;108(11):163A. 29. Ghosh S, Pendurthi UR, Steinoe A, Esmon CT, Rao LV. Endothelial cell protein C receptor acts as a cellular receptor for factor VIIa on endothelium. J Biol Chem. 2007;282(16):11849–11857. 30. Lopez-Sagaseta J, Montes R, Puy C, Diez N, Fukudome K, Hermida J. Binding of factor VIIa to the endothelial cell protein C receptor reduces its coagulant activity. J Thromb Haemost. 2007;5(9):1817–1824. 31. Esmon CT. Crosstalk between inflammation and thrombosis. Maturitas. 2004;47(4):305–314. 32. Camerer E, Gjernes E, Wiiger M, Pringle S, Prydz H. Binding of factor VIIa to tissue factor on keratinocytes induces gene expression. J Biol Chem. 2000;275(9):6580–6585. 33. Lisman T, Mosnier LO, Lambert T, et al. Inhibition of fibrinolysis by recombinant factor VIIa in plasma from patients with severe hemophilia A. Blood. 2003;99:175–179. 34. He S, Blomback M, Jacobsson EG, Hedner U. The role of recombinant factor VIIa (FVIIa) in fibrin structure in the absence of FVIII/FIX. J Thromb Haemost. 2003;1(6):1215–1219.

vascular compartments may explain the prolonged prophylactic action of rFVIIa under some clinical conditions and provide a mechanism of action to explain the effectiveness of rFVIIa observed in prophylaxis. Notwithstanding, the implications of our findings for improving hemostasis in a clinical setting must be confirmed in patients with hemostatic disorders. The existence of an additional reservoir of rFVIIa may provide new insights into the physiological and pathological implications of FVIIa in hemostasis. Acknowledgements: The authors would like to thank Montserrat Vin˜as, Fulgencio Navalon, Marc Pino, and Patricia Molina from the Service of HemotherapyHaemostasis of Hospital Clinic, and Maria Calvo, from the Unit of Confocal Microscopy, SCT, Universitat de Barcelona, for their technical assistance. We also thank the Delivery Room Staff of the Hospital de Sant Joan de Deu and the Hospital de la Maternitat in Barcelona, Spain, for providing the umbilical cords. This work has been partially supported by grants SAF2006-08003 and SAF2009-10365, FIS PI060260, PET2007_0169, and Red HERACLES RD06/0009 from the Spanish government, SGR 2005-00952 from the Generalitat of Catalunya, and by an unrestricted grant from Novo Nordisk.

REFERENCES 1. Darby SC, Keeling DM, Spooner RJ, et al. The incidence of factor VIII and factor IX inhibitors in the hemophilia population of the UK and their effect on subsequent mortality, 1977–99. J Thromb Haemost. 2004;2(7):1047–1054. 2. Hedner U, Ingerslev J. Clinical use of recombinant FVIIa (rFVIIa). Transfus Sci. 1998;19(2):163–176. 3. Monroe DM. Further understanding of recombinant activated factor VII mode of action. Semin Hematol. 2008;45(2 Suppl 1):S7– S11. 4. Hedner U, Erhardtsen E. Potential role for rFVIIa in transfusion medicine. Transfusion. 2002;42(1):114–124. 5. Hedner U, Ezban M. Tissue factor and factor VIIa as therapeutic targets in disorders of hemostasis. Annu Rev Med. 2008;59:29–41. 6. Hoffman M, Monroe DM. The action of high-dose factor VIIa (FVIIa) in a cell-based model of hemostasis. Semin Hematol. 2001; 38(4):6–9. 7. Roberts HR, Hoffman M, Monroe DM. A cell-based model of thrombin generation. Semin Thromb Hemost. 2006;32:32–38. 8. Galan AM, Tonda R, Altisent C, Maragall S, Ordinas A, Escolar G. Recombinant factor VIIa (NovoSeven(R)) restores deficient coagulation: experience from an ex vivo model. Semin Hematol. 2001;38:10–14. 9. Lisman T, De Groot PG. Mechanism of action of recombinant factor VIIa. J Thromb Haemost. 2003;1(6):1138–1139. 10. Galan AM, Tonda R, Pino M, Reverter JC, Ordinas A, Escolar G. Increased local procoagulant action: a mechanism contributing to the favorable hemostatic effect of recombinant FVIIa in PLT disorders. Transfusion. 2003;43:885–892. 11. Lindley CM, Sawyer WT, Macik BG, et al. Pharmacokinetics and pharmacodynamics of recombinant factor VIIa. Clin Pharmacol Ther. 1994;55(6):638–648. 12. Girard P, Nony P, Erhardtsen E, et al. Population pharmacokinetics of recombinant factor VIIa in volunteers anticoagulated with acenocoumarol. Thromb Haemost. 1998;80(1):109–113. 13. Erhardtsen E. Pharmacokinetics of recombinant activated factor VII (rFVIIa). Semin Thromb Hemost. 2000;26(4):385–391. 14. Ludlam CA. The evidence behind inhibitor treatment with recombinant factor VIIa. Pathophysiol Haemost Thromb. 32(Suppl2002;1):13–18.

JCD 2009; 1:(1). OCTOBER 2009

6

www.slm-hematology.com


JOURNAL OF COAGULATION DISORDERS

ORIGINAL ARTICLE

Dental and Periodontal Health in Children with Hemophilia E Alpkılıc¸ Baskirt1, H Albayrak2, G Ak1, A Pınar Erdem2, E Sepet2 and B Zulfikar3 Affiliations: Departments of Oral Medicine and Oral Surgery1 and Pedodontics2, Faculty of Dentistry, Istanbul University and 3 Department of Paediatric Haematology-Oncology, Cerrahpasa Medical Faculty, Istanbul University, Istanbul, Turkey Submission Date: 9th June 2009, Revision Date: 14th July 2009, Acceptance Date: 30th July 2009

A B S T R A C T Aim The aim of this study was to determine the dental and periodontal health of children with hemophilia A in different aspects.

Methods The gingival index, plaque index, and dmf(t)-dmf(s), DMF(T)-DMF(S) scores of 36 children (aged between 6 and 12 years) with hemophilia A and 39 healthy children were analyzed. Type of hemophilia, dietary habits, and frequency of tooth brushing, educational and economic level of the parents, and parents’ dental habits were determined by a questionnaire.

Results All the analyses were carried out by means of commercial statistical software. Statistical analyses were performed by chi-square, Kruskal–Wallis and independent t tests. The difference in plaque index scores between the study and control groups (P50.077) was not statistically significant. Gingival index scores of the children with hemophilia were statistically higher than the control group (t53.10; P50.003). No difference was found in dmf(t)-dmf(s) scores, but DMF(T)-DMF(S) scores of the study group were statistically higher than the control group (P50.001; P50.012). The frequency of tooth brushing was defined statistically different between the two groups (P50.044). The frequency of sugar consumption by the children with hemophilia was found to be statistically higher than that of the control group (P50.006). Although the maternal educational levels of the two groups were similar (P50.130), paternal education levels and the economic levels of the study group were statistically lower than those of the control group (P50.002).

Conclusion We conclude that children with hemophilia A have significantly higher GI and DMF(T)-DMF(S) scores compared with matched, healthy control subjects. Expanded preventive measures, educational, and recall programs should be organized for these ‘‘special needs’’ patients. Keywords: hemophilia, dental, periodontal, children, DMF Correspondence: Esra Alpkilic Baskirt, Halıcılar Avenue, Oksuzler Street, No: 18/6 Fatih, Istanbul, Turkey. Tel: +9-0542321-21-42; fax: +9-0212-531-22-31; e-mail: esra_alpkilic@yahoo.com

Alpkilic et al [2] reported that prevalence of tooth brushing was significantly higher in the healthy control group. Healthy control subjects have a more regular brushing habit than people with hemophilia. As a result, poor oral hygiene causes an increase in dental caries. Severe gingivitis and periodontitis can also be seen in these patients.

INTRODUCTION Hemophilia is a life-threatening inherited bleeding disorder characterized by a lifelong defect in the clotting mechanism [1]. Like every part of the body, hemophilia also has effects on the oral region. Spontaneous bleeding can occur in hemophilia patients with periodontal disease. In thinner regions of the gingival, there are a number of enlarged capillaries near the surface. Therefore, minor trauma such as tooth brushing or food abrasion and infection can cause bleeding in gingival tissues [1]. Patients with hemophilia neglect their oral hygiene because of the bleeding during tooth brushing. In a recent study, JCD 2009; 1:(1). OCTOBER 2009

Mild hemophilia may not be diagnosed until adolescence or even later, particularly if major surgery, severe trauma, or dental extractions have been avoided. In a study that investigated the nature of bleeding episodes leading to the diagnosis of 132 patients with hemophilia A, 14% of hemophilia 7

www.slm-hematology.com


Journal of Coagulation Disorders

patients and 30% of cases with mild type were reported to have been initially diagnosed following an episode of severe oral bleeding. The most common site of oral bleeding was the labial frenulum and the tongue [3, 4]. Thus, the dentist may be the first to diagnose a patient with hemophilia.

RESULTS The qualitative and quantitative data gained from the study and control groups are presented in Tables 1 and 2. The data gained from comparison of the hemophilia group (severe, moderate, mild) are presented in Table 3.

However, there have been several reports related to the surgical procedures of hemophilia; there are few data for the dental health of children with hemophilia and their parents or adults. The purpose of the study reported here was to investigate the dental and periodontal health of children with hemophilia A and to evaluate the relation of children’s dental health with parents’ educational and economic status.

The difference in mean age scores between the study and control groups was not statistically significant. GI scores of the children with hemophilia were determined statistically higher than control group. DMF(T) scores of the study group were found higher than those of the control group. DMF(S) scores of the study group were statistically higher than those of the control group. No difference was found in dmf(t)-dmf (s) scores between the study and control groups.

PATIENTS AND METHODS

A statistically significant difference was found in the prevalence of tooth brushing between the study and control groups. In the study group, the number of participants who never brushed his/her teeth is significantly higher than in the control group (P,0.05). It was found that children with no systemic disease consume candy more frequently than those with hemophilia (P,0.05). Consumption of foods such as cakes, biscuits, and cookies was found to be higher in healthy children compared with the study group (P,0.05). Intergroup comparison for maternal education level revealed no significant difference, whereas paternal education level and families’ financial circumstances were statistically lower in the study group than in the control group.

The study group included 36 children suffering from hemophilia A in the age range 6–12 years who were registered with the Haemophilia Society of Turkey. Fourteen patients had mild hemophilia A, whereas 12 had moderate and 10 had severe type. The control group included 39 healthy children who were referred to Istanbul University Faculty of Dentistry, Department of Pedodontics. They were matched with children with hemophilia with regard to age and gender. A questionnaire including questions about the presence of other hemophilic members of the family, type of hemophilia, dietary habits, frequency of tooth brushing, educational and economic level of the parents, and parents’ dental hygiene habits was administered to the participants.

The GI, PI, DMF(T), and DMF(S) scores revealed no significant difference between the groups with severe, moderate, and mild types of hemophilia. However, values of dmf(t) and dmf(s) in patients with mild hemophilia were found to be significantly higher than those with moderate and severe hemophilia.

All participants including the study and control group were examined in a dental chair under a standard dental light. The gingival index (GI) (Loe & Silness) [5], plaque index (PI) (Silness & Loe) [6], and dmf(t)-dmf(s), DMF(T)-DMF(S) [7, 8] scores of children with hemophilia A and healthy children were compared. The DMF(T/S) index, which was developed by the World Health Organization, was created to express the caries experience of people and to obtain data regarding the dental health of populations. The D component is used for untreated caries, M for missing teeth due to caries, and F for filling (dental restorations for caries treatment). The T means index per tooth (as opposed to S per surface). Also, dmf(t/s) is the form of the index used for deciduous teeth [7, 8]. In the study group, each patient had an initial consultation prior to dental examination and treatment to determine the required type of replacement therapy procedure that was organized with the collaboration of the hematologist and dentist.

DISCUSSION Tooth brushing is the main component of preventive dentistry and the removal of dental plaque [5, 6, 9]. In this study, GI values were significantly higher in Table 1. Gingival, Plaque, dmf, and DMF Indices Scores of the Study and Control Groups Study group (n536)

The data was subjected to statistical analysis by using Graphpad Prisma V.3. Statistical analyses were performed by Chi-Square, Kruskal Wallis and independent t tests. The level of significance was p , 0.05. JCD 2009; 1:(1). OCTOBER 2009

Control group Independent (n539) t test

P

Age

9.54 ¡ 2.39

9.49 ¡ 1.70

0.11

0.913

GI

0.39 ¡ 0.48

0.15 ¡ 0.14

3.10

0.003*

PI

0.88 ¡ 0.42

0.72 ¡ 0.37

1.79

0.077

DMF(T) DMF(S)

3.44 ¡ 3.30 5.78 ¡ 6.64

1.37 ¡ 1.62 2.45 ¡ 4.04

3.41 2.58

0.001* 0.012*

dmf(t)

3.44 ¡ 3.43

3.24 ¡ 2.62

0.27

0.788

dmf(s)

5.89 ¡ 6.57

6.32 ¡ 6.84

–0.25

0.803

*The statistically significant values which provide the rule P#0.05. GI, gingival index; PI, plaque index.

8

www.slm-hematology.com


Dental and Periodontal Health in Children with Hemophilia

Table 2. The Data gained by the Survey applied to the Parents of the Study and Control Groups

Did your child breast-feed?

Yes

Study group (n536) (%)

Control group (n539) (%)

Chi-square test

92.3

79.5

x2 2.64

No

7.7

20.5

P50.104

Did you add sugar or sweet food to your child’s night milk?

No

53.8

56.4

x2 0.05

Sugar, honey

46.2

43.6

Did you do any dental cleaning when your child was young?

Yes

7.7

17.9

x2 1.83

No

92.3

82.1

P50.176

How often does your child brush his/her teeth?

Twice/more than twice a day Once a day

30.8 33.3

7.7 53.8

Once/twice a week

20.5

25.6

Once/twice a month

2.6

7.7

12.8

5.1

Never In your opinion, does your child consume candies frequently?

No

61.5

30.8

Yes

38.5

69.2

How often does your child eat cariogenic foods (i.e., cakes, biscuits)

Never

12.8

0.0

Once/twice a week Once/twice a day

64.1 23.1

53.8 46.2

Maternal educational level

University

How often do you brush your teeth? (mother)

Paternal educational level

How often do you brush your teeth? (father)

P50.0820

x2 9.79 P50.044* x2 7.42 P50.006* x2 8.34 P50.015*

3.1

20.5

High school

15.6

23.1

Secondary school

15.6

7.7

x2 7.12

Primary school

65.6

48.7

P50.130

Twice/more than twice a day

21.1

23.1

Once a day

42.1

51.3

x2 1.15

Once/twice a week University

36.8 2.6

25.6 18.4

P50.561

High school

23.1

50.0

Secondary school

12.8

5.3

Primary school

61.5

26.3

Twice/more than twice a day

17.9

27.8

x2 15.11 P50.002*

Once a day

43.6

41.7

Once/twice a week

28.2

25.0

x2 1.40

Never

10.3

5.6

P50.705

*The statistically significant values which provide the rule P#0.05.

the hemophilia group than in the control group. Therefore, we thought that higher index results in children with hemophilia were a consequence of neglected or insufficient tooth brushing. When caries are considered, children with hemophilia must be seen as a high-risk group. In this study, their DMF(T) and DMF(S) values were significantly higher than in the control group. The factor that increased the GI values affected the high prevalence of caries in the study

group. Although the consumption of candies was not frequent in the haemophilia group, the DMF(T) and DMF(S) values were found significantly higher; it might depend on irregular consumption time of the cariogenic foods. The data gained from the questionnaire and clinic examinations of the study group disclosed the need for and necessity of preventive treatments and dental education in patients with hemophilia.

Table 3. Evaluation of Patients by the Type of Hemophilia Severe

Moderate

Mild

Kruskal–Wallis test

P

10.20 ¡ 2.49

8.92 ¡ 2.27

9.57 ¡ 2.41

1.70

0.428

GI PI

0.41 ¡ 0.29 0.84 ¡ 0.36

0.30 ¡ 0.47 0.99 ¡ 0.38

0.35 ¡ 0.33 0.72 ¡ 0.43

2.06 1.04

0.357 0.594

DMF(T)

2.78 ¡ 3.11

2.30 ¡ 2.10

4.45 ¡ 3.96

1.46

0.483

DMF(S)

3.67 ¡ 4.36

3.50 ¡ 4.17

9.00 ¡ 8.84

1.71

0.425

dmf(t)

3.00 ¡ 3.74

2.20 ¡ 3.01

6.00 ¡ 2.88

7.67

0.022*

dmf(s)

3.71 ¡ 3.94

4.60 ¡ 7.52

9.80 ¡ 8.54

5.99

0.05*

Age

*The statistically significant values which provide the rule P#0.05.

www.slm-hematology.com

9

JCD 2009; 1:(1). OCTOBER 2009


Journal of Coagulation Disorders

In Northern Ireland, Boyd and Kinirons [10] found better scores on dental indices in children with hemophilia. The prevalence of caries experienced and of untreated caries was lower in the primary and permanent dentition of patients with hemophilia [10]. Another study by Sonbol et al [4] in the UK, in which 38 children with severe hemophilia were compared with healthy control subjects, similar results including lower caries scores were found. On the other hand, a recent evaluation in Poland demonstrated no significant difference in caries prevalence in 80 children (aged between 4 and 18 years) with congenital bleeding disorders. However, worse dental status was seen in children with severe forms of hemophilia A and von Willebrand disease compared with other sick children [11]. Boyd and Kinirons [10] found that the reasons why the pediatric hemophilic population of Northern Ireland has less decay and a higher restorative index was the better motivated patients and parents about the prognosis and effects of hemophilia on oral tissues, the importance of attendance for dental treatment, and dental hygiene habits. Patients with hemophilia receive a more vigorous dental prevention program than the general population [10].

CONCLUSION The results of our study revealed that children with hemophilia constitute a special group for dental care. Dental management of patients with hemophilia should begin with prevention of dental disease. The parents’ approach to dental care has an importance to people with hemophilia in relation to dental care. Parents should be advised about the significance of and necessity for oral care. Preventive care should be delivered as early as possible, and patients should be recalled for regular dental visits. In this way, the need for active treatment may be reduced to a minimum level. Disclosures: The authors have no financial interests to disclose related to the contents of this article.

REFERENCES 1. Patton LL. Bleeding and clotting. In: Greenberg MS, Glick M, eds. Burket’s Oral Medicine Diagnosis and Treatment. 10th ed. Hamilton: BC Decker; 2003:454–478. 2. Alpkilic Baskirt E, Ak G, Zulfikar B. Oral and general health related quality of life among young haemophilia patients. Haemophilia. 2009;15(1):193–198. 3. Sonis A, Musselman RJ. Oral bleeding in classic haemophilia. Oral Surg Oral Med Oral Pathol. 1982;53:363–366. 4. Sonbol H, Pelargidou M, Lucas VS, et al. Dental health indices and caries related microflora in children with severe haemophilia. Haemophilia. 2001;7:468–474. 5. Loe H, Silness J. Periodontal disease in pregnancy I. Prevalance and severity. Acta Odaont Scand 1963;25:533-48. 6. Silness J, Loe H. Periodontal disease in pregnancyII. Correlation between oral hygiene and periodontal condition. Acta Odont Scand 1964;22:121-35. 7. Becker T, Levin L, Shochat T, Einy S. How much does the DMFT index underestimate the need for restorative care? J Dent Educ. 2007;71(5):677–681. 8. Oral Health Surveys: Basic Methods. Geneva: World Health Organization; 1987. 9. Newman MG, Takei HH, Carranza FA. Carranza’s Clinical Periodontology. 9th ed. Philadelphia: WB Saunders; 2002:543. 10. Boyd D, Kinirons M. Dental caries experiences of children with haemophilia in Northern Ireland. Int J Paediatr Dent. 1997;7: 149–153. 11. Mielnik-Blaszczak M. Evaluation of dentition status and oral hygiene in Polish children and adolescents with congenital hemorrhagic diatheses. Int J Paediatr Dent. 1999;9:99–103.

Parents of participants in the control group showed a higher tendency for regular dental visits. Low prevalence of regular dental visits in patients with hemophilia may be related to the low frequency of the families’ dental visits. Financial income and education level in the study group were lower than those of the control group. The most important issue related to this study is the serious lack of studies concerning dental health in children with hemophilia related with the socio-economic and educational status of parents. It is considered that the results found in this study depend closely upon the deficiency of patient information on the subject, education level of the families, socio-economic and cultural dissimilarities. There was no difference between the tooth brushing prevalence of parents in both groups. We know that the frequency of dental hygiene habits is directly related to people’s PI level. The children who were educated by careless parents about dental hygiene habits do not give the sufficient care and importance to tooth brushing. So, no difference was found between the PI level of the study and control groups. Also, at the deciduous dentition period, children take their mothers as a role model in dental hygiene habits; as a result of this situation, there was no difference between the dmf(t) and dmf(s) levels of children in the two groups. But at the mixed and permanent dentition period, children with hemophilia grow up and become aware of their illness. So they are afraid of gum bleeding and neglect tooth brushing. Because of neglected and insufficient tooth brushing, the DMF(T) and DMF(S) levels of the hemophilia group become statistically worse than those of the control group. JCD 2009; 1:(1). OCTOBER 2009

10

www.slm-hematology.com


JOURNAL OF COAGULATION DISORDERS

REVIEW ARTICLE

Gastrointestinal Angiodysplasia and Acquired Von Willebrand Syndrome: A Review of an Enigmatic Association Renu Saxena and Prashant Sharma Affiliation: Department of Hematology, All India Institute of Medical Sciences, Ansari Nagar, New Delhi 110029, India Submission date: 15th June 2009, Revision date: 29th June 2009, Acceptance date: 14th August 2009

A B S T R A C T Von Willebrand disease (VWD), the commonest inherited bleeding disorder, is characterized by reduced levels and/or abnormal function of von Willebrand factor (VWF). It may rarely be acquired (von Willebrand syndrome, VWS) in association with a variety of underlying conditions including myeloid, lymphoid and nonhematologic malignancies, autoimmune disorders, specific pharmaceutical agents, valvular heart defects and endocrine abnormalities. Gastrointestinal bleeding due to intestinal angiodysplasia is a well-documented complication of VWD as well as the acquired VWS, predominantly seen in elderly patients. Its high prevalence in disease subtypes associated with reduced or absent high molecular weight (HMW) VWF multimers suggests that the specific loss of these molecules is the primary pathologic event in abnormal hemostasis under high shear stress blood flow conditions. This link, first postulated specifically by Warkentin and colleagues, has now been borne out by both laboratory and epidemiological studies. VWF HMW multimer deficiency also provides a plausible link between senile aortic stenosis and gastrointestinal bleeding in Heyde syndrome. Enhanced proteolysis of HMW multimers by the shear-dependent VWF cleaving metalloproteinases as well as avid platelet binding of uncoiled HMW multimers results in accelerated clearance from circulation. The resultant deficiency mimicking VWD type 2A frequently resolves after surgical correction of the cardiac defect. The diagnosis and management of occult gastrointestinal bleeding from angiodysplasia in VWD or VWS are often challenging as is differentiating the acquired from the inherited forms of the disorder. Multiple therapeutic approaches have been tried with varying degrees of success. Keywords: acquired von Willebrand disease, angiodysplasia, aortic stenosis, Heyde syndrome, high molecular weight multimers, high shear stress, occult gastrointestinal bleeding Correspondence: Dr Renu Saxena, Professor and Head, Department of Hematology, All India Institute of Medical Sciences, Ansari Nagar, New Delhi 110029, India. Tel: +91-11-2659-4670; fax: +91-11-2658-8663; e-mail: renusax@hotmail.com

by platelets in myeloproliferative neoplasms [6, 7]), a reduced rate of synthesis of VWF (for example, in hypothyroidism or in patients on valproate therapy [8, 9]) or by accelerated proteolytic degradation of VWF (as in ciprofloxacin therapy [10] and in patients with congenital and acquired cardiac defects [11]).

INTRODUCTION Von Willebrand disease (VWD) is the commonest inherited bleeding disorder worldwide. The disease is characterized by low levels and/or abnormal function of the plasma protein von Willebrand factor (VWF). It may rarely be acquired, often in older patients, in association with a large number of underlying conditions (Table 1). It is then sometimes referred to as von Willebrand syndrome (VWS) to distinguish it from the inherited form [1, 2].

Among this last category are a group of patients, often elderly, who present with acquired VWD, gastrointestinal (GI) angiodysplasia and frequently, cardiovascular lesions, such as aortic stenosis, ventricular septal defect and or primary pulmonary hypertension. The cardiac association was first recognized in 1958 [12] although the precise link to VWS was proposed much later. We have, in the past, encountered 2 patients with acquired VWS and bleeding angiodysplasia [4, 13]. This review explores the nature of the relationship between these seemingly dissimilar disorders, the insights into the pathophysiology of

Multiple possible mechanisms may account for VWS, including the production of inhibitory or immune complex-forming antibodies against VWF (for example, in association with lymphoid or plasma cell proliferations or auto-immune disorders [3–5]), the adsorption of VWF by malignant cells (for example, in solid tumors such as adrenal cortical carcinomas or JCD 2009; 1:(1). OCTOBER 2009

11

www.slm-hematology.com


Journal of Coagulation Disorders

Table 1. Known Associations of Acquired Von Willebrand Syndrome Lymphoid neoplasms: Acute lymphoblastic leukemia, Plasma cell dyscrasias (MGUS, Waldenstro¨m macroglobulinemia, plasma cell myeloma), Chronic lymphocytic leukemia, Hairy cell leukemia, B-cell NHL Myeloid neoplasms: Chronic myeloid leukemia, Primary myelofibrosis, Essential thrombocythemia, Polycythemia vera, Acute myeloid leukemia Endocrinopathies: Hypothyroidism, Diabetes mellitus Autoimmune disorders: Systemic lupus erythematosus, Scleroderma, Mixed connective tissue disorder, Autoimmune hemolytic anemia, ANCA positive vasculitis Solid tumors: Nephroblastoma, Adrenal cortical carcinoma, Lung cancer, Peripheral neuroectodermal tumor, Gastric carcinoma Pharmaceutical agents: Ciprofloxacin, Valproate, Hydroxyethyl starch, Griseofulvin Miscellaneous: Angiodysplasia, Hemoglobinopathies, Hypercholesterolemia, Ehlers-Danlos syndrome, EBV infection, Postallogeneic bone marrow transplantation, Hydatidosis, Pesticide poisoning, Uremia, Congenital and acquired valvular heart disease (aortic stenosis, hypertrophic pulmonary osteoarthropathy, ventricular septal defect), Primary pulmonary hypertension ANCA, anti-neutrophil cytoplasmic antibody; EBV, Epstein-Barr virus; MGUS, monoclonal gammopathy of undetermined significance; NHL, non-Hodgkin’s lymphoma.

these intriguing illnesses and our attempts at diagnosing and managing these patients.

lengths of .2 micrometers, permitting visualization by electron microscopy. Uncoiling enhances VWF–platelet adhesion by means of the Gp Ib–IX receptor complexes. Thereafter, surface-bound VWF induces platelet adhesion and aggregation with the generation of an activated platelet–phospholipid surface facilitating clotting by formation of the primary hemostatic plug. The larger HMW VWF multimers are hemostatically the most active in this process on mucosal surfaces. This process, deficient in acquired VWS, is the primary cause of bleeding in these patients as discussed later [14–17].

VWF: RELEVANT PATHOPHYSIOLOGY VWF is a multimeric glycoprotein, synthesized by endothelial cells and megakaryocytes. It mediates platelet adhesion to the subendothelium of damaged blood vessels under high shear conditions or conditions of endothelial denudation and injury and also acts as a carrier for factor VIII. VWF is present in plasma, platelets, megakaryocytes, endothelium and subendothelium. The protein circulates in plasma as variably sized multimers ranging in molecular mass from 500 to over 10000 kDa (comprised of 2 to 20 or more monomeric 250 kDa pro-VWF units in dimeric protomers) [14, 15].

SELECTIVE DEFICIENCY OF HMW MULTIMERS: THE VWS/VWD TYPE 2A VWS type 2A has long attracted scientific attention due to its strong association with bleeding in a subset of acquired VWS patients [18, 19]. This qualitative variant displays diminished VWF-dependent platelet adhesion due to a lower proportion of the largest VWF multimers (Figure 1). Factor VIII and VWF antigen levels are near-normal or only modestly decreased. VWF/ristocetin cofactor assay shows markedly decreased activity, thus pointing toward abnormal VWF function [18–21].

High molecular weight (HMW) VWF multimers are those larger than 10 dimers or so in size. These are stored within in the Weibel-Palade bodies in endothelial cells and in platelets, whereas smaller multimers are secreted. The molecules also express ABO glycoprotein antigens, with carbohydrates comprising approximately one-fifth of the total VWF mass [14, 15]. ADAMTS-13 belongs to the family of ‘‘a disintegrinlike and metalloproteinases with thrombospondin-1 motifs’’ that cleaves VWF between tyr1605 and met1606 in its A2-domain. Apart from proteolysis and clearance of VWF, it is also central in maintaining the normal distribution of variably sized VWF multimers [16]. The lower VWF levels in blood group O individuals are possibly due to accelerated proteolysis in these persons.

Inherited VWD type 2A may be caused by mutations that interfere with the assembly or secretion of HMW multimers (e.g. impaired intracellular transport or defective posttranslational processing) or by mutations that render circulating multimers susceptible to proteolysis by ADAMTS-13 (e.g. mutations involving the A2 domain of VWF) [14, 15, 22, 23]. The acquired VWS type 2A has multiple associations, the most relevant in the present context being the selective degradation of HMW multimers in patients with cardiac defects [18, 19].

Under normal circumstances, the FVIII–VWF complex circulates as a loosely coiled protein complex and does not interact avidly with platelets or endothelial cells. However, if subendothelial collagen is exposed by vascular injury, VWF tethers to the exposed subendothelium. The high fluid shear milieu of the microcirculation also induces conformational changes in the multimeric molecule. VWF uncoils and attains JCD 2009; 1:(1). OCTOBER 2009

Distinguishing VWS type 2A from the more common inherited type 1 disorder (showing mild to moderately reduced levels of all VWF multimers) is also helpful since the association of the former with bleeding GI 12

www.slm-hematology.com


The genesis of GI angiodysplasia is postulated to be from an age-related progressive dilatation of the precapillary vascular sphincter. This occurs due to chronic, intermittent, low-grade obstruction of the submucosal veins at the point where they pierce the muscularis propria of the intestine. The obstructive pathology ultimately results in acquired anomalous arteriovenous communications with increased blood flow directly from arterioles to the dilated submucosal veins. These malformations are similar to the lesions of hereditary hemorrhagic telangiectasia or Osler Weber Rendau syndrome, and they may also be accurately termed vascular ectasias or telangiectasias [11, 25–29]. This pathogenetic hypothesis is functionally corroborated by the angiographic appearance of a normal arterial phase, the presence of an abnormal vascular bundle in the capillary phase and subsequent rapid filling and delayed drainage of large veins. Angiography performed during active bleeding additionally reveals extravasations of the contrast dye. The relative frequency of angiodysplasia in the cecum also possibly reflects greater wall tension coupled with absent serosal support in this location [11, 25–29]. Figure 1. Multimer analysis in a case of acquired VWS type 2A (right lane) shows selective loss of HMW multimers. NPP: Normal pooled plasma control

HEYDE SYNDROME, AND WARKENTIN’S HYPOTHESIS In 1958, in a letter to the New England Journal of Medicine, EC Heyde from Vancouver, Washington, USA, published 10 cases of coexisting calcific aortic stenosis and idiopathic severe gastrointestinal bleeding [12]. The combination of calcific aortic stenosis and iron deficiency anemia due to gastrointestinal bleeding was later eponymously termed Heyde syndrome. Also in 1958, a review by Goldman of 37423 hospital admissions over 6 years studied 256 patients with aortic stenosis and 780 patients with GI hemorrhage. He found a 3-fold higher than predicted incidence of gastrointestinal bleeding in cases of aortic stenosis. While chance alone would have accounted for about 5 cases with both aortic stenosis and GI hemorrhage, instead 15 cases of the association were seen [30]. In 1965, Cattell was said to have opined that patients with aortic stenosis bled from ascending colon lesions that were not demonstrable pathologically. He putatively recommended blind right hemi-colectomy in such patients for the treatment refractory anemia [11].

angiodysplasia is much stronger [19]. This further implicates the deficiency of HMW multimers of VWF as a predisposing factor to bleeding GI angiodysplasia.

GASTROINTESTINAL BLEEDING IN VON WILLEBRAND DISEASE AND ANGIODYSPLASIA Mucosal bleeding, including bleeding originating from the GI tract is fairly common in the more severe subtypes of VWD [24, 25]. A relatively frequent occurrence in patients with either inherited VWD or the acquired VWS, compared to age-matched subjects without an inherited bleeding disorder, is occult bleeding from submucosal vascular abnormalities including angiodysplasia. The underlying acquired platelet and coagulation abnormality in VWS may in fact, be underrecognized and go undiagnosed due to the presence of the anatomic lesion [20, 21, 24, 25]. Endoscopically and on gross examination, angiodysplasias are flat to vaguely elevated, bright red vascular lesions 2 to 12 mm in diameter consisting of clusters of radiating flares of submucosal peripheral vessels (arterioles, capillaries and venules) oriented around a prominent central draining vein. Histologically, despite their increased size, the vascular structures in angiodysplasia usually appear completely normal. The cecum and ascending colon are the most common sites of incidence although the lesions may occur anywhere in the small and large intestine or even the stomach [11, 25–27]. www.slm-hematology.com

By strict criteria today, Heyde syndrome would refer to a triad of aortic stenosis, acquired coagulopathy (VWS-2A) and anemia arising from bleeding intestinal angiodysplasia or bleeding of idiopathic (presumed GI) origin [31]. While the link between the valve defect and coagulopathy on the one hand, and the coagulopathy and the GI bleeding on the other appears well established, the link between the aortic stenosis and intestinal angiodysplasia has been controversial. Prior retrospective studies and cohort studies show conflicting conclusions [11, 32, 33]. The evidence for aortic 13

JCD 2009; 1:(1). OCTOBER 2009


Journal of Coagulation Disorders

stenosis being the primary cause of the coagulopathy is much more robust than the proof of its casual association with angiodysplasia.

septal defect, hypertrophic obstructive cardiomyopathy and aortic stenosis, or in primary pulmonary hypertension. This increased shear stress enhances the proteolysis of VWF by ADAMTS-13 to an extent sufficient to deplete large VWF multimers. A bleeding disorder resembling VWD type 2A is thereby generated. Supportive evidence for this pathogenetic theory is found in the fact that the VWF multimer distribution improves if the underlying cardiovascular condition is treated successfully [11, 19, 20, 36, 37].

In 1992, in a unifying theory, Warkentin et al were the first to suggest that aortic stenosis and angiodysplasia were both simply age-related processes and were unrelated except for the cause-and-effect relationship of VWS type 2A between them (Figure 2) [34]. This was based on the selective deficiency of HMW multimers in bleeding aortic stenosis patients that had been demonstrated experimentally previously [18, 19].

The predisposition to GI bleeding in VWS patients with a type 2A defect may be explained by pathologic microvascular features of angiodysplasia. The major mechanism involves the high fluid shear flow rates in the microcirculation. Under normal physiologic situations, the highest wall shear rates (.4000/second) occur in the arterioles. This wall shear rate is directly proportional to the blood flow velocity and inversely proportional to vessel diameter. Due to their fast blood flow and small size, angiodysplasias lead to increased wall shear rates. The very largest multimers of VWF are required to maintain hemostasis under such conditions. This would explain why many patients have bleeding GI angiodysplasia without a generalized bleeding tendency, because small and intermediatesized VWF multimers remain available for GpIb-VWFsubendothelial interactions at the lower physiologic shear rates [10, 18, 19, 24, 36, 37].

THE EPIDEMIOLOGICAL LINK BETWEEN VWD AND BLEEDING GI ANGIODYSPLASIA This link has been explored by many groups and is now well established. Apart from the studies by Goldman [30] above, Fressinaud et al, in an international survey of 4503 VWD patients from 297 centers, found the incidences of angiodysplasia to be 0% in type 1, 2% in type 2, 4.5% in type 3 and 11.7% in acquired VWD. The median patient age was 55 years for types 2 and 3 VWD and 69 years for acquired VWD. The increased prevalence of angiodysplasia in older patients suggested that the prevalence of angiodysplasia in older patients with VWD types 2 and 3 was probably similar to that in acquired VWS. Significantly, angiodysplastic bleeding only occurred in patients lacking HMW multimers of VWF, that is to say, patients with types 2 and 3 VWD [35].

INVESTIGATIONS FOR ANGIODYSPLASIA Various investigative modalities include fiberoptic endoscopy, push enteroscopy for small intestinal lesions, helical computed tomographic mesenteric angiography or radionucleotide scanning, video capsule endoscopy, histology on colectomy specimens and at autopsy. Each carries different sensitivities and specificities. Angiodysplasia can be particularly difficult to document pathologically, especially postmortem [11, 25–31].

PATHOPHYSIOLOGY OF BLEEDING IN VWS WITH ANGIODYSPLASIA Pathologic increases in fluid shear stress frequently occur within cardiovascular lesions, such as ventricular Age-related connective tissue degradation in the elderly Senile aortic stenosis, gastrointestinal angiodysplasia

INVESTIGATIONS FOR ACQUIRED VWD (VWS)

Peri-lesional high shear stress blood flow conditions generated in the cardiac defect

The rare VWS may not be considered a differential diagnostic possibility in a patient with a bleeding disorder or, even more commonly, with unexplained anemia. The detailed hemostatic history is frequently the most rewarding hemostatic screen and should always be elicited to confirm the acquired nature of the bleeding disorder. Bleeding symptoms are typically mucocutaneous (oronasopharyngeal, gastrointestinal, ecchymosis) or postoperative. Soft-tissue hematomas are less frequent [14, 20, 21, 23]. The identification of a coexisting, potentially pathogenetically linked disorder, such as dysproteinemia, a lymphoproliferative or myeloproliferative disorder, an autoimmune disorder, or hypothyroidism can provide a diagnostic clue to the presence of VWS, although most patients with angiodysplasia will not provide such a history [2–11].

HMW multimers of VWF uncoil VWF cleavage sites for ADAMTS13 are exposed Increased proteolysis with selective loss of HMW multimers (VWS type 2A) PLUS accelerated clearance by enhanced platelet-VWF binding Requirement for HMW multimers to maintain hemostasis in microcirculation compromised Mucosal bleeding from pre-existing age-related angiodysplasia Figure 2. The pathogenesis of Heyde syndrome (Warkentin and colleagues [11,34,40])

JCD 2009; 1:(1). OCTOBER 2009

14

www.slm-hematology.com


Laboratory testing should include the template bleeding time, screening tests of hemostasis (platelet count, prothrombin time, activated partial thromboplastin time, thrombin time) and specific tests for VWD (VWF antigen level, ristocetin cofactor activity, factor VIII coagulant activity, ristocetin-induced platelet aggregation and VWF multimer analysis). Distinguishing inherited from acquired VWD by laboratory testing alone is difficult and a clinical hemostatic history and family studies are usually the clinching factors. Measurement of VWF propeptide may be helpful in this distinction if inhibitors or accelerated destruction is the cause of acquired VWS (normal to high propeptide level). VWF propeptide or the VWF antigen II results from endoproteolytic cleavage of VWF in the endothelial cells before secretion and has a circulatory half-life of 2 hours [2, 12–14, 17, 21, 23, 38].

[41]. This approach is obviously only feasible for patients with detectable cardiac flow defects due to structural abnormalities.

SUMMARY LEARNING POINTS N Gastrointestinal bleeding from angiodysplasia in inherited or acquired VWD, sometimes with valvular heart defects, is rare but documented. N The coagulopathy in such cases is often found to be similar to inherited VWD type 2A, with selective loss of the very highest molecular weight multimers of VWF from circulation. N It is highly likely that the specific loss of these multimers occurs through the shear-dependent VWF cleaving metalloproteinase ADAMTS-13 as well as enhanced platelet binding of uncoiled HMW multimers resulting in their increased clearance from circulation. N HMW multimers are vital in platelet activation and thrombin generation for adequate hemostasis in high-pressure arteriovenous communications as seen in angiodysplasia. Their loss results in failure of procoagulant mechanisms with occult or manifest bleeding from the gastrointestinal lesions. N Correction of the valvular defects in Heyde syndrome results in correction of both the bleeding disorder as well as the multimeric profile, providing strong evidence in support of this proposed pathogenesis. N The diagnosis and management of bleeding angiodysplasia in acquired VWD remain challenging.

Up to 20% of patients with VWS may have inhibitors of VWF (usually detected as inhibition of ristocetin cofactor activity). Inhibitor screening is typically performed by mixing the patient’s and normal plasmas, followed by comparison of the observed and expected ristocetin cofactor activity. This test may be extended by heating the test plasma at 56 ˚C for in vitro dissociation of immune complexes. This results in diminished residual VWF/RCo (determined after incubation at 37 ˚C of serial dilutions of test plasma mixed with normal pooled plasma). However, antibodies directed against nonfunctional VWF domains may be undetectable by these approaches. Staphylococcal protein A and crossed immunoelectrophoresis have also been explored [2, 12–15, 17, 21, 23, 37–39].

Disclosures: The authors have no financial interests to disclose related to the contents of this article.

TREATMENT OPTIONS FOR BLEEDING ANGIODYSPLASIA IN VWS/VWD

REFERENCES 1. Rodeghiero F, Castaman G. von Willebrand disease: epidemiology. In: Lee CA, Berntorp EE, Hoots WK, eds. Textbook of Hemophilia. Oxford: Blackwell Publishing Ltd; 2005:265–271. 2. Nitu-Whalley IC, Lee CA. Acquired von Willebrand syndrome: A report of 10 cases and review of the literature. Haemophilia. 1999; 5:318–326. 3. Wautier JL, Levy-Toledano S, Caen JP. Acquired von Willebrand’s syndrome and thrombopathy in a patient with chronic lymphocytic leukaemia. Scand J Haematol. 1976;16:128. 4. Mahapatra M, Mishra P, Makharia G, Kumar R, Saxena R. Acquired von Willebrand’s disease associated with gastrointestinal angiodysplasia and monoclonal gammopathy. J Assoc Physicians India. 2006;54:963. 5. Simone JV, Cornet JA, Abildgaard CF. Acquired von Willebrand’s syndrome in systemic lupus erythematous. Blood. 1968;31:806–812. 6. Facon T, Caron C, Courtin P, et al. Acquired type II von Willebrand’s disease associated with adrenal cortical carcinoma. Br J Haematol. 1992;80:488–494. 7. Budde U, Van Genderen PJJ. Acquired von Willebrand disease in patients with high platelet counts. Semin Thromb Hemost. 1997;23:425–431. 8. Dalton RG, Savidge GF, Matthews KB. Hypothyroidism as a cause of acquired von Willebrand’s disease. Lancet. 1987;1:1007. 9. Kreuz W, Linde R, Funk M, et al. Induction of von Willebrand type I by valproic acid. Lancet. 1990;335:1350–1351.

Therapeutic options for angiodysplasia include surgical recourses such as endoscopic therapy (thermocoagulation, electrocoagulation and laser photocoagulation), angiographic therapy with vasopressin infusion, catheter embolization or surgical resection including partial or extended colectomy. Medical therapeutic options include oral or frequently, intravenous iron replacement therapy, estrogen, progesterone, octreotide, anti-fibrinolytics, thalidomide or VWF/factor VIII concentrates. Prophylactic treatment with VWF/FVIII concentrate may also be considered [25–27]. Warkentin et al reported durable reversions of clinical and biologic hemostatic abnormalities lasting over 10 years in 2 patients who had undergone surgical treatment for severe aortic stenosis with acquired VWS and bleeding [40]. In a study by Vincentelli et al on 50 patients, aortic valve replacement in patients with stenosis also resulted in significant improvement in biologic parameters at 6 months. However, not all of their patients had a complete, sustained correction www.slm-hematology.com

15

JCD 2009; 1:(1). OCTOBER 2009


Journal of Coagulation Disorders

35. Fressinaud E, Meyer D. International survey of patients with von Willebrand disease and angiodysplasia. Thromb Haemost. 1993:70:546. 36. Siedlecki CA, Lestini BJ, Kottke-Marchant KK, et al. Sheardependent changes in the three-dimensional structure of human von Willebrand factor. Blood. 1996;88:2939–2950. 37. Tsai HM, Sussman II, Nagel RL. Shear stress enhances the proteolysis of von Willebrand factor in normal plasma. Blood. 1994;83:2171–2179. 38. Fressinaud E, Meyer D. von Willebrand disease: biological diagnosis. In: Lee CA, Berntorp EE, Hoots WK, eds. Textbook of Hemophilia. Oxford: Blackwell Publishing Ltd; 2005:272–278. 39. Kumar S, Pruthi R, Nichols WS. Acquired von Willebrand Disease. Mayo Clin Proc. 2002;77:181–187. 40. Warkentin TE, Moore JC, Morgan DG. Gastrointestinal angiodysplasia and aortic stenosis. N Engl J Med. 2002;347:858–859. 41. Vincentelli A, Susen S, Le Tourneau T, et al. Acquired von Willebrand syndrome in aortic stenosis. N Engl J Med. 2003;349: 343–349.

10. Castaman G, Lattuada A, Mannucci PM, Rodeghiero F. Characterization of two cases of acquired transitory von Willebrand syndrome with ciprofloxacin: evidence for heightened proteolysis of von Willebrand factor. Am J Hematol. 1995;49:83–86. 11. Warkentin TE, Moore JC, Anand SS, Lonn EM, Morgan DG. Gastrointestinal bleeding, angiodysplasia, cardiovascular disease, and acquired von Willebrand syndrome. Transfus Med Rev. 2003;17:272–286. 12. Heyde EC. Gastrointestinal bleeding in aortic stenosis. N Engl J Med. 1958;259:196. 13. Gupta PK, Kannan M, Chatterjee T, et al. Acquired von Willebrand’s disease associated with gastrointestinal angiodysplasia: a case report. Haemophilia. 2006;12:452–455. 14. Sadler JE, Budde U, Eikenboom JC, et al. Update on the pathophysiology and classification of von Willebrand disease: a report of the Subcommittee on von Willebrand Factor. J Thromb Haemost. 2006;4(10):2103–2114. 15. Sadler JE. von Willebrand factor. J Biol Chem. 199;266:22777– 22780. 16. Moake JL. von Willebrand factor, ADAMTS-13, and thrombotic thrombocytopenic purpura. Semin Hematol. 2004;41:4–14. 17. Mann KG, Ziedins KB. Overview of hemostasis. In: Lee CA, Berntorp EE, Hoots WK, eds. Textbook of Hemophilia. Oxford: Blackwell Publishing Ltd; 2005:1–4. 18. Pickering NJ, Brody JI, Barrett MJ. von Willebrand syndromes and mitral-valve prolapse: linked mesenchymal dysplasias. N Engl J Med. 1981;305:131–134. 19. Gill JC, Wilson AD, Endres-Brooks J, Montgomery RR. Loss of the largest von Willebrand factor multimers from the plasma of patients with congenital cardiac defects. Blood. 1986;67:758–761. 20. Rinder MR, Richard RE, Rinder HM. Acquired von Willebrand disease: a concise review. Am J Hematol. 1997;54:139–145. 21. Tefferi A, Nichols WL. Acquired von Willebrand disease: concise review of occurrence, diagnosis, pathogenesis and treatment. Am J Med. 1997;103:536–540. 22. Mancuso DJ, Tuley EA, Westfield LA, et al. Structure of the gene for human von Willebrand factor. J Biol Chem. 1989;264:19514– 19527. 23. Budde U, Schneppenheim R. von Willebrand disease: molecular aspects. In: Lee CA, Berntorp EE, Hoots WK, eds. Textbook of Hemophilia. Oxford: Blackwell Publishing Ltd; 2005:257–264. 24. Federici AB. Clinical diagnosis of von Willebrand disease. Haemophilia. 2004;10:169–176. 25. Makris M. Gastrointestinal bleeding in von Willebrand disease. Thromb Res. 2006;118:S13–S17. 26. Howard OM, Buchanan JD, Hunt RH. Angiodysplasia of the colon: Experience of 265 cases. Lancet. 1982;2:16–19. 27. Richter JM, Christensen MR, Colditz GA, et al. Angiodysplasia. Natural history and efficacy of therapeutic interventions. Dig Dis Sci. 1989;34:1542–1546. 28. Boley SJ, Sprayregen S, Sammartano RJ, et al. The pathophysiologic basis for the angiographic signs of vascular ectasias of the colon. Radiology. 1977;125:615–621. 29. Boley SJ, Sammartano R, Adams A, et al. Degenerative lesions of aging. Gastroenterology. 1977;72:650–660. 30. Goldman MJ. Aortic stenosis and gastrointestinal bleeding. N Engl J Med. 1958;259:941. 31. Massyn MW, Khan SA. Heyde syndrome: a common diagnosis in older patients with severe aortic stenosis. Age Ageing. 2009;38: 267–270. 32. Bhutani MS, Gupta SC, Markert RJ, et al. A prospective controlled evaluation of endoscopic detection of angiodysplasia and its association with aortic valve disease. Gastrointest Endosc. 1995;42:398–402. 33. Greenstein RJ, McElhinney AJ, Reuben D, et al. Colonic vascular ectasias and aortic stenosis: coincidence or causal relationship? Am J Surg. 1986;151:347–351. 34. Warkentin TE, Moore JC, Morgan DG. Aortic stenosis and bleeding gastrointestinal angiodysplasia: is acquired von Willebrand’s disease the link? Lancet. 1992;340:35–37.

JCD 2009; 1:(1). OCTOBER 2009

16

www.slm-hematology.com


JOURNAL OF COAGULATION DISORDERS

REVIEW ARTICLE

Renal Hemophilic Pseudotumors Emel Gu¨rkan Affiliation: Department of Haematology, C ¸ ukurova University Medical School, Adana, Turkey Submission date: 1st July 2009, Revision date: 2nd August 2009, Acceptance date: 10th August 2009

A B S T R A C T Severe hemophilic patients are prone to develop long-term sequelae such as pseudotumor formation caused by repeated and unresolved hematomas leading to subsequent encapsulated mass lesion. Among these, renal hemophilic pseudotumors are seen as a rather rare event. Once developed, this serious condition should be approached carefully, first by maintaining adequate hemostasis with appropriate factor replacement therapy and, later on, complete surgical excision should be targeted. Without surgical intervention, complete resolution is generally not possible. Other treatment modalities such as radiotherapy may be helpful in addition to surgery preoperatively in order to reduce bleeding complications. However, experience of this approach is limited to those pseudotumors of other sites such as bone and pelvic pseudotumors. Rarely, a true pseudotumor may mimic a malignancy; therefore, accurate assessment by radiological means is required. Considering the difficulties in management, it is of great importance to constitute early substitutive treatment for the prevention of pseudotumors. Keywords: renal pseudotumor, hemophilia, management Correspondence: Emel Gu¨rkan, C ¸ ukurova University Faculty of Medicine, Department of Hematology, 01330 Adana, Turkey. Tel: +90-322-3386060; e-mail: egurkan@cu.edu.tr

the location, two types of pseudotumors have been described; proximal pseudotumors mainly occur in adults, in the pelvis and femur, and distal pseudotumors mostly involve the hands and feet [4]. In proximal pseudotumors, the pattern of formation is mainly caused by repeated and unresolved hematomas leading to subsequent encapsulation with progressive enlargement of the mass. In adults, most pseudotumors are localized within the long bones. Distal pseudotumors usually develop more rapidly and are found mainly in children and adolescents [5]. Conversely, proximal pseudotumors develop slowly, sometimes over years, and are seen in adults.

INTRODUCTION Hemophilias are inherited bleeding disorders with a reported incidence of 1 in 5000 live births exclusively affecting the male population. Two disorders, hemophilia A and hemophilia B, both referred to by the term ‘‘hemophilia’’, show similar bleeding problems. Approximately 85% of all hemophilias exhibit factor VIII deficiency (hemophilia A), and the remaining 15% have factor IX deficiency (hemophilia B) [1]. The degree of severity of clinical manifestations varies among patients. The level of factor deficiency correlates well with disease severity. Joints and muscles are common bleeding sites in most hemophilia patients. Contractures, pain, and limitation of motion are debilitating late complications influencing quality of life in hemophilia patients. Bleeding into muscles and soft tissues comprises 10–30% of all bleeding episodes in hemophilia patients [2, 3]. Occasionally, these bleeding episodes become pseudotumors in time. A pseudotumor is formed by encapsulation of a hematoma with fibrous layers [4]. Pseudotumors in hemophilia patients usually involve the extremities and, less frequently, the pelvic region. There are also reports of the formation of pseudotumors in rarer sites such as the paranasal sinuses, mandible, and the orbital region [5]. A hemophilic pseudotumor is a rare complication defined as a progressive cystic swelling involving muscle and bone [6]. Clinically, according to JCD 2009; 1:(1). OCTOBER 2009

PATHOGENESIS AND CLINICAL FINDINGS The incidence of pseudotumor formation is reported as 1–2% in patients with severe hemophilia [7–9]. Renal hemophilic pseudotumor formation is very rare event so that identification of the risk factors for the development of such a lesion is not widely known. The only reported case in the literature developed spontaneously [10]. It is likely that minor traumas leading to silent hematomas organize over time without complete resolution. Progressive cystic swelling with repeated hemorrhages into the cavity results in encapsulation of the hematoma with fibrous tissue. Vascular tissue may envelope underlying capsule in the outer region. Calcifications or ossifications can also be seen [4]. As 17

www.slm-hematology.com


Journal of Coagulation Disorders

in other cases with renal pseudotumors, those in hemophilia patients that are histologically normal can lead to confusion with a true ‘‘tumor’’ [4]. Renal pseudotumors may have the potential to destroy renal function over time causing a mass effect. Clinically, progression is very slow and usually asymptomatic until they become palpable on physical examination or detected incidentally by imaging procedures. Painless asymptomatic hematuria can be seen. Lesions may appear multilocular. Such pseudotumors frequently remain painless and asymptomatic. Even giant mass lesions can be seen (Figure 1). Abdominal ultrasonography is very helpful in diagnosis. Lobulation or a septated fibrous structure with the encapsulated pseudotumor can easily be detected. Computed tomography (CT) and/or magnetic resonance imaging (MRI) of the abdominal structures are usually required before surgical intervention. It is important to investigate primary malignancies, as mass lesions could be mistaken for pseudotumors of hemophilia. There are reports in the literature demonstrating that malignancies such as sarcomas and non-Hodgkin lymphoma may mimic a pseudotumor in hemophilia patients [11]. Differentiation of renal hemophilic pseudotumor from malign lesions such as renal cell carcinoma is easily made depending on MRI findings and clinical history. A high-quality CT or MRI examination usually differentiates renal pseudotumors from renal neoplasms [12]. It is of note that there are many benign and malignant conditions mimicking pseudotumors radiologically including fibrosarcoma, plasmacytoma, malignant fibrous histiocytoma, telangiectatic osteosarcoma, and metastatic disease from primary tumors in organs [13, 14]. Benign bone tumors such as aneurysmal bone cysts and giant cell tumors are considerations, as well as unusual infectious processes such as echinococcosis [15]. Tumor-like lesions in outward appearance, such as a hernia on the body

surface, have also been reported [16, 17]. Plain radiographic films are not usually informative in cases of renal pseudotumors. If present, calcifications surrounding a soft tissue mass could be recognizable. The kidney may be found to be enlarged on physical examination and imaging procedures. Strenuous palpation of the intra-abdominal mass lesion should be avoided in hemophilia patients, as this may potentially cause new hemorrhages by increasing intracystic pressure and may even result in rupture of the capsule of a renal pseudotumor. A careful preliminary workup should also be performed before proceeding to surgery because hemophilia patients with factor inhibitors are more prone to develop pseudotumors.

MANAGEMENT The management of proximal pseudotumors consists of embolization, irradiation, percutaneous management, or surgical removal [4]. Treatment approaches including immobilization factor replacement therapy seem to be inadequate as these applications do not always prevent the development and organization of a pseudotumor [2]. Renal pseudotumors secondary to causes other than hemophilia are usually benign and require no treatment [18]. In contrast, pseudotumors originating from renal structures in hemophilia patients should be treated immediately. Joint bleeding usually resolves with appropriate factor replacement therapy; however, this is not always the case for soft tissue hematomas in hemophilia patients. Preoperative preparation of the patient should be done cautiously in order to provide adequate hemostasis. This should target a factor level of 100% in the perioperative period, which should be maintained at around 40–50% postoperatively for at least 14 days. Patients should be monitored carefully for recurrences by routine imaging techniques. Inadequate resection may lead to recurrences. Prophylactic factor treatment should be considered according to the individual patient. Despite factor replacement therapy with other supportive modalities helping in the resolution of early intramuscular bleeding episodes, in the case of chronic hematomas and pseudotumors, surgical excision may still be unavoidable in order to prevent the erosion of adjacent tissue structures. Radiotherapy is recommended in hemophilic bone pseudotumors both alone or in combination when it is impossible to perform complete surgical excision of the lesion. Recommended radiation dosage is small fractions of 2 Gy to the lesion site or less to a total dosage of 6–23.5 Gy [5]. Therapeutic arterial embolization of hemophilic pseudotumors is considered in large lesions especially in pseudotumors of the pelvic region to reduce tumor size and bleeding complications perioperatively [5]. Radiotherapy might be considered as an option for renal hemophilic pseudotumors before surgery. On the other hand, therapeutic arterial

Figure 1. Giant encapsulated intrarenal hematoma in a severe hemophilia A patient

JCD 2009; 1:(1). OCTOBER 2009

18

www.slm-hematology.com


embolization for renal pseudotumors seems unfeasible because of imminent renal ischemia. There is a considerable risk of massive bleeding due to the highly vascularized capsule of the pseudotumor. In the author’s view, it seems reasonable to use radiotherapy before surgery especially when the lesion size is large and it is not completely resectable. We had a 30-year-old severe hemophilic patient in our centre who was admitted for mild abdominal pain and intermittent hematuria. There was a homogeneous, lobular, well-defined cystic lesion located through the left of the abdominal cavity in close proximity to the left kidney, which was detected ultrasonographically 2 years ago and known not to exist before. In his medical history, he experienced mild–moderate intensity bleeding episodes once in every 3–5 years, which were managed on episodic basis. He had a percutaneous drainage operation for a left iliopsoas hematoma 10 years before. During follow-up, he was hospitalized for hematuria and received factor replacement therapy. Six months later, he experienced abdominal pain, and hematuria restarted prior to hospitalization. A trial for percutaneous evacuation was not successful and complicated by a localized infection. Surgical intervention performed later was very effective as the pseudotumor mass was completely excised. He has been followed up for 5 years without any recurrence. There was moderate bleeding during the operation, which was compensated with six packs of erythrocyte suspension. The amount of total factor replacement was as huge as the tumor size (Figure 1), i.e., approximately 150 000 units of factor VIII was infused. Fortunately, our patient did not develop inhibitor formation later on. Maintenance was done for just a month with factor VIII, and tranexamic acid was continued for somewhat longer. To the best of our knowledge, our case is unique in the literature, and no similar case has been reported related to renal hemophilic pseudotumor [10]. Our case was unique for its location and demonstrates that untreated hemorrhages, wherever they occur, may lead to the formation of a pseudotumor.

of the lesions. Therefore, it is important to emphasize the role of early substitutive treatment for the prevention of pseudotumors, which are otherwise very challenging to manage. Disclosure: The author has no financial interests to disclose related to the contents of this article.

REFERENCES 1. White GC 2nd, Rosendaal F, Aledort LM, et al. Factor VIII and Factor IX Subcommittee. Definitions in hemophilia. Recommendation of the scientific subcommittee on factor VIII and factor IX of the scientific and standardization committee of the International Society on Thrombosis and Haemostasis. Thromb Haemost. 2001;85:560. 2. D’Young AI. Conservative physiotherapeutic management of chronic haematomata and haemophilic pseudotumours: case study and comparison to historical management. Haemophilia. 2009;15:253–260. 3. Beeton K, Alltree J, Cornwall J. Rehabilitation of muscle dysfunction in haemophilia. Haemophilia. 1990;4:532–537. 4. Rodriguez-Merchan EC. Haemophilic cysts (pseudotumours). Haemophilia. 2002;8:393–401. 5. Espandar R, Heidari P, Rodriguez-Merchan EC. Management of haemophilic pseudotumours with special emphasis on radiotherapy and arterial embolization. Haemophilia. 2009;15:448–457. 6. Rodriguez-Merchan EC. The haemophilic pseudotumour. Haemophilia. 2002;8:12–16. 7. Mohanty SS, Bhasme VK, Garg H, et al. The haemophilic pseudotumour – surgical treatment by excision and filling the defect with calcium-phosphate cement granules. Haemophilia. 2007;13:217–220. 8. Gupta S, Mohapatra BB, Gihai Set al. Haemophilic pseudotumour of the paranasal sinuses: management with radiotherapy and factor replacement therapy. Haemophilia. 2001;7:595–599. 9. Maclachlan J, Gough-Palmer A, Hargunani R, et al. Haemophilia imaging: a review. Skeletal Radiol. 2008; epub Sep 20. ¨ c¸al F. Renal haemophilic pseudotumour. 10. Gu¨rkan E, O Haemophilia. 2005;11:559–560. 11. Allen DJ, Goddard NJ, Mann HA, et al. Primary malignancies mistaken for pseudotumours in haemophilic patients. Haemophilia. 2007;13:383–386. 12. Israel GM, Bosniak MA. How I do it: evaluating renal masses. Radiology. 2005;236:441–450. 13. Magallo´n M, Monteagudo J, Altisent C, et al. Hemophilic pseudotumor: multicenter experience over a 25-year period. Am J Hematol. 1994;45:103–108. 14. Allen DJ, Goddard NJ, Mann HA, et al. Primary malignancies mistaken for pseudotumours in haemophilic patients. Haemophilia. 2007;13:383–386. 15. Chang CC, Chen YC, Pan RY, et al. A hemophiliac pseudotumor mimicking an incisional hernia in outward appearance. Eur J Haematol. 2008;81:79–80. 16. Chang CC, Chang ST, Chang HY, et al. Multicentric giant cell tumours in an adolescent with haemophilia. Haemophilia. 2007; 13:199–201. 17. Chang CC, Chen YC, Pan RY, et al. A hemophiliac pseudotumor mimicking an incisional hernia in outward appearance. Eur J Haematol. 2008;81:79–80. 18. Wespes E, Van Gansbeke D, Schulman CC. Renal pseudotumors. World J Urol. 1984;2:89-91.

CONCLUSIONS Repeated bleeding into muscles and other soft tissues may lead to a serious long-term sequela of the disease defined as pseudotumors. Renal pseudotumours as well as other proximal pseudotumors grow slowly and may achieve large sizes. A few reports in the literature suggest that these are more commonly seen in adult patients with severe hemophilia and more frequently in those patients with factor inhibitors. These do not usually respond well to conservative treatment such as immobilization and factor replacement therapy. Renal pseudotumors generally are clinically asymptomatic, painless lesions. They may cause intermittent hematuria. Delays in the treatment of such cases are complicated by bleeding, infection, or even fistulization www.slm-hematology.com

19

JCD 2009; 1:(1). OCTOBER 2009


JOURNAL OF COAGULATION DISORDERS

REVIEW ARTICLE

Prophylaxis in Patients with Severe Hemophilia and Inhibitor Victor Jimenez-Yuste1,2, Maria Teresa Alvarez2, Monica Martin-Salces2, E Carlos Rodriguez-Merchan3, Nora Butta4, Ihosvany Fernandez-Bello4, Isabel Rivas4 and Ana Rodriguez de la Rua2 Affiliations: 1Department of Hematology, Autonoma University, Madrid, Spain; 2Hematology Department, Hospital Universitario La Paz, Madrid, Spain; 3Department of Orthopedics and Hemophilia Unit, Autonoma University, Madrid, Spain; 4 Research into Hemostasis Unit, Hospital Universitario La Paz, Madrid, Spain Submission date: 4th July 2009, Revision date: 4th August 2009, Acceptance date: 17th September 2009

A B S T R A C T Joint hemorrhage, the most common manifestation of severe hemophilia, frequently leads to recurrent hemarthrosis, chronic synovitis, and arthropathy. The main goal of replacement therapy is to prevent this pathology. On-demand treatment slows, but does not prevent, the progression of arthropathy. It has been shown that prophylaxis is superior to enhanced, episode-based therapy in preventing joint damage. However, patients often develop antibodies that inhibit or neutralize factor VIII (FVIII) or factor IX (FIX) replacement therapy in what is today the most serious complication of hemophilia and its treatment. The effects of bypassing agents in treating the bleeding episodes are not as predictable as those of replacement therapy with the deficient factor in patients with inhibitors. Consequently, these patients have higher levels of arthropathy than patients without inhibitors. Prophylaxis for patients with inhibitors has gained attention over the last decade, and some papers have reported that bypassing agents may help to prevent arthropathy. Nevertheless, the justification of prophylaxis in patients with inhibitors is still questioned because of concerns regarding its cost, complications, and efficacy. Keywords: Haemophlia, inhibitors, prophylaxis, arthropaty Correspondence: Victor Jimenez-Yuste, Servicio de Hematologı´a y Hemoterapia, Hospital Universitario La Paz, Paseo de la Castellana 261, 28046 Madrid, Spain. Tel: +34-91-207-1945; fax: +34-91-7277225; e-mail: vjimenez.hulp@salud.madrid.org

Mainstay therapies for acute bleeds are: agents that ‘‘bypass’’ the need for FVIII or FIX (activated prothrombin complex concentrates (aPCC); FEIBAH; Baxter Bioscience, Vienna, Austria) and recombinant activated factor VII ((rFVIIa) NovoSevenH; Novo Nordisk, Bagsvaerd, Denmark). However, the effects are not as predictable as replacement therapy with the deficient factor, so the control of bleeding complications is still more difficult than in patients without inhibitors [7]. This is why inhibitor patients have more severe joint morbidity than patients without inhibitors, and older adults experience significant orthopedic disabilities [8, 9]. Because of the serious and disabling consequences of persistent inhibitors, there is considerable clinical and research interest in establishing effective bypassing agent regimens that will prevent bleeding in inhibitor patients in much the same way that prophylaxis works in non-inhibitor patients [6, 10, 11].

INTRODUCTION Patients with hemophilia A often develop severe bleeding complications in multiple joints, which lead to hemophilic arthropathy, and this situation may be prevented by prophylactic replacement therapy. Prophylactic therapy has shown that it can reduce the number of hemarthroses and decrease the risk of joint damage in boys with severe hemophilia compared with enhanced, episode-based treatment [1]. The results from the early studies on prophylaxis by Nilsson and colleagues in Sweden indicated that prophylaxis should begin at a very early age, before the joints were affected [2, 3]. Nowadays, this proposal for primary prophylaxis has became the standard of care in many countries [4]. Nevertheless, the development of antibodies that inhibit or neutralize replacement therapy with factor VIII (FVIII) or factor IX (FIX) is today the most serious complication of hemophilia and its treatment. Inhibitors occur in 20–30% of patients with severe hemophilia A but in only 5% of patients with hemophilia B [5, 6]. JCD 2009; 1:(1). OCTOBER 2009

This article will review the reported experiences in the prevention of arthropathy in patients with inhibitors, focusing on issues related to the efficacy and safety of bypassing agents. 21

www.slm-hematology.com


Journal of Coagulation Disorders

Table 1. FEIBA Experiences in Prophylaxis Patients

dose (U/kg)

Kreuz [18]

22

50–100

Daily

Mean 1 bleed/year (0–6)

No

Dimichele and Negrier [12]

14

15–100

Daily–weekly

Mean bleeding decrease 53%

No

Leissinger [21]

5

50–100

Daily–three doses per week

Mean bleeding decrease 73–83%

No

Escuriola-Ettingshausen [22]

7

50–100

12 h–3 weeks

Mean 2.5 bleeds (0–8)

Jimenez-Yuste [11]

5

50

Valentino [26]

6

50–100

Three times/week–every 48 h Daily–every third day

Mean number of bleeds before 14 and during 5 Mean reduction in bleeding 83%

First author (reference)

Frequency

Adverse events

Catheter sepsis, 5 Cutaneous infections, 2 No No

FEIBA in periods of up to 12 years has been safe and effective and no child in the study series had presented with an important bleeding episode [19].

FEIBA EXPERIENCE (Table 1) Clinical studies have shown that FEIBA is effective in controlling more than 80% of bleeding episodes, including acute and perioperative bleeding [12].

DiMichele and Negrier [12] conducted a retrospective post-licensure survey of FEIBA, concluding that prolonged prophylactic treatment with FEIBA decreased the frequency of bleeding episodes and was both safe and effective. Their study analyzed data on 14 patients who received FEIBA and included information on 15 prophylactic treatment periods. The duration of prophylaxis was available in 12 cases (mean 19.5 months; range 0.25–26 months); half of these were short term (,6 months, n56). Before starting prophylaxis, 12 of the 14 patients exhibited evidence of arthropathy in their knees or elbows. Six patients had developed at least one target joint. The mean dose per FEIBA infusion was 69 U/kg (range 15–100 U/kg), and the frequency of administration varied from once per day to once per week, with most patients being dosed every other day. The frequency of bleeding episodes decreased in ten prophylactic treatment periods and remained unchanged in three; information was incomplete for two periods. Quantitative data regarding the reduction in bleed frequency were available for eight treatment periods and showed a mean decrease of 53% (range 10– 85%). Clinical joint status was unchanged in eight patients, improved in three patients, but worsened in two patients. The mean inhibitor titer decreased from 99 BU immediately prior to prophylaxis to 11 BU during prophylaxis in the nine patients for whom the inhibitor titer data were available.

Nevertheless, there are concerns about the safety of long-term continuous administration of FEIBA, especially regarding thrombotic complications. The incidence of adverse thrombotic events reported in pharmacovigilance and post-licensure studies is low, ranging from 4 to 8.5 per 100 000 FEIBA infusions [13, 14]. Another concern arises from the anamnestic response in the inhibitor titers associated with residual FVIII antigen contained in FEIBA. However, anamnesis does not interfere with the effectiveness of FEIBA and, moreover, inhibitor titers do not necessarily remain elevated during ongoing FEIBA therapy, often declining over time [15]. Nevertheless, an anamnestic response could be a drawback when immune tolerance induction (ITI) is postponed to allow the inhibitor titer to decline to ,10 BU and, usually, FEIBA is not recommended as first-line prophylaxis because of its potential to prolong anamnesis [16]. Different authors have explored the benefits of longterm continuous aPCC administration in different clinical situations, for instance in association with FVIII in ITI or as the only agent in patients who had either failed or were not candidates for ITI. FEIBA has been a component of the Bonn ITI protocol for 30 years [17]. Kreuz et al [18] prospectively evaluated FEIBA prophylaxis during ITI in 22 children. The FEIBA dose varied from 50 to 100 U/kg/day. The median annual incidence of joint bleeding at the time of the study was one per patient, and no boy suffered a life-threatening hemorrhage. No evidence of arthropathy was seen in six of the eight boys who were evaluated radiographically; in fact, joint pathology in the two boys with arthropathy was minimal. The investigators concluded that this experience compares favorably with data on severe hemophilia patients without inhibitors who were receiving prophylaxis [18]. The same group did a prospective study of five children who had failed ITI. They observed that JCD 2009; 1:(1). OCTOBER 2009

Response during prophylaxis

Hilgartner et al [20] retrospectively described FEIBA prophylaxis in seven inhibitor patients (aged 4– 10 years) with target joints. FEIBA was administered at a dose ranging from 50 to 100 U/kg and from three times per week to every other day for up to 6.5 years. Joints that were normal at the beginning of the study remained normal; however, previously documented joint disease continued to progress despite prophylaxis. Most of the patients selected had multiple target joints at the beginning of their prophylaxis, suggesting that significant joint damage was already present and may have contributed to recurrent bleeding during the study. 22

www.slm-hematology.com


Leissinger et al [21] retrospectively evaluated the prophylactic use of FEIBA to reduce the frequency of bleeding episodes and to slow or prevent joint damage in five inhibitor patients. Median treatment duration was 15 months, and dosages ranged from 50 to 75 U/kg three times per week in four patients, while one patient received 100 U/kg daily. In three patients, range of motion (ROM) before prophylaxis was normal in all joints and remained so at the end of the study. The number of total and joint bleeds was reduced in all three patients while they were receiving prophylaxis. One patient with advanced arthropathy at the start of prophylaxis had functional improvement over the observed prophylaxis. The fifth patient already had severe joint disease, and prophylaxis had a protective effect on his remaining musculoskeletal function. The frequency of bleeding episodes was reduced in all patients. The authors concluded that FEIBA can be effective in reducing the frequency of bleeding episodes and maintaining or improving orthopedic status, especially in patients with normal joint function.

associated with bypassing agent treatment. FEIBA was used in all the patients without association with ITI. Two of the five patients presented increased inhibitor levels at the end of prophylaxis. No thromboembolic complications were detected in any patient.

Escuriola-Ettingshausen et al [22] evaluated longterm FEIBA prophylaxis in seven patients given 50– 100 U/kg FEIBA three times weekly up to as much as twice daily over 0.1–13.1 years. The median annual incidence of spontaneous joint bleeding was 2.5 episodes. There were no thrombotic complications. Ewenstein et al [23], in another study of FEIBA prophylaxis in 14 patients, reported no adverse event in any patient treated with FEIBA at doses ranging from 15 to 100 U/kg administered every other day or every third day for up to 26 months. However, they did observe a rise in the inhibitor titer in some patients 1– 3 months after beginning prophylaxis. Later, after 6 months of treatment, the inhibitor titer decreased and stabilized at levels comparable to or lower than the baseline titers. Siegmund et al [24] and Schino et al [25] have reported similar results.

rFVIIA EXPERIENCE (Table 2)

More recently, Valentino [26] retrospectively evaluated the use of FEIBA in six patients with hemophilia A or B and high-titer inhibitors. When FEIBA was administered regularly, most of them had a reduction in the number of hemorrhages, an improvement in orthopedic status, and an improvement in quality of life. He also observed in one patient that, in comparison to on-demand therapy, FEIBA prophylaxis was associated with a 25% reduction in monthly cost, suggesting that long-term aPCC prophylaxis could be a cost-effective treatment option for some patients with high-titer inhibitors, particularly those who bleed frequently. This author concluded that prophylaxis with FEIBA can reduce hemorrhages and halt further joint deterioration in patients with hemophilia and inhibitors.

Nowadays, rFVIIa is a well-established and effective hemostatic treatment for hemophilia patients who develop inhibitors to FVIII or FIX. rFVIIa has been employed prophylactically in hemophilia patients with inhibitors. Early experiences come from Brackmann et al [27], who described the use of rFVIIa in association with ITI in four patients ranging in age from 0.5 to 26 years and with a dosage regimen varying from 90 mg/kg administered twice a day up to two or three times a week. The duration of prophylaxis ranged from 2 to 27 months. The decrease in bleeding episodes was slightly lower than when using an aPCC prophylaxis. No adverse events were reported. Cooper et al [28] described secondary prophylaxis with rFVIIa over 80 days during physiotherapy in a patient with severe hemophilia B and inhibitor. In order to reduce knee flexion contractures, a short-term secondary prophylaxis was used. Bleeding was controlled by starting with 240 mg/kg rFVIIa (every 6–8 h) and then the dose interval was increased (every 12– 24 h). Two bleeds were reported in the 11 weeks of prophylaxis and the patient’s knee flexion improved.

Jimenez-Yuste et al [11] performed a retrospective analysis of the efficacy, cost, and safety of FEIBA in five inhibitor patients with hemophilia. The median duration of prophylaxis was 13 months, and median age at initiation of prophylaxis was 11 years. The median number of bleeds per patient prior to prophylaxis was 14 and 5 during prophylaxis. The number of bleeding episodes decreased in four out of five patients. The total number of bleeding episodes increased in only one patient. At the end of prophylaxis, two patients showed an improvement in orthopedic status and three patients had the same as before. None of the patients had life-threatening bleeds during prophylaxis. The median cost of factor treatment per patient per month in this study was higher before prophylaxis than during prophylaxis. In three of the five patients, the cost of prophylaxis was less than their previous on-demand treatment. In the other two patients, the cost of prophylaxis was also less than ITI www.slm-hematology.com

Saxon et al [29] reported on one patient receiving short-term secondary rFVIIa prophylaxis over 21 weeks. Prior to prophylaxis, the patient had had 2.1 bleeds per week in his right ankle. Prophylaxis with rFVIIa (90 mg/kg daily) resulted in a decrease in the frequency of bleeds. During the prophylaxis period, the patient experienced a mean of 0.4 bleeds per week. No adverse events were reported. Bryant et al [30] reported the use of secondary prophylaxis with rFVIIa during 5 months in a patient with severe hemophilia A and inhibitor. Prophylaxis with 300 mg/kg rFVIIa three times a week associated with 23

JCD 2009; 1:(1). OCTOBER 2009


Journal of Coagulation Disorders

Table 2. rFVIIa Experiences in Prophylaxis Patients

Dose (mg/kg)

Brackmann [27]

4

90

Cooper [28]

1

Variable

Saxon [29]

1

Bryant [30]

First author (reference)

Frequency

Response during prophylaxis

Adverse events

Twice daily

Decrease in bleeding slightly lower than FEIBA prophylaxis

No

240 mg/kg rFVIIa every 6–8 h, decreasing to once daily

Two bleeds in the 11 weeks of prophylaxis

No

90

Daily

Prior to prophylaxis: mean 2.1 bleeds per week. During prophylaxis: mean of 0.4 bleeds per week

No

1

300

Three times a week associated with prednisone

No bleeds with prednisone. When the prednisone was stopped, four bleeds occurred in the following 4 weeks. Reintroduction of prednisone with the rFVIIa resulted in no more bleeds

No

Young [31]

2

Variable

Variable

Reduction in bleed frequency and hospitalization

No

Morfini [32]

13

Variable

Variable

In 12/13 cases, prophylaxis considerably reduced the number of bleeding episodes

No

Daily

No episodes of clinical hemarthrosis

No

Daily. Randomized 1:1 to receive prophylaxis with either 90 or 270 mg/kg for 3 months

Reduction in bleeding frequency 45% with the 90 mg/kg dose and 59% with the 270 mg/kg dose

No

Daily

Median bleeds per patient prior to prophylaxis: 4 (mean hemarthrosis 1 (0–1) and, during prophylaxis, 1 bleeding episode (0–5) (mean hemarthrosis 0 (0–4)

No

Jimenez-Yuste [33]

1

90

Konkle [34]

22

90 or 270

Jimenez-Yuste [11]

5

90–100

prednisone (0.1 mg/kg/day) resulted in no spontaneous bleeds over 4 months. When the prednisone was stopped, four bleeds occurred in the following 4 weeks. Reintroduction of prednisone with the rFVIIa resulted in no more bleeds. There was no associated adverse event.

before starting ITI. No episodes of clinical hemarthrosis were observed before or during prophylaxis. The authors concluded that this experience with primary prophylaxis against hemarthrosis with rFVIIa opened a new and interesting field for investigating the efficacy and utility of rFVIIa in patients with hemophilia and inhibitors prior to the onset of joint bleeding.

Young et al [31] described the outcome of long-term (12–25 months) secondary prophylaxis with rFVIIa in two patients with hemophilia and high-titer inhibitors. Overall, prophylaxis with rFVIIa was associated with a reduction in bleed frequency and hospitalization in both patients.

Konkle et al [34] published the only randomized, prospective, clinical trial of rFVIIa. Thirty-eight male patients entered a 3-month preprophylaxis period to confirm high baseline bleeding frequency (mean > four bleeds per month). Twenty-two patients were randomized 1:1 to receive daily rFVIIa prophylaxis with either 90 or 270 mg/kg for 3 months, followed by a 3month post-prophylaxis period.

Morfini et al [32] reported on 13 patients in a retrospective study on rFVIIa prophylaxis carried out in different European centers. The NovoSeven dosage regimen and the duration of treatment varied widely. In 12/13 cases, prophylaxis with rFVIIa considerably reduced the number of bleeding episodes compared with earlier treatment. Eight/nine patients were satisfied or very satisfied with rFVIIa treatment and, in cases reporting subjective quality of life (QoL), all patients reported improved, much improved, or significantly improved QoL. The authors concluded that rFVIIa prophylaxis improved QoL and dramatically reduced the number of bleeding episodes with good patient compliance.

During the prophylaxis period, treatment with 90 mg/ kg rFVIIa significantly reduced bleeds per month from 5.6 to 3.0. Treatment with 270 mg/kg rFVIIa reduced the bleeding frequency from 5.3 to 2.2 bleeds per month. A significant reduction in bleeding frequency was maintained after treatment with both 90 and 270 mg/kg rFVIIa (4.1and 2.7 respectively) during the postprophylaxis period. The effective reduction in bleeding frequency with rFVIIa prophylaxis compared with the preprophylaxis period was 45% with the 90 mg/kg dose (P,0.0001) and 59% with the 270 mg/kg dose (P,0.0001). During the post-prophylaxis period, the reduction in the number of bleeds per month persisted, and the number of

Jimenez-Yuste et al [33] reported the use of rFVIIa prior to any clinically evident joint bleed in a 4-yearold with hemophilia A and a high inhibitor titer. The boy received 90 mg/kg every day over 15 months of prophylaxis awaiting the fall of the inhibitor titer JCD 2009; 1:(1). OCTOBER 2009

24

www.slm-hematology.com


hemophilia could prevent recurrent joint bleeding and preserve an excellent musculoskeletal status after 25 years of follow-up [2, 3]. These results were promising; they also observed that deterioration of a joint was often due to progressive destruction of joints already affected before the start of treatment. This finding indicated that prophylaxis should begin at a very early age, before joints were affected. This proposal regarding so-called primary prophylaxis has since been corroborated in many different reports [1, 37, 38].

bleeds remained significantly less than during the preprophylaxis period for both treatment groups (90 mg/kg rFVIIa, 27%, P,0.01; 270 mg/kg rFVIIa, 50%, P,0.0001). Although a similar reduction was observed with all types of bleed, the effect was most pronounced for spontaneous joint bleeds No thromboembolic events were reported during prophylaxis. The study concluded that clinically relevant reductions in bleeding frequency compared with conventional on-demand therapy were achieved during prophylaxis without raising safety concerns and provided evidence for the concept of secondary rFVIIa prophylaxis in inhibitor patients with frequent bleeds. In a further analysis of the Konkle study data [34], Hoots et al [35] showed that prophylaxis with rFVIIa improves quality of life for inhibitor patients who already have some level of arthropathy.

Hemophilic patients with inhibitors are a vulnerable population who suffer increased morbidity and risk of mortality compared with the non-inhibitor hemophilic population. Bypassing agents have been shown to be effective in controlling bleeding complications in patients who have developed inhibitory antibodies against FVIII or FIX. It should, however, be emphasized that, in contrast to FVIII and FIX substitution, neither of these bypassing agents can completely normalize thrombin generation [39, 40]. Consequently, inhibitor patients have progressive joint disease with significant orthopedic disabilities that are more prevalent than in hemophiliacs without inhibitors [8, 9]. Although different studies have evaluated the orthopedic status of patients with hemophilia, the recent European Study on Orthopaedic Status of Haemophilia Patients with Inhibitors (ESOS) was the first to describe in detail the orthopedic status of inhibitor patients [41]. This study found that, when compared with their non-inhibitor counterparts, patients with inhibitors had higher levels of arthropathy, were hospitalized more frequently due to orthopedic or musculoskeletal complications, and had greater difficulties with mobility and daily activities due to pain/discomfort [41].

Jimenez-Yuste et al [11] retrospectively analyzed the efficacy, cost, and safety of rFVIIa as prophylaxis in inhibitor patients with hemophilia. The median treatment duration in this group was 9 months with a median age of 2 years. The median number of bleeds per patient prior to rFVIIa prophylaxis was four (range 3–10) (mean hemarthrosis 1 (range 0–1)) and, during prophylaxis, one bleeding episode (range 0–5) (mean hemarthrosis 0 (range 0–4)). In all cases, the total number of bleeds was lower during rFVIIa prophylaxis than before. Orthopedic status remained normal in three patients, improved in one case, and worsened in one. None of the patients had life-threatening bleeds during prophylaxis. The median cost of total factor per patient per month prior to prophylaxis was higher than during prophylaxis. In two of the five, the cost of prophylaxis was less than their previous treatment of ITI plus on-demand bypassing agents. In three patients, the cost of prophylaxis was more expensive than their previous FVIII treatment.

Inhibitor eradication with ITI therapy currently remains the best long-term option in patients with hemophilia A, particularly in patients with high antibody titers [42, 43]. Regarding outcome predictors, it seems that a low inhibitor titer (,10 BU) at the beginning of an ITI therapy course is the most consistent predictor of success in most studies [44]. Nevertheless, it can take 12–18 months or longer before antibody titers decline to ,10 BU, the level at which ITI therapy can be initiated.

The treatment can be considered primary prophylaxis as all had one or no joint bleeds before starting prophylaxis. Recently, Mannuci and Palhares de Miranda [36] published an international survey of attitudes towards secondary prophylaxis with recombinant factor VIIa in hemophilia A patients with inhibitors. A total of 31 interviews were performed with physicians and 20 interviews with providers. Overall, rFVIIa secondary prophylaxis was considered a desirable option by 90% of the respondent physicians and 94% of the providers. Almost 80% of the physicians interviewed in the survey envisaged using secondary prophylaxis.

DISCUSSION

The decision to attempt ITI therapy for FIX inhibitors must be very carefully weighed against the relatively high risk of adverse reactions and the relatively low likelihood of success, particularly in those patients whose inhibitor is associated with an allergic phenotype. In this group of patients, allergic reactions could also occur with the use of FEIBA. So, in these patients, if ITI therapy is not indicated, prophylaxis with rFVIIa could be an alternative therapy.

The possibility of primary prophylaxis to prevent joint disease comes from the studies of Nilsson and colleagues in Sweden. They observed that full-dose prophylaxis started in the first few years of life in boys with severe

The use of prophylaxis during ITI has also been recommended [16]. It has been speculated that, by avoiding an inflammatory environment in which the tolerogenic delivery of FVIII is not possible, prophy-

www.slm-hematology.com

25

JCD 2009; 1:(1). OCTOBER 2009


Journal of Coagulation Disorders

bypassing agent prophylaxis is clear. Large, long-term, controlled clinical trials are needed to obtain more data and provide support for the use of prophylaxis with bypassing agents as a standard of care in patients with hemophilia and high-titer inhibitors.

laxis during immune tolerance may prevent major hemorrhage, thereby shortening the duration of ITI and improving the likelihood of success. However, data supporting this possibility are limited. In patients undergoing ITI therapy who experience early joint bleeding or intracranial hemorrhage (ICH), prophylaxis with a bypassing agent may be considered [16]. In an attempt to prevent the sporadic development of catheter-associated vessel thrombosis, it is recommended that prophylaxis with bypassing agents should be discontinued as soon as any indication of FVIII recovery is detected.

Disclosures: V. Jimenez-Yuste has received a fee for speaking by Novo Nordisk A/S and Baxter. The other authors stated that they had no interests which might be perceived as posing a conflict or bias.

REFERENCES 1. Manco-Johnson MJ, Abshire TC, Shapiro AD, et al. Prophylaxis versus episodic treatment to prevent joint disease in boys with severe hemophilia. N Engl J Med. 2007;357(6):535–544. 2. Lofqvist T, Nilsson IM, Berntorp E, Pettersson H. Haemophilia prophylaxis in young patients—a long-term follow-up. J Intern Med. 1997;241(5):395–400. 3. Nilsson IM, Berntorp E, Lofqvist T, Pettersson H. Twenty-five years’ experience of prophylactic treatment in severe haemophilia A and B. J Intern Med. 1992;232(1):25–32. 4. Chambost H, Ljung R. Changing pattern of care of boys with haemophilia in western European centres. Haemophilia. 2005; 11(2):92–99. 5. Wight J, Paisley S. The epidemiology of inhibitors in haemophilia A: a systematic review. Haemophilia. 2003;9(4):418–435. 6. Berntorp E, Shapiro A, Astermark J, et al. Inhibitor treatment in haemophilias A and B: summary statement for the 2006 international consensus conference. Haemophilia. 2006;12(Suppl 6):1–7. 7. Leissinger CA. Prevention of bleeds in hemophilia patients with inhibitors: emerging data and clinical direction. Am J Hematol. 2004;77(2):187–193. 8. Soucie JM, Cianfrini C, Janco RL, et al. Joint range-of-motion limitations among young males with hemophilia: prevalence and risk factors. Blood. 2004;103(7):2467–2473. 9. Gringeri A, Mantovani LG, Scalone L, Mannucci PM. Cost of care and quality of life for patients with hemophilia complicated by inhibitors: the COCIS Study Group. Blood. 2003;102(7):2358– 2363. 10. Leissinger CA. Prophylaxis in haemophilia patients with inhibitors. Haemophilia. 2006;12(Suppl 6):67–73. 11. Jimenez-Yuste V, Alvarez MT, Martin-Salces M, et al. Prophylaxis in 10 patients with severe haemophilia A and inhibitor: different approaches for different clinical situations. Haemophilia. 2009;15(1):203–209. 12. Dimichele D, Negrier C. A retrospective postlicensure survey of FEIBA efficacy and safety. Haemophilia. 2006;12(4):352–362. 13. Ehrlich HJ, Henzl MJ, Gomperts ED. Safety of factor VIII inhibitor bypass activity (FEIBA): 10-year compilation of thrombotic adverse events. Haemophilia. 2002 Mar;8(2):83–90. 14. Aledort LM. Factor VIII inhibitor bypassing activity (FEIBA)— addressing safety issues. Haemophilia. 2007;14:39–43. 15. Ewing NP. Anamnestic responses in hemophilia patients with inhibitors on continuous prophylaxis with factor eight inhibitor bypassing activity, vapor heated (FEIBA VH). J Thromb Haemost. 2005;3(Suppl 1):abstract P2036. 16. DiMichele DM, Hoots WK, Pipe SW, Rivard GE, Santagostino E. International workshop on immune tolerance induction: consensus recommendations. Haemophilia. 2007;13(Suppl 1):1–22. 17. Brackmann HH, Gormsen J. Massive factor-VIII infusion in haemophiliac with factor-VIII inhibitor, high responder. Lancet. 1977;2(8044):933. 18. Kreuz W, Escuriola-Ettinghausen C, Martinez I, Mentzer D, Figura S, Klarmann D. Factor VIII inhibitor bypass activity (FEIBA) for prophylaxis during immune tolerance induction (ITI) in patients with high-responding inhibitors. Blood. 2000; 96(266a):abstract 1140.

Because of the serious and disabling consequences of persistent inhibitors, there is considerable clinical and research interest in establishing effective bypassing agent regimens to prevent bleeding in these patients in much the same way that prophylaxis works in noninhibitor patients [10, 11, 20, 26–29, 31, 36, 45, 46]. Nowadays, prophylaxis is a potential strategy for preventing episodes of joint bleeding and protecting joint damage before, during, and after failed ITI. However, there is a lack of data to support any specific agent or regimen or even to allow a recommendation on their use in different clinical conditions; furthermore, most of the existing reports are still only based on meeting abstracts. Although the effect of FEIBA can last longer than that of NovoSeven, their efficacy rates are similar, suggesting that the biological effect of rFVIIa is actually much longer than indicated by its short halflife. FEIBA contains residual FVIII antigen and may cause an anamnestic response in the inhibitor titer. This is crucial when ITI is postponed to allow the inhibitor titer to decline to ,10 BU. In this setting, FEIBA is not recommended as first-line prophylaxis because of its potential to protract anamnesis, and rFVIIa is the preferred agent [16]. Moreover, some reports suggest the possibility that rFVIIa could protect against joint bleeding during the 12–18 months that may be needed before antibody titers decline to ,10 BU and it is possible to initiate ITT; the use of rFVIIa would allow inhibitor patients to start ITI and protect their satisfactory orthopedic status [33]. However, there are some concerns about efficacy, thrombotic complications, and cost. So the rationale for prophylaxis may be even more difficult for patients with inhibitors. Although there is an obviously important cost associated with prophylaxis, it has been pointed out by authors such as Leissinger et al [21] that, if prophylaxis is initiated early, costs may be similar to those associated with repeated bleeds, hospitalizations, and surgeries in older, larger patients with inhibitors who have joint damage. With few therapeutic options available to patients with hemophilia and inhibitors, the need for more data assessing the safety, efficacy, and cost-effectiveness of JCD 2009; 1:(1). OCTOBER 2009

26

www.slm-hematology.com


38. Berntorp E. Prophylactic therapy for haemophilia: early experience. Haemophilia. 2003;9(Suppl 1):5–9 (discussion). 39. Mehta R, Parameswaran R, Shapiro AD. An overview of the history, clinical practice concerns, comparative studies and strategies to optimize therapy of bypassing agents. Haemophilia. 2006;12(Suppl 6):54–61. 40. Negrier C, Dargaud Y, Bordet JC. Basic aspects of bypassing agents. Haemophilia. 2006;12(Suppl 6):48–53. 41. Morfini M, Haya S, Tagariello G, et al. European study on orthopaedic status of haemophilia patients with inhibitors. Haemophilia. 2007;13(5):606–612. 42. Astermark J, Morado M, Rocino A, et al. Current European practice in immune tolerance induction therapy in patients with haemophilia and inhibitors. Haemophilia. 2006;12(4):363–371. 43. Dimichele DM. Immune tolerance: critical issues of factor dose, purity and treatment complications. Haemophilia. 2006;12(Suppl 6):81–6. 44. DiMichele D. Immune tolerance therapy dose as an outcome predictor. Haemophilia. 2003;9(4):382–386. 45. Konkle B, Friedrich U, Abrams Z. Secondary prophylactic treatment with rFVIIa in patients with haemophilia A or b and inhibitors with high requirements for on-demand treatment. XXVII International Congress of The World Federation of Hemophilia. Haemophilia. 2006;12(abstract 14 PO):363. 46. Hedner U. Potential role of recombinant factor FVIIa in prophylaxis in severe hemophilia patients with inhibitors. J Thromb Haemost. 2006;4(11):2498–2500.

19. Kreuz W, Escuriola-Ettinghausen C, Martinez I, Mentzer D, Figura S, Klarmann D. Efficacy and safety of factor VIII inhibitor bypass activity (FEIBA) for long-term prophylaxis in patients with high-responding inhibitors. Blood. 2000;96(265a): abstract 1140. 20. Hilgartner MW, Makipernaa A, Dimichele DM. Long-term FEIBA prophylaxis does not prevent progression of existing joint disease. Haemophilia. 2003;9(3):261–268. 21. Leissinger CA, Becton DL, Ewing NP, Valentino LA. Prophylactic treatment with activated prothrombin complex concentrate (FEIBA) reduces the frequency of bleeding episodes in paediatric patients with haemophilia A and inhibitors. Haemophilia. 2007;13(3):249–255. 22. Escuriola-Ettinghausen C, Martinez-Saguer I, Funk M, et al. Long-term prophylaxis with FEIBAH in patients with highresponding inhibitors. J Thromb Haemost. 2003;1(Suppl 1): abstract P1628. 23. Ewenstein B, Giangrande P, Morfini M, Tjonnfjord GE, Kraut E, Luu H. Evaluation of FEIBA for prophylaxis in patients with inhibitors. Haemophilia. 2004;10(Suppl 3):114(22 PO 10). 24. Siegmund B, Richter H, Pollmann H. Prophylactic treatment with FEIBA of a haemophilia A patient with inhibitor: what are the costs, what are the benefits? Haemophilia. 2005;11(6):638–641. 25. Schino M, Centra A, Pisedu G, Sbrighi P. APCC (FEIBA) home treatment prophylaxis in inhibitor haemophilia patients. Haemophilia. 2000;6:294 (abstract 37). 26. Valentino LA. The benefits of prophylactic treatment with APCC in patients with haemophilia and high-titre inhibitors: a retrospective case series. Haemophilia. 2009;15(3):733–742. 27. Brackmann HH, Effenberger E, Hess L, Schwaab R, Oldenburg J. NovoSeven in immune tolerance therapy. Blood Coagul Fibrinolysis. 2000;11(Suppl 1):S39–44. 28. Cooper HA, Jones CP, Campion E, Roberts HR, Hedner U. Rationale for the use of high dose rFVIIa in a high-titre inhibitor patient with haemophilia B during major orthopaedic procedures. Haemophilia. 2001;7(5):517–522. 29. Saxon BR, Shanks D, Jory CB, Williams V. Effective prophylaxis with daily recombinant factor VIIa (rFVIIa-Novoseven) in a child with high titre inhibitors and a target joint. Thromb Haemost. 2001;86(4):1126–1127. 30. Bryant P, Carr M, Martin E, Sutton J. High dose recombinant activated factor VII in a pediatric patient with factor VIII deficiency and high titer inhibitor. Blood. 2003;102(11):104b–105b (abstract 4128). 31. Young G, McDaniel M, Nugent DJ. Prophylactic recombinant factor VIIa in haemophilia patients with inhibitors. Haemophilia. 2005;11(3):203–207. 32. Morfini M, Auerswald G, Kobelt RA, et al. Prophylactic treatment of haemophilia patients with inhibitors: clinical experience with recombinant factor VIIa in European Haemophilia Centres. Haemophilia. 2007;13(5):502–507. 33. Jimenez-Yuste V, Quintana M, Alvarez MT, Martin-Salces M, Hernandez Navarro F. Primary prophylaxis with rFVIIa in a patient with severe Haemophilia A and inhibitor. Blood Coagul Fibrinolysis. 2008;19 (6):719–20. 34. Konkle BA, Ebbesen LS, Erhardtsen E, et al. Randomized, prospective clinical trial of recombinant factor VIIa for secondary prophylaxis in hemophilia patients with inhibitors. J Thromb Haemost. 2007;5(9):1904–1913. 35. Hoots K, Ebbesen LS, Konkle B, et al. Secondary prophylaxis with recombinant activated factor VII improves health-related quality of life of haemophilia patients with inhibitors. Haemophilia. 2008;14(3):466–75.Haemophilia. 2008 May;14(3):670. 36. Mannucci PM, Palhares de Miranda PA. International survey of attitudes towards secondary prophylaxis with recombinant factor VIIa in haemophilia A patients with inhibitors. Haemophilia. 2009;15(1):345–347. 37. Berntorp E, Astermark J, Bjorkman S, et al. Consensus perspectives on prophylactic therapy for haemophilia: summary statement. Haemophilia. 2003;9(Suppl 1):1–4.

www.slm-hematology.com

27

JCD 2009; 1:(1). OCTOBER 2009


JOURNAL OF COAGULATION DISORDERS

REVIEW ARTICLE

Safety of Radiosynovectomy in Hemophilic Synovitis: it is Time to Re-evaluate! Cuneyt Turkmen Affiliation: Department of Nuclear Medicine, Istanbul Faculty of Medicine, Istanbul University, Istanbul, Turkey Submission date: 2nd July 2009, Revision date: 11th August 2009, Acceptance date: 18th August 2009

A B S T R A C T The hemophiliac joint has become an increasing target for radiosynovectomy since the first description by Ahlberg in 1977. Radiosynovectomy with the beta-emitting nuclide has been effective in dramatically reducing the frequency of hemarthroses and resolving chronic synovitis in hemophilic patients. Because the joint-surface erosions secondary to chronic synovitis often occur in early childhood and progress to advanced arthropathy by late adolescence, these patients need therapy in the early stages of life. However, several issues, including radiopharmaceutical choice and the potential side-effects, have been questioned, particularly in younger patients concerning radiation safety. Although the clinicians have been a bit confused about the safety of radiosynovectomy in the pediatric population, recent advances have prompted clinicians to take a renewed interest in the potential side-effects of radiosynovectomy. There is a growing realization that the benefits of properly performed radiosynovectomy should always outweigh any short- or long-term side-effects. Keywords: radiosynovectomy, radiosynoviorthesis, safety, hemophilia, synovitis, yttrium, rhenium Correspondence: Cuneyt Turkmen, Istanbul University, Istanbul Faculty of Medicine, Department of Nuclear Medicine, C ¸ apa, 34390 Istanbul, Turkey. Tel: +90-212-4142000; fax: +90-212-4142056; e-mail: cturkmen@istanbul.edu.tr

available, started a discussion about the risk/benefit of this procedure [4].

INTRODUCTION Hemophilic arthropathy following repeated hemarthrosis is the primary morbidity of hemophilia [1]. If we try to see the world from a hemophilic child’s point of view, joint bleeding, which causes lifelong disability and interrupts the physical and social activities of a child, is a ‘‘nightmare’’. It was in order to stop the vicious circle of target joints that radiosynovectomy, also known as radiosynoviorthesis, was proposed by Ahlberg 30 years ago [2]. The procedure consists of the intra-articular installation of radiopharmaceuticals within an affected joint to selectively destroy the pathologic synovium. Radiosynovectomy has gained more and more acceptance by providing a simple, economic, and quite effective treatment modality in hemophilic synovitis over the last decade.

As radiosynovectomy in children or young adults should only be performed if the estimated benefit outweighs the potential risks, a good understanding of the possible side-effects of therapy is mandatory for correct risk estimation and justification of the therapy. For a reliable risk–benefit analysis, we need correctly analyzed available data about the safety of radiosynovectomy relating to different radiosynovectomy agents. This review focuses on the safety of radiosynovectomy in hemophilic synovitis in the light of recent advances.

RADIOPHARMACEUTICALS FOR RADIOSYNOVECTOMY

In fact, widespread use of radiosynovectomy has paralleled the availability of new-generation radiosynovectomy agents, which exhibit minimal leakage of activity from the treated joint, as well as studies that have led to high efficacy of the therapy [3].

In the search for the ideal compound, a large number of different radionuclides in different chemical forms have been used since radiosynovectomy was first introduced in 1952 using colloidal gold (Au-198) [5]. The desired requirements in an ideal radiopharmaceutical for radiosynovectomy are: (1) having beta emission at an appropriate energy level for effective ablation of inflamed synovium, but not so great as to damage the underlying articular cartilage or overlying

Recently, two leukemia cases, reported in hemophilic patients after treatment with 32P radiosynovectomy in the United States, where yttrium-90 (Y-90) and rhenium-186 (Re-186) radiocolloids are not commercially JCD 2009; 1:(1). OCTOBER 2009

29

www.slm-hematology.com


Journal of Coagulation Disorders

skin; (2) incorporation in a colloidal preparation with a particle size small enough to permit phagocytosis, but large enough to avoid extra-articular leakage; and 3) having a physical half-life short enough to minimize prolonged radiation in vivo, but long enough for delivery from the site of manufacture to the hospital and effective therapy [6].

of multiple joints with different radiopharmaceuticals can be performed in one session. The recommended limit for the total administered activity per session is 400 MBq. If indicated, up to three radiosynovectomy treatments are recommended with a 3- to 6-month interval between them [9].

Today, related to joint size, three radionuclides have gained widespread acceptance for radiosynovectomy in Europe: Y-90 citrate or silicate colloid for knee joints; rhenium-186 (Re-186) sulfide colloid for midsized joints; and erbium-169 (Er-169) colloid for all finger and toe joints. Phosphorus-32 (P-32) is mainly used for radiation synovectomy in the United States, and dysprosium-165 (Dy-165) ferric hydroxide is also in use. P-32 and Dy-165, with beta energy ranging between the energy of Y-90 and Re-186, are usually applied to knee joints and mid-sized joints. While using P-32 and Dy-165 in a wide range of joint size seems to be an advantage, they are no longer mentioned in European guidelines because these radiopharmaceuticals have disadvantages such as half-life and high lymphatic transport [7]. Investigation of other potentially useful radionuclides, such as holmium-166 hydroxyapatite, Re-188 microspheres, and Sm-153 particulate hydroxyapatite, is still relatively new. A summary of commercially available and routinely applied radionuclides for radiosynovectomy is given in Table 1.

SIDE-EFFECTS

DOSE CONSIDERATIONS

More severe complications are local skin and needle track ulcerations caused by radionecrosis, which may occur if the radionuclides flush back out of the needle during retraction or throughout the puncture channel from the joint after injection. This can easily be avoided by flushing the needle with steroids or 0.9% saline and compression of the injection site after application of the radionuclide. Necrosis of periarticular tissue is the worst local complication in radiosynovectomy and is caused by accidental paraarticular injection of the radionuclide. Only a few cases related to needle track ulceration or periarticular necrosis have been reported in the literature [10, 11].

Local complications Local complications after radiosynovectomy are very rare. A temporary increase in joint pain and swelling due to radiation-induced synovitis 6–48 h after treatment may be observed. Lymphedema or fever may occur in rare cases. These symptoms are usually selflimited without further intervention and can be treated simply by cooling the joint with ice packs or, if necessary, with anti-inflammatory drugs. Ultrasound or magnetic resonance imaging (MRI) of the joint is useful to evaluate joint space, the structure of the synovial, and the amount of effusion to ensure homogeneous distribution of the radiopharmaceuticals. In cases of loculated effusions, radionuclide administration is not recommended because of the high local radiation burden with danger of necrosis. Fluoroscopic or ultrasound guidance by injection air or contrast media into the joint is also important to verify successful intra-articular injection especially in mid-sized to small joints.

It is difficult to determine exactly the amount of activity of any radionuclide needed to achieve a therapeutic response. The absorbed dose is not only dependent on the type of radionuclide and the amount of activity used, but also on various other factors such as the size of the joint cavity, synovial thickness, distribution of the colloids in the joint fluids, and the inflammatory activity of the joint. Approximately 100 Gy per 100 g of synovial tissue should be absorbed to have an optimal effect [8]. Recommended activities and radiopharmaceuticals for the various joints are summarized in Table 1. Activity levels may be adjusted to 50–100% of the recommended activity in adults for the respective joint depending on the age and size of the child. Treatment

To prevent from radionecrosis, avoiding the use of an inappropriate radionuclide as well as intra-articular

Table 1. Properties of Radiopharmaceuticals used for Radiosynovectomy

Radiopharmaceutical

Emission Particle (b: beta; size (nm) c: gamma)

Energy (MeV)

Half-life (days)

Penetration mean/max (mm)

Main use/adult dose (MBq)

Y-90 silicate or citrate colloid

100

b

2.2

2.7

3.6/11

Knee joint; 185

Re-186 sulfide colloid

5–10

b, c

1.0

3.72

1.2/3.6

Hip, shoulder; 74–185 Elbow, wrist, ankle; 74

10

b

0.34

9.4

500–2000 3000–8000

b b, c

1.7 1.29

14 0.1

Er-169 citrate colloid P-32 colloid Dysprosium-165 Ferric hydroxide macroaggregate

JCD 2009; 1:(1). OCTOBER 2009

30

0.3/1

Small joints of hands and feet; 37

2.6/7.9 1.3/5.7

Knee, elbows and ankles; 10–40 Knee, hip; 10 000–11 000 Ankle 7400– 9200

www.slm-hematology.com


injection with sufficient technique is mandatory. An unacceptably high rate of serious complications including severe necrosis was reported after the injection of Y-90 into an ankle joint, which should be treated with a radionuclide of lower energy and a lower tissue range such as Re-186 [12]. In cases of radionecrosis or para-articular injection, no reliable recommendations exist owing to limited experience. One alternative would be to wait and watch carefully until demarked necrotic tissue can be resected. Another alternative would be to perform surgery immediately to remove as much of the para-injected activity as possible by flushing the tissue and by resecting radioactive tissue around the injection site [13].

cause both focal clustering of chondrocytes and fibrillation of the collagenous matrix of the articular cartilage [19]. Similar results with focal damage of chondrocytes and surrounding extracellular matrix were observed in eight patients with rheumatoid arthritis after treatment of the knee joint with 185 MBq of Y-90 [21]. Contradictory results with no microscopic signs of cartilage degeneration were found after the application of 150 MBq of Y-90 into the knee joints of dogs [20]. Although not as large as that to cartilage, the absorbed dose to bone surface and red bone marrow is also of interest in radiosynovectomy. The bone surface dose was described as being 25% of the synovial surface dose in the case of Y-90, down to 4% in the case of Re-186, and negligible for Er-169 [22]. The dose to the bone surface further decreases through the thickening of the synovial membrane in hemophilic synovitis. For the bone surface, a maximum dose of 18 Gy was calculated for 185 MBq of Y-90, which is not considered to be a dose sufficient to cause significant bone damage or necrosis. When there are already bony changes manifested on X-rays, it is better to avoid the procedure considering both its efficacy and safety. The dose to the bone marrow in large or mid-sized joints is considered negligible because the distance to the radiation source is greater than the mean tissue penetration of radionuclides used for radiosynovectomy.

A distribution scan acquired with a gamma camera after radionuclide injection is also helpful to verify successful intra-articular injection and proper distribution within the joint. KavaklÄą et al [14, 15] reported their experience of radiosynovectomy with Re-186 and Y-90 in two large series. As they observed no serious adverse events, they conclude that Y-90 for the knee joint and Re-186 for mid-sized joints are safe agents.

Radiation effects to articular cartilage and bone There is a potential risk of irradiation of healthy intra- and extra-articular tissues such as cartilage and bone from the radioactivity within the injected joint. Beta-emitting colloidal particles are phagocytosed by inflamed hypertrophic synovial tissue, including that part of the synovial lining that lies adjacent to the hyaline cartilage at its margins. Some irradiation of cartilage and subchondral bone during radiosynovectomy is thus inevitable. Although mature cartilage has been considered to be resistant to radioactivity, minor injury to articular cartilage remains a concern for radiosynovectomy especially in the pediatric population. Some studies using P-32 or Y-90 for radiosynovectomy reported injuries of both articular cartilage and the growth plate [16, 17]. Additionally, synovial damage, subsequent joint inflammation, and fibrosis may also contribute to articular cartilage damage after radiosynovectomy. Extensive fibrosis of the subsynovial tissues has been reported after radiosynovectomy with Y-90, Dy-165 ferric hydroxide macroaggregate, or P-32 [18–20]. Fibrosis may hamper filtration and resorption of synovial fluid, and lead indirectly to articular cartilage damage [13].

Whole-body radiation exposure Whole-body radiation-absorbed doses were reported ranging from 9 to 99 mSv after the application of 200 MBq of Y-90 into the knee joint with a median of 37 mSv [23]. In the same study, the total gonadal dose was 0.1 mSv in women and 0.2 mSv in men. van der Zant et al [24] calculated the effective dose for 75 MBq of Re-186 as 0.15 mSv using tissue weight factors defined by the International Commission on Radiological Protection. In another dosimetric study, Manil et al [25] calculated the effective dose for 70 MBq of Re-186 as 26 mGy. It might be possible that this calculation greatly overestimated the dose to the whole body, because the authors nevertheless concluded that their results were within certain methodological limitations. The primary disadvantage for radiation exposure in radiosynovectomy is the leakage of injected radioactivity away from the joint. The absorbed dose of regional lymph nodes may occasionally be quite large. After radiation synovectomy with Re-186, the activity in lymph nodes was reported to be up to 6% of the injected activity [25]. van der Zant et al [24] showed that, 24 h after radiosynovectomy with Re-186 in the ankle joints, the maximal leakage of the radioisotope to a single lymph node and the liver was 4% and 5.5% respectively. We have found that mean proportions of

According to experimental data and morphological in vitro studies, a radiogenic decrease in collagen synthesis must be taken into account if potential sideeffects of radiation synovectomy are discussed. However, because similar effects can be observed in hemophilic arthritis as well, the clinical impact of these findings is not clear. Intra-articular injection of 15 MBq of Y-90 in rabbit knee joints was reported to www.slm-hematology.com

31

JCD 2009; 1:(1). OCTOBER 2009


Journal of Coagulation Disorders

first days after treatment. One single active use of the joint was reported to result in leakage of up to 40% of the injected activity [35]. Therefore, strict immobilization for at least 48 h is regarded as mandatory to reduce extra-articular drainage to a minimum. As well as joint immobilization, the choice of radionuclide with a short half-life is also an attractive approach to minimize the radiation dose due to leakage.

the injected Re-186 activity were 0.2¡0.7% in the lymph nodes and 4.7¡2.8% in the hepatosplenic area using quantitative gamma camera imaging 48 h after radiosynovectomy [26]. Grmek et al [27] reported that patients received a 0.8– 3.7 mSv radiation dose owing to the leakage of 186-Re from the treated joint and its retention in the body. The highest doses were established in the spleen (26.0¡10.7 mGy), the liver (17.6¡7.2 mGy), and red marrow (3.0¡0.8 mGy). The contribution of gamma radiation to the effective dose was less than 0.1 mSv in radiosynovectomy of the ankle, 0.4 mSv in the elbow, and 0.6 mSv in the shoulder joint [27]. Reported extraarticular leakage rates from acceptable and effective doses received by patients after radiosynovectomy using Re-186 colloid are quite low.

Genotoxic effect and cancer risk Although the biology of radiation carcinogenesis is not yet understood in detail, some general features have become widely accepted. There are reports of radiation-induced membrane damage, but the biological effects of ionizing radiation result largely from DNA damage, caused directly by ionizations within the DNA molecule or indirectly from the action of chemical radicals formed as a result of local ionizations. Ordinarily, a high proportion of radiationinduced DNA damage is repaired by the cell, with long-term biological consequences resulting from a proportion that is unrepairable or misrepaired.

With regard to the leakage rates of pure beta emitter agents, available data are limited, and the results of studies with different methodology are sometimes contradictory. Y-90 citrate silicate, with a particle size of 10–100 nm, was reported to leak 5–10% at 24 h following administration and 15–25% at 5 days [28]. In contrast to this study, Spooren et al [29] could not observe any extra-articular activity resulting from the leakage of Y-90 silicate. In agreement with our experience, Gedik et al did not observe any visible extra-articular leakage in whole-body Bremsstrahlung imaging obtained after radiosynovectomy with Y-90 citrate [30, 31]. Gedik et al also reported that there were no significant differences in terms of extraarticular leakage between Y-90 citrate or silicate and R-186 sulfide [30].

Exposure to ionizing radiation may cause both deterministic and stochastic biologic effects. Deterministic effects are those that typically occur soon after exposure and that increase in magnitude with increasing doses above a threshold dose level. Deterministic doses such as the intended dose on the synovial surface results in cell death. Stochastic effects of radiation typically occur later after exposure, and the probability but not the magnitude of the effects is dose dependent. A threshold dose level for stochastic effects is generally not assumed. Examples of stochastic effects include cancer induction and genetic changes.

Leakage data on Er-169 colloid report a maximum rate of 14% of the injected activity [32].

Present knowledge of the late effects of exposure to ionizing radiation is limited, as the dose–response assessments rely heavily on high-dose exposure studies and animal experiments. Epidemiologic studies have demonstrated that children are considerably more sensitive to the carcinogenic effects of ionizing radiation than adults. Furthermore, children live longer and thus have a larger window of opportunity for expressing radiation damage. Based on epidemiologic studies of Japanese atomic bomb survivors and children and infants irradiated for benign diseases such as tinea capitis and skin hemangioma, a distinct pattern of risk for radiation-related tumors has emerged. Dose-related increased risks for cancers of the thyroid gland, breasts, brain, non-melanoma skin cancer, and leukemia have been observed in adults who were irradiated for benign diseases in childhood [36].

For Dy-165 and P-32 colloidal particles, low leakage rates are also reported [33, 34]. In 125 joints treated with P-32, leakage rates of less than 2% of the injected activity were found by Siegel et al in locoregional lymph nodes [34]. In fact, monitoring the distribution of the injected pure beta emitter radionuclide (e.g., P-32 or Y-90) is difficult because the particles are poorly detected externally using a counter (Geiger-Muller), and less than 1% of the beta particles are converted to Bremsstrahlung, so that quantitative analysis of the low-count Bremsstrahlung image is subject to large statistical fluctuations in gamma camera imaging. Moreover, because of these discordant results about leakage rates after radiosynovectomy even with the same radionuclide, it can be speculated that, considering the safety of radiosynovectomy, the optimization of this therapy technique is as important as the physical properties of the selected radiopharmaceutical. The degree of leakage is influenced by the particle size and radiochemical stability of the radiopharmaceutical and also by the degree of joint movement during the JCD 2009; 1:(1). OCTOBER 2009

The International Commission on Radiation Protection recommendations (2005) stated that ‘‘the weight of evidence on fundamental cellular processes supports the view that in the low dose range up to a few tens of mSv, it is scientifically reasonable to assume that in 32

www.slm-hematology.com


general and for practical purposes cancer risk will rise in direct proportion to absorbed dose in organs and tissues’’ [37]. There should be a differentiation of tumor entities and their likelihood of being theoretically induced by radiosynovectomy agents. For instance, there is no likelihood of a thyroid cancer being induced by nuclides other than radioiodine. In this context, leakage of b emitters causes little stochastic exposure especially in the hematopoietic system, thus resulting in a low probability of causation of leukemia. No published studies have directly attributed cancer to radiosynovectomy, and it is important to recognize how difficult it would be to perform such a study. The lifetime risk of fatal cancer in the general population is approximately 1 in 5. To perform a study to detect an increase from 0.2000 (the 1 in 5 risk in the general population) to 0.2002 (the 1 in 5 risk seen in the general population plus a 1 in 5000 potential risk from a radiosynovectomy) would require hundreds of thousands to millions of subjects and extremely careful matching of the subjects in the study to ensure an accurate result.

such as I-131 in benign thyroid diseases [41]. Also, effective doses in radiological imaging range easily in the same magnitude of ,20 mSv, when computed tomography (CT) is used repeatedly. It is almost twice the dose of an abdominal CT image [42]. If we adapted this rough estimation for radiosynovectomy patients, it is obvious that the average doses to the patients after radiosynovectomy were not sufficiently high to make the statistical detection of any increase in childhood leukemia rates likely. There have been two cases of acute lymphocytic leukemia reported in hemophilia patients receiving chromic phosphate-32. Both children, aged 9 and 14 years, had uncomplicated radiosynovectomy and developed leukemia within 1 year [4]. Interestingly both patients have a history of autoimmune disorder, and the interval between exposure and the development of leukemia is less than the expected peak of radiation-induced leukemia. A recent survey of hemophilia treatment centers in the USA identified 1017– 1027 P-32 injections into 563–577 patients who have had radiosynovectomy since 1988. There are no other reports of malignancy associated with radiosynovectomy in this country. As a consequence of these arguments, the Medical and Scientific Advisory Council of the National Hemophilia Foundation recommends discussion about the risk–benefit ratio of radiosynovectomy, including the potential risk of cancer, with all individuals or with their parents considering the procedure, and written informed consent should be obtained which clearly documents that these two cases of malignancy were discussed [43].

The Chernobyl accident might shed light on the late effects of protracted low-dose radiation exposure. A comprehensive dose reconstruction program estimated radiation dose to the thyroid from the inhalation or ingestion of I-131, other short-lived radioactive iodine isotopes (I-132, I-133, and I-135), radioactive tellurium isotopes (Te-131m and Te-132), and radioactive cesium isotopes (Cs-134 and Cs-137), and external gamma-ray exposures from radioactive elements deposited on the ground. The accident initiated a major and early increase in childhood thyroid cancer that resulted from ingestion of iodine-131 by young children living in the most heavily contaminated areas of Belarus, Ukraine, and Russia [38]. On the other hand, no evidence of an increasing number of childhood leukemia or other malignancy cases was noted over time. In fact, there is no contradiction with radiobiological experience if current risk estimates and the magnitude of the doses after Chernobyl are considered. The average added radiation exposures even in the more contaminated areas are unlikely to be more than about 10 mSv since the accident. For children, the average doses were probably slightly larger than the average doses to the total population. However, with the familiar estimates of the excess risk of leukemia in childhood (ages up to 15 years) of about 0.004/Sv and with 10 mSv, one would expect an increase in the probability of a child developing leukemia before the age of 15 by 4/100 000 compared with the spontaneous probability of about 60/100 000 during childhood [39, 40].

Until now, an increased risk of cancer after radiosynovectomy with Y-90 and Re-186 radiocolloids has not been reported. A number of studies have been published, in which the kinetics of induction of chromosomal aberrations in peripheral blood lymphocytes from patients who were undergoing radiosynovectomy was analyzed [44–48]. Most of these studies, performed in the 1970s, revealed an increased incidence of radiation-induced chromosomal aberrations in patients treated with various colloidal forms (e.g., resin colloid, ferric hydrate colloid, citrate colloid) of Y-90. In fact, the radiopharmaceuticals used in the majority of these studies were not optimal when compared with radiopharmaceuticals in use today. As there are apparent differences between the form used and the leakage rates of radiosynovectomy agents, the results of these studies are incomparable, like comparing apples and oranges. Biological dosimetry using the measurement of the yield of dicentric chromosomes in human lymphocytes is an established technique. It works well for acute radiation exposures when the blood sample is taken within a few weeks after exposure. This technique is capable of estimating whole-body doses of low-linear energy transfer (LET) radiation down to about

The effective doses for radiosynovectomy given in the dosimetric studies seem to be in the low-dose range. For example, the effective dose with Re-186 remains approximately 30 times lower than with other treatments www.slm-hematology.com

33

JCD 2009; 1:(1). OCTOBER 2009


Journal of Coagulation Disorders

100 mGy. Recently, we studied the cytogenetic analyses such as chromosomal aberration analysis, micronuclei, and sister chromatid exchange as an indicator of radiation induced-cytogenetic damage in 38 hemophilic children undergoing radiosynovectomy using Y90 or Re-186 [26, 31]. The results of our studies indicate that high radiation doses, which would induce genotoxic effects, are not obtained by peripheral blood lymphocytes in children after radiosynovectomy. Dicentric aberrations are the main interest in these types of studies as the formation of dicentric chromosomes in human peripheral lymphocytes is a specific effect of ionizing radiation [49]. We could not detect any persistent dicentric chromosomal aberrations after the therapy, and there was no statistically significant increase in the number of chromosomal aberrations in children who were treated with Y-90 or Re-186 radiosynovectomy. In agreement with our findings, Falcon de Vargas and Fernandez-Palazzi and Voth et al reported that there was no significant increase in the number of dicentric chromosomes following therapy in patients who were treated with Y-90 radiosynovectomy [50–52].

Furthermore, the long-term risks of cancer in patients with rheumatoid arthritis who have been treated with Y-90 were discussed by Vuorela et al [54]. Only nine cases of cancer were found among the patients who had received Y-90, whereas the expected number, based on the incidence among the population in the region, was 14.9. They concluded that radiosynovectomy with Y-90 did not appreciably increase the risk of cancer [54].

In a study carried out on 31 hemophilic patients (age range 9–24 years) with no chromosomal aberrations, only non-specific chromosomal structural changes (breakages) were observed 6 months after Re-186 injection for radiosynovectomy, and these changes were reversible until 1 year after the injection [50]. In contrast to these findings, Manil et al reported a significant cumulative increase in dicentric aberrations 7 days after radiosynovectomy in 45 rheumatoid arthritis patients treated with Re-186 [25]. However, as there was no follow-up after 7 days in this study, it is unclear whether this significant increase in dicentrics would be persistent afterwards.

The response rate of different radiopharmaceuticals currently in use for radiosynovectomy appears not to differ significantly. But the side-effects, which are mainly related to the penetration and leakage rates of radiopharmaceuticals, depend on correct radiopharmaceutical choice for the joint involved and correct technique. Otherwise, irradiation of normal intraarticular as well as extra-articular tissues and its consequences would be more pronounced. There are only two reported cases of malignancy among people with hemophilia or rheumatoid arthritis who have received P-32 for radiosynovectomy in the United States. The Medical and Scientific Advisory Council of the National Hemophilia Foundation emphasize that, because of the low frequency and the short interval between exposure and the development of leukemia in these two cases, the casual relationship between radiation exposure and malignancy cannot be established conclusively, but possible causality cannot be ruled out. Today, Y-90, Re-186, and Er-169 have gained widespread acceptance for radiosynovectomy in Europe, and P-32 is no longer mentioned in European guidelines.

The long-lasting clinical practice and the lack of any well-documented cases of malignancy resulting from radiosynovectomy suggest a very low and acceptable risk compared with the benefit for the patient. The tumor morbidity rate as a result of whole-body irradiation was calculated as 0.4 per 1000 related to International Commission on Radiological Protection (ICRP) 60 risk data [55], and the genetic radiation risk related to United Nations Scientific Committee on the Effects of Atomic Radiation (UNSCEAR) data was described as being several orders of magnitude below one per 1000 [13].

SUMMARY

On the other hand, we observed an increased frequency of micronuclei in the peripheral lymphocytes of children 2 days after Y-90 therapy. But this effect was not persistent in the peripheral lymphocytes of the children in this study and had disappeared at the day 90 control. In agreement with our experience, Prosser et al analyzed 22 patients treated with Y-90 silicate, six of whom showed significantly elevated yields of micronuclei in their peripheral lymphocytes 2 weeks after therapy [53]. Micronuclei represent small, additional nuclei formed by the exclusion of chromosome fragments or whole chromosomes lagging at mitosis. Micronuclei rates, therefore, indirectly reflect chromosome breakage or impairment of the mitotic apparatus. Even if the relationship of cancer risk with micronuclei is not well substantiated, as is that with chromosomal aberrations, this method has been proven to be useful as a ‘‘biologic marker of early effects’’ in biomonitoring studies on the human population that is exposed to genotoxic agents. JCD 2009; 1:(1). OCTOBER 2009

In view of the literature, it can be said that, while radiation dose seems to be higher in Y-90 radiosynovectomy compared with Re-186, radiation doses are still below unsafe levels in children who undergo radiosynovectomy. Chromosomal aberrations and increased micronucleus frequency do occur in some patients as a result of radiosynovectomy, but these aberrations are neither specific nor persistent. Review of the literature on the safety of radiosynovectomy, together with clinical experience over 30 years, brings new evidence to bear and provides a potent argument 34

www.slm-hematology.com


14. Kavakli K, Aydog˘du S, Omay SB, et al. Long-term evaluation of radioisotope synovectomy with yttrium 90 for chronic synovitis in Turkish haemophiliacs: Izmir experience. Haemophilia. 2006; 12(1):28–35. 15. Kavakli K, Aydogdu S, Taner M, et al. Radioisotope synovectomy with rhenium186 in haemophilic synovitis for elbows, ankles and shoulders. Haemophilia. 2008;14(3):518–523. 16. Pavelka K, Meier-Ruge W, Muller W, Fridrich R. Histological study of effects of colloidal 90 yttrium on knee joint tissues of rabbits. Ann Rheum Dis. 1975;34:64–69. 17. Ubios AM, Cabrini RL, Silberman FS, Miranda FF. Radiation effects produced by the intra-articular injection of 32P. Clin Orthop. 1978;136:299–303. 18. Meier-Ruge W, Muller W, Pavelka K. Effect of yttrium 90 on experimental allergic arthritis in rabbits. Ann Rheum Dis. 1976; 35:60–66. 19. Zuckerman JD, Sledge CB, Shortkroff S, Venkatesan P. Treatment of antigen-induced arthritis in rabbits with dysprosium-165ferric hydroxide macroaggregates. J Orthop Res. 1989;7:50–60. 20. Howson MP, Shepard NL, Mitchell NS. Colloidal chromic phosphate 32P synovectomy in antigen-induced arthritis in the rabbit. Clin Orthop. 1988;229:283–293. 21. Kerschbaumer F, Bauer R, Falser N, Altmann H. Effects and side effects of radiosynovectomy with yttrium 90 on rheumatic joint cartilage. Arch Orthop Trauma Surg. 1979;93:95–102. 22. Johnson LS, Yanch JC, Shortkroff S, et al. Beta-particle dosimetry in radiation synovectomy. Eur J Nucl Med. 1995;22: 977–988. 23. Klett R, Puille M, Matter HP, et al. Activity leakage and radiation exposure in radiation synovectomy of the knee: influence of different therapeutic modalities. Z Rheumatol. 1999;58:207–212. 24. van der Zant FM, Jahangier ZN, Moolenburgh JD, et al. Radiation synovectomy of the ankle with 75 MBq colloidal 186 rhenium sulfide: effect, leakage, and radiation considerations. J Rheumatol. 2004;31:896–901. 25. Manil L, Voisin P, Aubert B, et al. Physical and biological dosimetry in patients undergoing radiosynoviorthesis with erbium-169 and rhenium-186. Nucl Med Commun. 2001;22:405– 416. 26. Turkmen C, Ozturk S, Unal SN, et al. Monitoring the genotoxic effects of radiosynovectomy with Re-186 in paediatric age group undergoing therapy for haemophilic synovitis. Haemophilia. 2007;13(1):57–64. 27. Grmek M, Milcinski M, Fettich J, Brecelj J. Radiation exposure of hemophiliacs after radiosynoviorthesis with 186Re colloid. Cancer Biother Radiopharm. 2007;22(3):417–422. 28. Davis MA, Chinol M. Radiopharmaceuticals for radiation synovectomy: evaluation of two yttrium-90 particulate agents. J Nucl Med. 1989;30:1047–1055. 29. Spooren PFMJ, Raske JJ, Arens RPJH. Synovectomy of the knee with 90Y. Eur J Nucl Med. 1985;10:441–445. 30. Gedik GK, Ugur O, Atilla B, et al. Comparison of extrarticular leakage values of radiopharmaceuticals used for radionuclide synovectomy. Ann Nucl Med. 2006;20:183–88. 31. Turkmen C, Ozturk S, Unal SN, et al. The genotoxic effects in lymphocyte cultures of children treated with radiosynovectomy by using yttrium-90 citrate colloid. Cancer Biother Radiopharm. 2007;22(3):393–399. 32. Ruotsi A, Hypen M, Rekonen A, Oka M. Erbium-169 versus triamcinolone hexacetonide in the treatment of rheumatoid finger joints. Ann Rheum Dis. 1979;38:45–47. 33. Pirich C, Pilger A, Schwameis E, et al. Radiation synovectomy using 165Dy ferric-hydroxide and oxidative DNA damage in patients with different types of arthritis. J Nucl Med. 2000;41: 250–256. 34. Siegel HJ, Luck JV Jr, Siegel ME, Quinones C. Phosphate-32 colloid radiosynovectomy in hemophilia: outcome of 125 procedures. Clin Orthop Relat Res. 2001;392:409–417.

against the risk of genotoxicity arising from Y-90 and Re-186 radiosynovectomy. On the other hand, because of the complexity of biomonitoring of genotoxicity, we need further investigations to understand the biological effects of the radiation exactly. Additionally, for correct justification and safe application of therapy, close interdisciplinary collaboration with hematologist, orthopedist, and nuclear medicine physician is strongly recommended. It should also be kept in mind that radiopharmaceuticals should be used only under the guidance of nuclear medicine physicians who are qualified through specific training in the safe use and handling of radionuclides and whose experience and training have been approved by the appropriate government agency authorized to license the use of radionuclides. Patients should receive written and verbal information about the procedure, and written informed consent should be obtained prior to treatment. Informed consent should include the rationale for treatment, treatment alternatives, potential side-effects, and outcome. Disclosures: The author has no financial interests to disclose related to the contents of this article.

REFERENCES 1. Rodriguez-Merchan EC. Pathogenesis, early diagnosis, and prophylaxis for chronic hemophilic synovitis. Clin Orthop. 1997;343:6–11. 2. Ahlberg A. Synoviorthesis with radioactive gold in hemophilia. Rev Rhum Mal Osteoartic. 1977;44(1):41–44. 3. Deutsch E, Brodack JW, Deutsch KF. Radiation synovectomy revisited. Eur J Nucl Med. 1993;20:1113–1127. 4. Dunn AL, Manco-Johnson M, Busch MT, Balark KL, Abshire TC. Leukemia and P32 radionuclide synovectomy for hemophilic arthropathy. J Thromb Haemost. 2005;3:1541–1542. 5. Fellinger K, Schmid J. Die Lokale Behandlung der rheumatischen Erkrankungen (Local therapy of rheumatic diseases). Wien Z Inn Med. 1952;33:351–363. 6. Ug˘ur O, Gedik GK, Atilla B, Rubello D. Radiosynovectomy: current status in the management of arthritic conditions. Nucl Med Commun. 2008;29(9):755–758. 7. Clunie G, Fisher M. EANM procedure guidelines for radiosynovectomy. Eur J Nucl Med. 2003;30:BP12–16. 8. Schneider P, Farahati J, Reiners C. Radiosynovectomy in rheumatology, orthopedics, and hemophilia. J Nucl Med. 2005; 46:48S–54S. 9. Dunn AL, Busch MT, Wyly JB, Abshire TC. Radionuclide synovectomy for hemophilic arthropathy: a comprehensive review of safety and efficacy and recommendation for a standardized treatment protocol. Thromb Haemost. 2002;87(3):383–393. 10. Savaser AN, Hoffmann KT, Sorensen H, Banzer DH. Radiosynoviorthesis in the treatment plan of chronic inflammatory joint diseases. Z Rheumatol. 1999;58:71–78. 11. Peters W, Lee P. Radiation necrosis overlying the ankle joint after injection with yttrium-90. Ann Plast Surg. 1994;32:542– 543. 12. Bickels J, Isaakov J, Kollender Y, Meller I. Unacceptable complications following intra-articular injection of yttrium 90 in the ankle joint for diffuse pigmented villonodular synovitis. J Bone Joint Surg Am. 2008;90(2):326–328. 13. Brenner W, Radionuclide joint therapy. In: Eary JF, Brenner W, eds, Nuclear Medicine Therapy. New York: Informa Healthcare; 2007:21–44.

www.slm-hematology.com

35

JCD 2009; 1:(1). OCTOBER 2009


Journal of Coagulation Disorders

35. Gratz S, Gobel D, Behr TM, Herrmann A, Becker W. Correlation between radiation dose, synovial thickness, and efficacy of radiosynoviorthesis. J Rheumatol. 1999;26:1242–1249. 36. Kleinerman R. Cancer risks following diagnostic and therapeutic radiation exposure in children. Pediatr Radiol. 2006;36(Suppl 14): 121–125. 37. International Commission on Radiological Protection. 2005 Recommendations of the International Commission on Radiological Protection. ICRP; 2005:30. 38. Cardis E, Kesminiene A, Ivanov V, et al. Risk of thyroid cancer after exposure to 131I in childhood. J Natl Cancer Inst. 2005;97: 724–732. 39. Preston DL, Kusumi S, Tomonaga M, et al. Cancer incidence in atomic bomb survivors. Part III. Leukemia, lymphoma and multiple myeloma, 1950–1987. Radiat Res. 1994;137(Suppl):68–97. 40. Gapanovich VN, Iaroshevich RF, Shuvaeva LP, et al. Childhood leukemia in Belarus before and after the Chernobyl accident: continued follow-up. Radiat Environ Biophys. 2001;40(4):259–267. 41. International Commission on Radiological Protection (ICRP). Radiation dose to patients from radiopharmaceuticals. Publication 53. Annals ICRP. 1987;18:1–4. 42. Brenner DJ, Hall EJ. Computed tomography—an increasing source of radiation exposure. N Engl J Med. 2007;29(357):2277– 2284. 43. National Hemophilia Foundation Home Page. October 14, 2006, MASAC recommendation number 176, http://www.hemophilia. org (July 1, 2009). 44. de la Chapelle A, Rekonen A, Oka M, et al. Chromosome damage after intra-articular injections of radioactive yttrium: Effect of immobilization on the biological dose. Ann Rheum Dis. 1972; 31(6):508–12. 45. Gumpel JM, Stevenson AC. Chromosomal damage after intraarticular injection of different colloids of yttrium 90. Rheumatol Rehabil. 1975;14(1):7–12. 46. Lloyd DC, Reeder EJ. Chromosome aberrations and intraarticular yttrium-90. Lancet. 1978;18;1(8064):617. 47. Doyle DV, Glass JS, Gow PJ, et al. A clinical and prospective chromosomal study of yttrium-90 synovectomy. Rheumatol Rehabil. 1977;16(4):217–222. 48. Jalava S, Salonius AL. Letter: Chromosomes of patients treated with yttrium-90. Lancet. 1974:27;1(7861):807. 49. Rodrigues AS, Oliveira NG, Gil OM, Leonard A, Rueff J. Use of cytogenetic indicators in radiobiology. Radiat Prot Dosimetry. 2005;115:455–460. 50. Falcon de Vargas A, Fernandez-Palazzi F. Cytogenetic studies in patients with hemophilic hemarthrosis treated by 198Au, 186Rh, and 90Y radioactive synoviorthesis. J Pediatr Orthop B. 2000; 9(1):52–54. 51. Voth M, Klett R, Lengsfeld P, et al. Biological dosimetry after yttrium-90 citrate colloid radiosynoviorthesis. Nuklearmedizin. 2006;45(5):223–228. 52. Fernandez-Palazzi F, Caviglia H. On the safety of synoviorthesis in haemophilia. Haemophilia. 2001;7(Suppl. 2):50–53. 53. Prosser JS, Izard BE, Brown JK, et al. Induction of micronuclei in peripheral blood lymphocytes of patients treated for rheumatoid or osteoarthritis of the knee with dysprosium-165 hydroxide macroaggregates or yttrium- 90 silicate. Cytobios. 1993;73(292):7– 15. 54. Vuorela J, Sokka T, Pukkala E, et al. Does yttrium radiosynovectomy increase the risk of cancer in patients with rheumatoid arthritis? Ann Rheum Dis. 2003;62(3):251–53. 55. IRCP. IRCP-60: Recommendations of the IRCP. International Commission on Radiation Protection; 1991.

JCD 2009; 1:(1). OCTOBER 2009

36

www.slm-hematology.com


JOURNAL OF COAGULATION DISORDERS

REVIEW ARTICLE

Immunotolerance Induction Treatments in Hemophilia Saturnino Haya, Pilar Casan˜a, Andre´s Moret, Ana R Cid, Noelia Cabrera, Lydia Abad and Jose´ A Aznar Affiliation: Haemostasis and Thrombosis Unit, Hospital Universitario La Fe, Valencia, Spain Submission date: 1st July 2009, Revision date: 16th August 2009, Acceptance date: 24th August 2009

A B S T R A C T Hemophilia is the major congenital hemorrhagic disorder. The administration of the deficient factor can trigger the development of antibodies, also known as inhibitors, capable of neutralizing the infused protein function. This is considered the most serious complication in hemophilia and usually occurs in the first years of life after the first doses of factor concentrate are given; hence, the main objective in these patients is to permanently eradicate these inhibitors. Current treatments are based on immunotolerance induction (ITI) involving continued exposure to the deficient factor. The prognosis factors for success in ITI in severe hemophilia A have been obtained mainly from four registries and from several patient series. The strongest predictors of success are a low inhibitor titer at the beginning of the ITI and a maximum inhibitor titer. However, there are many other points to clarify: FVIII doses; the use of plasma-derived or recombinant concentrates, especially whether the plasma-derived concentrate contains large amounts of von Willebrand factor or not; concomitant infections during ITI; age at the start of ITI and delay in beginning it. Another controversial point is the use of immunosuppressants to help in the treatment of patients with a poor prognosis or in rescue treatments. In mild– moderate hemophilia, the incidence of inhibitors is lower and immunosuppressants can play an important role. Inhibitors in hemophilia B are rare. ITI usually leads to allergenic events and nephrotic syndrome, and the success rate is lower than in hemophilia A; therefore, ITI in hemophilia B must be carefully studied. Keywords: hemophilia A, hemophilia B, inhibitors, immunotolerance induction, treatment Correspondence: Saturnino Haya, Unidad de Hemostasia y Trombosis, Hospital Universitario La Fe, Avd. de Campanar 21, 46009 Valencia, Spain. Tel/fax: +34-961-973052; e-mail: haya_sat@gva.es

produces little or no increase in the inhibitor titer, showing titers ,5 BU, or ‘‘high responders’’ when factor exposure produces a fast increase in the inhibitor titer, showing titers .5 BU.

INTRODUCTION Hemophilia is the major congenital hemorrhagic diathesis. Hemophilia A results from molecular defects in the F8 gene, which cause a fall in or dysfunction of factor VIII (FVIII) in plasma. Hemophilia B is caused by defects in the F9 gene, reducing factor IX (FIX) plasma levels. Hemophilia is classified as severe when factor activity is below 1 IU/dL, moderate if between 1 and 5 IU/dL and mild when .5 IU/mL and ,40 IU/mL [1]. Treatment usually involves the administration of the deficient factor, but this treatment can trigger an immune response leading to the appearance of antibodies, or inhibitors, capable of neutralizing the infused protein function.

The management of hemophiliac patients with inhibitors is one of the greatest challenges for physicians who treat hemophilia, requiring considerable human and economic resources. In these patients, the immediate objective is the treatment of hemorrhagic episodes and, in the long term, permanent inhibitor elimination. To achieve inhibitor eradication, current treatments are based on continuous exposure to the triggering factor, the deficient factor. This treatment is maintained until the patient becomes tolerant to the administered product. A patient is considered tolerant when the inhibitor titer is negative and the recovery and half-life of the infused factor are normal. This is known as immunotolerance induction treatment (ITI).

Nowadays, the development of inhibitors is the most serious complication in hemophilia, which will determine to a great extent the treatment and the quality of life of hemophiliac patients. This problem is more frequent in hemophilia A (21–33%), especially in severely affected patients [2–4], than in hemophilia B. The appearance of inhibitors usually occurs in the first years of life, after receiving the first doses of factor concentrate [2–5]. Inhibitor patients can be classified as ‘‘low responders’’ when factor exposure JCD 2009; 1:(1). OCTOBER 2009

IMMUNOTOLERANCE INDUCTION TREATMENTS IN SEVERE HEMOPHILIA A Inhibitor eradication is the main objective in hemophiliacs who develop this condition. When tolerance is 37

www.slm-hematology.com


Journal of Coagulation Disorders

achieved, deficient factor concentrates can be used again to manage the bleeding episodes. This is the first therapeutic option as it has shown the best efficacy– cost relationship and also allows the use of traditional prophylactic treatments to prevent articular hemorrhages and the subsequent arthropathy. Indeed, the ability to use concentrates of the deficient factor improves the quality of life of these patients, which clearly deteriorates when the inhibitor is present.

from the previous two variables [22]. In the North American Registry [23], the good prognosis factors are considered to be a lower maximum inhibitor titer, a lower titer during ITI, and starting the treatment with titers below 10 BU/mL. Regarding the doses, there was an inverse relation between the dose given and the success of the treatment; patients with lower doses responded better to the treatment but, moreover, in the successful cases, the response was faster when higher doses were used. Using immunomodulation concomitantly did not influence the success rate. The German Registry [24] shows two variables of good prognosis: a lower maximum titer and a low inhibitor titer at the start of the ITI. In the Spanish Registry, updated in 2005 [25], three factors for good prognosis were observed: an inhibitor titer below 10 BU/mL at the start of the ITI, the use of doses ,100 IU/kg/day, and the absence of infection of the central venous access devices (CVAD).

To achieve inhibitor elimination, the same immunosuppressant treatments used in acquired hemophilia were tested initially, but success was substantially lower and diverse. In this way, there are papers that explore different cyclophosphamide regimes administered together with FVIII to avoid the anamnesic response with poor results [6]. In most of the studies, although the anamnesic response cannot be totally avoided, it is delayed and of lower intensity [7, 8]. Complete inhibitor eradication was reported in a few patients [9, 10], and sometimes the patient changed from being a high responder to being a low responder [11], but most of the cases failed [10, 12, 13].

The results on the ITI prognostic factors come from non-controlled studies, in which diverse doses of FVIII are used and sometimes different criteria for success or failure are established. This has led us to define some variables that may affect ITI, but there are still many aspects that need to be clarified.

After these discouraging results, Brackmann and Gormsen reported a case in which they managed to eradicate a high-titer inhibitor by administrating elevated doses of FVIII and activated prothrombin complex concentrate (APCC) every day [14]. This led to the eradication treatment lately known as the Bonn protocol [15], which uses 100–150 IU/kg/12 h FVIII with 50 U/kg/12 h APCC or without it, depending on the bleeding profile of the patient. In spite of the results obtained with this scheme, on account of its high cost, other protocols with lower doses and similar results appeared: Aznar et al [16] with daily administration of 50 IU/kg FVIII and steroid treatment at the end of ITI; Ewin et al [17] with the same doses but without the steroid treatment; and Mauser-Bunschoten et al [18, 19], who used even lower doses, 25 IU/kg every 48–72 h, obtained remarkably results.

Factors that can Influence the ITI Result Inhibitor Titer at the Start of ITI The parameter in all the series and registries concordant with a good prognosis is low inhibitor titer at the beginning of an ITI. In a compilation of the International and North American registries [26], patients who started the ITI with titers ,10 BU/mL achieved success in 85% of the cases and in a mean time of 11 months, compared with 33% and 15 months when the titer was .10 BU/mL. Studies in which the start of the ITI was delayed until the patients reached titers ,10 BU/mL showed good results [19, 27]. From this, it is recommended to delay the ITI until the inhibitor titer declines below 10 BU/mL. This fall usually lasts several months. During this time, products that may contain FVIII must be avoided, and the inhibitor titer should be measured periodically.

The Malmo¨ group designed an intensive ITI protocol, where, together with high FVIII doses, immunosuppressants and immunoglobulins were administered, and immunoadsorption was carried out in high inhibitor titer patients [20]. In recent years, this treatment scheme has been used less, mainly because of its difficult application in pediatric patients.

Maximum Inhibitor Titer In most of the series or registries, patients with higher inhibitor titers have worse prognosis [23, 24, 28]. Patients who showed a historical peak titer .500 BU/mL usually failed the tolerance treatment. In the North American registry, besides the historical peak titer, a high titer during the ITI is also considered to be a poor prognostic factor.

In the 1990s, several registries were published after retrospectively collecting all data related to patients under ITI. The major objective was to obtain more information about the prognostic factors that could influence the ITI response. The International Registry shows two variables associated with a higher ITI success: the use of doses .100 IU/kg/day and a low inhibitor titer (,10 BU/mL) at the start of the ITI [21]. In the 2001 update of this registry, a lower maximum inhibitor titer and a younger age at the start of ITI were also considered good prognostic factors, apart JCD 2009; 1:(1). OCTOBER 2009

FVIII Doses during ITI One of the most controversial points is the dose of factor concentrate used in an ITI. The North American 38

www.slm-hematology.com


Registry as well as the Spanish Registry show better prognosis when low doses are used; this contrasts with the International Registry, where doses >100 IU/kg/ day are associated with a better success rate. The analysis made by Kroner [26] of the International and North American registries shows that, in patients with a good prognostic profile, defined as a historical inhibitor titer ,200 BU/mL and starting ITI with a titer ,10 BU/mL, the success rate is not dependent on the FVIII dose used. In this same analysis, patients with a historical inhibitor titer .200 BU/mL and a starting titer .20 BU/mL would only have a reasonable chance of success if they are treated with doses of 200 IU/kg/day or above. Using high doses of factor also implies that patients who achieve tolerance do so in a shorter period of time.

obtained when the delay between detection and the start of the ITI is shorter [31, 32], although this is not confirmed in the registries. As the more significant parameter at this moment is the inhibitor titer at the start of ITI, it is natural to begin ITI once the inhibitor falls below 10 BU/mL or even below 5 BU/mL. To reach these titers, products containing FVIII should be avoided, and bleeding episodes during this time should be treated with recombinant activated factor VII (rFVIIa).

In the last few years, at our center, we have been administering high doses, 200 UI/kg 3 days a week, with good results.

Infection of the Central Venous Access Devices

The age at the start of ITI is another controversial point. In the International Registry [22], it is a significant factor, showing poorer results in patients over 20 years old. However, this is not observed in other registries.

It is known that any infection during ITI, and especially in the central venous access devices (CVAD), usually causes an unspecific increase in the inhibitor titer. CVAD infection can slow or hinder the immunotolerance. In a systematic review made by Valentino LA et al [34], the use of CVAD in patients with inhibitors or in children with less than 6 years were shown as infection independent risk factors. The first point is inherent in our work. Regarding age, most of the patients submitted to ITI are in their first years of life, because the inhibitors usually appear at a young age, with the first administrations of factor.

In patients with an a priori good prognosis, the use of high doses may not be justified. To try to clarify this, there is an ongoing study called the ITI Study (www.itistudy.com). This study includes severe hemophiliac A children aged ,8 years, with a maximum inhibitor titer between 5 and 200 BU/mL and with less than 2 years since the inhibitor was detected. It is a clinical trial in which patients are randomized in two arms depending on the dose: 50 IU/kg 3 days a week or 200 IU/kg/day.

One obvious advantage of using treatment guidelines with fewer administrations, such as 3 days a week, is the chance to carry it out without CVAD. Even if a CVAD is used, the risk of infection would still be lower given it is used less often. This could improve the ITI success rate.

The Factor Concentrate Traditionally, the same FVIII concentrate that triggered the appearance of inhibitor has been used for ITI. Analyzing the data from the registries, there are no significant differences between the factor concentrate used. Despite this, some works show better results when low-purity concentrates are used. Patients who did not respond to high-purity concentrates achieved success after changing to a concentrate containing von Willebrand factor (VWF) [29]. In the Bonn series, patients entered into ITI in the 1990s showed worse results (50% success) compared with those treated before the 1990s (above 80%). This efficacy decrease was attributed mainly to the change from plasma to recombinant products. Nevertheless, the individual characteristics of these patients were not published, and the possibility of other influencing factors existing cannot be discarded. The authors explained that concentrates with VWF may be more efficient in ITI because the VWF could be masking the C2 domain from inhibitor action and the half-life of the infused FVIII could be extended.

Immunosuppressants Immunosuppressive treatment alone, in contrast to acquired hemophilia, did not succeed in inhibitor eradication and was used less often [10, 12, 34]. Owing to their potential adverse effects and their poor utility alone, they are not used as the first ITI line. Furthermore, there is consensus for not using immunosuppression in ITI patients under 6 years of age; as most patients included in ITI are children, this is another reason why they are not used. However, immunosuppressants can play an important role when administered together with high doses of factor, especially in patients with poor prognosis: those with high inhibitor levels, those who start ITI with a high titer, or those who have failed a previous ITI [9, 35]. Hence, immunosuppressant use is considered in rescue ITI where patients do not respond to treatments with factor exclusively or in adult patients with worse prognosis. In these cases, immunosuppressants could reduce response time and costs. The Malmo¨ scheme defends the use of factor concentrates at high doses

When to Start ITI Some groups still recommend starting ITI as soon as possible after inhibitor detection [30]. This can be based on some series in which a better result is www.slm-hematology.com

39

JCD 2009; 1:(1). OCTOBER 2009


Journal of Coagulation Disorders

with cyclophosphamide and also immunoglobulins because using factor concentrate with only one of them did not manage to eradicate the inhibitor [35].

[43], although sometimes it only neutralizes exogenous FVIII [44]. Inhibitor with type I and II kinetics has been described, but the second is more common in mild and moderate congenital hemophilia [45, 46].

In recent years, hopeful results have been reported for the administration of Rituximab and FVIII in patients who failed a previous ITI [36–38]. Data on this aspect are not very profuse, and randomized prospective studies are needed to establish the usefulness of this product in inhibitor treatment in congenital hemophilia; also, some reports have been published on multifocal progressive leukoencephalopathy related to Rituximab, and this may contribute to the low application of this reagent for inhibitor eradication.

Usually in mild and moderate hemophilia, the inhibitors produce a deterioration in the hemorrhagic symptomatology, with a similar clinical profile to that in severe hemophiliacs and many of them with behavior analogous to that in acquired hemophilia. Traditionally, it was considered that most of the inhibitors in mild and moderate hemophilia were transitory. In the series reported by Hay et al [39], one third of the patients showed transitory inhibitors, another third eradicated the inhibitors after ITI, and the rest maintained the inhibitor. Bleeding episodes are usually treated with APCC or rFVIIa and sometimes with 1-desamino-8-D-arginine vasopressin (DDAVP). Avoiding exposure to FVIII may accelerate the inhibitor titer decrease. DDAVP will be more efficient in patients who do not present cross-reaction with their own FVIII. There are plenty of publications in which the usefulness of DDAVP has been reported [39, 44, 46–48]. However, with big hemorrhages, it can be difficult to maintain satisfactory levels of FVIII with DDAVP.

A second ITI option must be considered, depending on the bleeding frequency, severity, and responsiveness to bypassing reagents.

When to Stop ITI in Patients who do not Achieve Success Most patients achieve tolerance during the first year of treatment but, in some cases, the response is delayed for 2–3 years or even more [22]. It seems natural to stop ITI or to change the ITI regimen if the inhibitor level does not decrease after 6 months of treatment. In patients who show a fall in inhibitor level but not its total eradication, maintaining ITI for a longer or shorter time can depend on secondary benefits such as low bleeding episodes.

There are few publications on ITI in mild or moderate patients, and most of them are about isolated cases. The largest series is the British one [39], in which 8 out of 26 patients received immunotolerance. The Malmo¨ scheme was used in four patients: two were successful and the other two were considered partially successful. The Dutch scheme with low doses failed in one patient and partially succeeded in another. The Bonn protocol was applied in two patients with similar results to the Dutch one. Some works have also been published documenting the use of immunosuppressants [49]. In the last few years, it has been reported that Rituximab use in congenital hemophilia with inhibitors showed a better response in mild than in severe hemophilia [37, 50].

IMMUNOTOLERANCE INDUCTION TREATMENTS IN MILD AND MODERATE HEMOPHILIA A Classically, inhibitor development in mild or moderate hemophilia A patients was considered rare, but this idea changed after a report in 1998 published by British hemophilia care centers [39]. In this study and after 7 years of monitoring, 15 out of 57 new inhibitor cases occurred in patients with mild or moderate severity. These data showed that the annual incidence of inhibitor in mild or moderate hemophilia A patients was 0.84 for every 1000 patients, whereas in the severe illness, it was 3.5. In a later study published by Sharathkumar et al [40], 4 out of 54 mild hemophiliacs developed inhibitors, representing 7.4%. If we only consider those treated, this percentage rises to 14%, and even more if administration was by continuous infusion, 57% (four patients out of seven).

Although the data in the literature are limited, it appears that a better response is achieved when immunosuppressants are used. This may be because inhibitors in mildly affected patients behave similarly to autoantibodies in acquired hemophilia. So, in mild and moderate hemophilia with inhibitors, a valid option could be to use immunosuppressants alone as the treatment to eradicate the inhibitors. As information about inhibitors in mild and moderate hemophilia is scarce, we encourage researchers to include the data from these patients in studies or registries. In this sense, there is a French–Belgian initiative and the INSIGHT study (INternational Study on etiology of inhibitors in patients with a moderate or mild form of hemophilia A influences of Immuno Genetic and Hemophilia Treatment factors).

As is the case with severe hemophilia A, there is a parental predisposition and the mutation type plays a fundamental role; mild A hemophiliacs who develop inhibitors usually have mutations in the A2 and C2 domains. Another risk factor related to the appearance of inhibitor in these patients is the intensive exposure and the continuous infusion [40–43]. The inhibitor usually neutralizes endogenous and exogenous FVIII JCD 2009; 1:(1). OCTOBER 2009

40

www.slm-hematology.com


3. Lusher JM, Arkin S, Abildgaard CF, Schwartz RS. Recombinant factor VIII for the treatment of previously untreated patients with hemophilia A. Safety, efficacy, and development of inhibitors. Kogenate Previously Untreated Patient Study Group. N Engl J Med. 1993;328(7):453–459. 4. Bray GL, Gomperts ED, Courter S, et al. A multicenter study of recombinant factor VIII (recombinate): safety, efficacy, and inhibitor risk in previously untreated patients with hemophilia A. Blood. 1994;83:2428–2435. 5. Ehrenforth S, Kreuz W, Scharrer I, et al. Incidence of development of factor VIII and factor IX inhibitors in haemophiliacs. Lancet. 1992;339(8793):594–598. 6. Ruggeri ZM, Mannucci PM, Allain JP, Frommel D. Preliminary trial of cyclophosphamide in the management of hemophiliacs with factor VIII inhibitors. Ann NY Acad Sci. 1975;240:412–418. 7. Nilsson IM, Hedner U, Holmberg L. Suppression of factor VIII antibody by combined factor VIII and cyclophosphamide. Acta Med Scand. 1974;195(1-2):64–72. 8. Stein RS. Letter: Hemophilia: cyclophosphamide and factor VIII concentrate. Ann Intern Med. 1974;81(5):706–707. 9. Stein RS, Colman RW. Hemophilia with factor VIII inhibitor. Elimination of anamnestic response. Ann Intern Med. 1973;79(1): 84–87. 10. Hultin MB, Shapiro SS, Bowman HS, et al. Immunosuppressive therapy of Factor VIII inhibitors. Blood. 1976;48(1):95–108. 11. Hedner U, Tengborn L. Management of haemophilia A with antibodies—the effect of combined treatment with factor VIII, hydrocortisone and cyclophosphamide. Thromb Haemost. 1985; 54(4):776–779. 12. Hruby MA, Schulman I. Failure of combined factor VIII and cyclophosphamide to suppress antibody to factor VIII in hemophilia. Blood. 1973;42(6):919–923. 13. Povlsen JV, Moller B, Jacobsen SE, et al. Failure to induce immune tolerance in a hemophilia A patient with high titre VIII:C inhibitor. Thromb Res. 1989;53(1):79–84. 14. Brackmann HH, Gormsen J. Massive factor-VIII infusion in haemophiliac with factor-VIII inhibitor, high responder [letter]. Lancet. 1977;2(8044):933. 15. Brackmann HH, Oldenburg J, Schwaab R. Immune tolerance for the treatment of factor VIII inhibitors—twenty years’ ’Bonn protocol’. Vox Sang. 1996;70 Suppl 1:30–35. 16. Aznar JA, Jorquera JI, Peiro A, Garcia I. The importance of corticoids added to continued treatment with Factor VIII concentrates in the suppression of inhibitors in haemophilia A. Thromb Haemost. 1984;51(2):217–221. 17. Ewing NP, Sanders NL, Dietrich SL, Kasper CK. Induction of immune tolerance to factor VIII in hemophiliacs with inhibitors. JAMA. 1988;259(1):65–68. 18. van Leeuwen EF, Mauser-Bunschoten EP, van Dijken PJ, Kok AJ, Sjamsoedin-Visser EJM, Sixma JJ. Disappearance of factor VIII:C antibodies in patients with haemophilia A upon frequent administration of factor VIII in intermediate or low dose. Br J Haematol. 1986;64:291–297. 19. Mauser-Bunschoten EP, Nieuwenhuis HK, Roosendaal G, van den Berg HM. Low-dose immune tolerance induction in hemophilia A patients with inhibitors. Blood. 1995;86(3):983–988. 20. Nilsson IM. Immune tolerance. Semin Hematol. 1994;31(2 Suppl 4):44–48. 21. Mariani G, Ghirardini A, Bellocco R. Immune tolerance in hemophilia—principal results from the International Registry. Report of the factor VIII and IX Subcommittee. Thromb Haemost. 1994;72(1):155–158. 22. Mariani G, Kroner B. Immunotolerance in hemophilia with factor VIII inhibitors: predictors of success. Haematologica. 2001; 86(11):1186–1193. 23. DiMichele DM, Kroner BL. The North American Immune Tolerance Registry: practices, outcomes, outcome predictors. Thromb Haemost. 2002;87(1):52–57. 24. Lenk H. The German Registry of immune tolerance treatment in hemophilia--1999 update. Haematologica. 2000;85(10 Suppl):45–47.

IMMUNOTOLERANCE INDUCTION TREATMENT IN HEMOPHILIA B The incidence of inhibitor in hemophilia B is much lower than in hemophilia A, around 3% of patients [51–54]. The development of inhibitors in hemophilia B is usually related to large deletions in the F9 gene. The higher incidence observed in some geographic areas may be because there are more patients with those mutations in these areas [55]. Patients who develop inhibitors frequently show allergic events [56]. In patients who show allergic events, we must desensitize them before administering FIX concentrates; this desensitization usually involves exposures to increasing amounts of factor, making the administration of antihistamines and corticoids necessary before each exposure. Information on ITI in hemophilia B is limited, and the response rates are low, especially in patients who have developed allergic events. The larger series is the Inhibitor Registry of the International Society on Thrombosis and Haemostasis, which shows a success rate of 15% (5 out of 34) [57]. Thirteen patients developed a nephrotic syndrome during ITI. This condition, which is related to allergic events, can be reversible in some patients by stopping factor administration, but other patients also need the administration of corticoids [58]. The North American Registry includes 16 patients who have completed ITI with a 31% success rate. Allergic reactions are the main adverse events reported in this group and the major reason for treatment withdrawal. In this registry, we can also observe a relation between the allergic events and the appearance of the nephrotic syndrome in B hemophiliacs with inhibitors submitted to ITI. The best ITI results in hemophilia B are from the Malmo¨ group [59, 60], who have 67% success (six out of nine patients). The authors defend the simultaneous use of immunoglobulins in high doses and cyclophosphamide to improve the results. Moreover, isolated cases have been reported where success was achieved after the use of other immunosuppressants, for example mycophenolate, corticoids, and immunoglobulins together with high doses of FIX, without the appearance of the nephrotic syndrome [61]. Using immunosuppressants may improve the results in these patients, reducing the allergic events and thereby accelerating ITI. However, starting ITI in hemophilia B must be carefully weighed because of the high rate of adverse events and the low chances of success. Disclosures: The authors have no financial interests to disclose related to the contents of this article.

REFERENCES 1. White GC, Rosendaal FR, Aledort LM, et al. Definitions in Hemophilia. Thromb Haemost. 2001;85:560. 2. de Biasi R, Rocino A, Papa ML, Salerno E, Mastrullo L, De Blasi D. Incidence of factor VIII inhibitor development in hemophilia A patients treated with less pure plasma derived concentrates. Thromb Haemost. 1994;71(5):544–547.

www.slm-hematology.com

41

JCD 2009; 1:(1). OCTOBER 2009


Journal of Coagulation Disorders

45. Bovill EG, Burns SL, Golden EA. Factor VIII antibody in a patient with mild haemophilia. Br J Haematol. 1985;61(2):323– 328. 46. Santagostino E, Gringeri A, Tagliavacca L, Mannucci PM. Inhibitors to factor VIII in a family with mild hemophilia: molecular characterization and response to factor VIII and desmopressin. Thromb Haemost. 1995;74(2):619–621. 47. Kesteven PJ, Holland LJ, Lawrie AS, Savidge GF. Inhibitor to factor VIII in mild haemophilia. Thromb Haemost. 1984;52(1):50– 52. 48. Cid AR, Casana P, Cabrera N, Haya S, Cortina V, Aznar JA. Inhibitor development in one patient and laboratory discrepancies in several families with both mild haemophilia and Arg531Cys mutation. Haemophilia. 2007;13(2):206–208. 49. Vlot AJ, Wittebol S, Strengers PF, et al. Factor VIII inhibitor in a patient with mild haemophilia A and an Asn618RSer mutation responsive to immune tolerance induction and cyclophosphamide. Br J Haematol. 2002;117(1):136–140. 50. Dunkley S, Kershaw G, Young G, et al. Rituximab treatment of mild haemophilia A with inhibitors: a proposed treatment protocol. Haemophilia. 2006;12(6):663–667. 51. Sultan Y. Prevalence of inhibitors in a population of 3435 hemophilia patients in France. French Hemophilia Study Group. Thromb Haemost. 1992;67(6):600–602. 52. Shapiro AD, Ragni MV, Lusher JM, et al. Safety and efficacy of monoclonal antibody purified factor IX concentrate in previously untreated patients with hemophilia B. Thromb Haemost. 1996; 75(1):30–35. 53. Parquet A, Laurian Y, Rothschild C, et al. Incidence of factor IX inhibitor development in severe haemophilia B patients treated with only one brand of high purity plasma derived factor IX concentrate. Thromb Haemost. 1999;82(4):1247–1249. 54. Shapiro AD, Di PJ, Cohen A, et al. The safety and efficacy of recombinant human blood coagulation factor IX in previously untreated patients with severe or moderately severe hemophilia B. Blood. 2005;105(2):518–525. 55. Ljung R, Petrini P, Tengborn L, Sjorin E. Haemophilia B mutations in Sweden: a population-based study of mutational heterogeneity. Br J Haematol. 2001;113(1):81–86. 56. Thorland EC, Drost JB, Lusher JM, et al. Anaphylactic response to factor IX replacement therapy in haemophilia B patients: complete gene deletions confer the highest risk. Haemophilia. 1999;5(2):101–105. 57. Key NS. Inhibitors in congenital coagulation disorders. Br J Haematol. 2004;127(4):379–391. 58. Ewenstein BM, Takemoto C, Warrier I, et al. Nephrotic syndrome as a complication of immune tolerance in hemophilia B [letter]. Blood. 1997;89(3):1115–1116. 59. Nilsson IM, Berntorp E, Zettervall O. Induction of split tolerance and clinical cure in high-responding hemophiliacs with factor IX antibodies. Proc Natl Acad Sci USA. 1986;83(23):9169–9173. 60. Berntorp E, Astermark J, Carlborg E. Immune tolerance induction and the treatment of hemophilia. Malmo protocol update. Haematologica. 2000;85(10 Suppl):48–50. 61. Klarmann D, Martinez S, I, Funk MB, et al. Immune tolerance induction with mycophenolate-mofetil in two children with haemophilia B and inhibitor. Haemophilia. 2008;14(1):44–49.

25. Haya S, Quintana M, Aznar J, et al. Registro Espan˜ol de Inmunotolerancia en hemofı´licos con inhibidores. Actualizacio´n a 2005. Haematologica. 2006;91(Spanish edition, Suppl 4):1–5. 26. Kroner BL. Comparison of the international immune tolerance registry and the North American immune tolerance registry. Vox Sang. 1999;77(Suppl 1):33–37. 27. Smith MP, Spence KJ, Waters EL, et al. Immune tolerance therapy for haemophilia A patients with acquired factor VIII alloantibodies: comprehensive analysis of experience at a single institution. Thromb Haemost. 1999; 81(1):35–38. 28. Haya S, Lo´pez MF, Aznar JA, Batlle J, the Spanish Immune Tolerance Group. Immune tolerance treatment in haemophilia patients with inhibitors: the Spanish Registry. Haemophilia. 2001;7(2):154–159. 29. Kreuz W, Mentzer D, Auerswald G, Becker S, Joseph-Steiner J. Successful immunotolerance therapy of FVIII inhibitor in children after changing from high to intermediate purity FVIII concentrate. Haemophilia. 1996;2(Suppl 1):19. 30. Astermark J, Morado M, Rocino A, et al. Current European practice in immune tolerance induction therapy in patients with haemophilia and inhibitors. Haemophilia. 2006;12(4):363–371. 31. Kreuz W, Ehrenforth S, Funk M, et al. Immune tolerance therapy in pediatric haemophiliacs with factor VIII inhibitors: 14 years follow-up. Haemophilia. 1995;1(1):24–32. 32. Mathias M, Liesner R, Hann I, Khair K. Immune tolerance in children with factors VIII and IX inhibitors: a single centre experience. Haemophilia. 2005;11(4):340–345. 33. Valentino LA, Ewenstein B, Navickis RJ, Wilkes MM. Central venous access devices in haemophilia. Haemophilia. 2004;10(2): 134–146. 34. Lian EC, Larcada AF, Chiu AY. Combination immunosuppressive therapy after factor VIII infusion for acquired factor VIII inhibitor. Ann Intern Med. 1989;110(10):774–778. 35. Nilsson IM, Berntorp E, Zettervall O. Induction of immune tolerance in patients with hemophilia and antibodies to factor VIII by combined treatment with intravenous IgG, cyclophosphamide, and factor VIII. N Engl J Med. 1988;318(15):947–950. 36. Mathias M, Khair K, Hann I, Liesner R. Rituximab in the treatment of alloimmune factor VIII and IX antibodies in two children with severe haemophilia. Br J Haematol. 2004;125(3): 366–368. 37. Carcao M, St Louis J, Poon MC, et al. Rituximab for congenital haemophiliacs with inhibitors: a Canadian experience. Haemophilia. 2006; 12(1):7–18. 38. Mateo J, Badell I, Forner R, Tizzano E, Foncuberta J. Rituximab en el tratamiento de un nin˜o afecto de hemofilia A severa con anticuerpos aloinmunes contra el factor VIII. Haematologica. 2004;89[Extraordinario 2]:153. 39. Hay CRM, Ludlam CA, Colvin BT, et al. Factor VIII inhibitors in mild and moderate-severity haemophilia A. Thromb Haemost. 1998;79(4):762–766. 40. Sharathkumar A, Lillicrap D, Blanchette VS, et al. Intensive exposure to factor VIII is a risk factor for inhibitor development in mild hemophilia A. J Thromb Haemost. 2003;1(6):1228– 1236. 41. Thompson AR, Murphy ME, Liu M, et al. Loss of tolerance to exogenous and endogenous factor VIII in a mild hemophilia A patient with an Arg593 to Cys mutation. Blood. 1997;90(5):1902– 1910. 42. White B, Cotter M, Byrne M, O’Shea E, Smith OP. High responding factor VIII inhibitors in mild haemophilia—is there a link with recent changes in clinical practice? Haemophilia. 2000;6(2):113–115. 43. Koestenberger M, Leschnik B, Muntean W. More on: mild hemophilia A and inhibitor development. J Thromb Haemost. 2004;2(4):676. 44. Peerlinck K, Jacquemin MG, Arnout J, et al. Antifactor VIII antibody inhibiting allogeneic but not autologous factor VIII in patients with mild hemophilia A. Blood. 1999;93(7):2267–2273.

JCD 2009; 1:(1). OCTOBER 2009

42

www.slm-hematology.com


JOURNAL OF COAGULATION DISORDERS

REVIEW ARTICLE

An Overview of Blastocystis hominis Infection and Published Experience in Hemophilic Population Carlos Aguilar1 and Jose´ F Lucı´a2 Affiliations: 1Department of Haematology, Hospital General Santa Ba´rbara, Soria, Spain and 2Department of Haematology, Hospital Universitario Miguel Servet, Zaragoza, Spain Submission date: 10th June 2009, Revision date: 16th July 2009, Acceptance date: 16th August 2009

A B S T R A C T Blastocystis hominis is a parasite with worldwide spread that infects from 10–15% to 30–50% of asymptomatic individuals in developed and developing countries respectively. Nine subtypes of the parasite have been reported, and the diagnosis usually involves microscopic observation of vacuolar or cystic forms of the parasite in stools. Controversy regarding the commensal or pathogenic nature of the infection has remained for decades, but recent epidemiological and microbiological studies support an emerging pathogenic potential for the parasite. Clinical manifestations of the disease include non-specific gastrointestinal symptoms (namely diarrhea, abdominal pain, nausea), which are usually self-limited. Specific treatment, most often with metronidazol or cotrimoxazol, is able to achieve more rapid resolution of symptoms and clearance of the infection. Bleeding has not been shown to be a common manifestation or complication of blastocystosis. Only two cases of enterocolitis caused by Blastocystis hominis infection have been reported in hemophiliacs, and only one of them presented with lower gastrointestinal bleeding, which rapidly resolved after treatment of the infection with metronidazol and a short course of recombinant factor VIII. There seem to be no specific features of Blastocystis hominis infection in hemophiliacs and, where bleeding occurs, it is unlikely to be severe and can probably be readily stopped with isolated doses of factor VIII/IX or DDAVP. Investigations to rule out other infectious or non-infectious causes of intestinal bleeding should be carried out in such cases. Keywords: hemophilia, gastrointestinal bleeding, Blastocystis hominis Correspondence: Dr Carlos Aguilar, Department of Haematology, Hospital General Santa Ba´rbara, Paseo Santa Ba´rbara s/ n, 42.002 Soria, Spain. Tel: +34-975-214156; fax: +34-975-234321; e-mail: caraguilar@excite.com

different subtypes with subtype 3 being the only one with a purely human origin. As a result of this new classification and the existing evidence that humans can be infected by numerous genotypes (many of which are of animal origin), Tan [1] has suggested that this subtype should be the only one referred to as B. hominis, whereas laboratories should report the rest as Blastocystis sp.

INTRODUCTION Blastocystis hominis is an anaerobic parasite that is frequently isolated from the gastrointestinal tract of humans and a wide range of animals [1]. It was originally discovered in 1911 and considered for decades as an innocuous yeast, but it was not until the 1970s that evidence suggested that B. hominis was actually a protozoan; this idea was supported by the fact that it was sensitive to antiprotozoal drugs and proved to be unable to grow on fungal media [2]. In recent years, there have been significant advances in determining its biology, and genetic analyses performed in 1996 showed that Blastocystis is not actually a fungus or a protozoan; since then, its classification has undergone major reviews which definitely place it into Stramenopiles [3]. Such analyses have also shown that, in contrast to previous knowledge, substantial morphological and genetic differences can be found among Blastocystis isolates from both humans and animals. Application of DNA studies to the isolates mentioned above resulted in the recognition of nine JCD 2009; 1:(1). OCTOBER 2009

EPIDEMIOLOGICAL AND MICROBIOLOGICAL ASPECTS Blastocystis can be found worldwide, but it has been reported to be more prevalent in developing countries, presumably because of poor hygiene, closer contact with animals, and the intake of contaminated food or water. This parasite has been found to be the one most commonly isolated from the human gastrointestinal tract [4]. Prevalence of the infection can range from 10–15% to 30–50% in healthy asymptomatic individuals from developed and developing countries respectively [5]. The mode of transmission remains unclear, 43

www.slm-hematology.com


Journal of Coagulation Disorders

transplants or chemotherapy, or individuals suffering from disturbances in intestinal function from other causes [8, 15–17].

but the fecal–oral mechanism has been claimed to be the most likely; contaminated water has been increasingly implicated as the source of Blastocystis infections [6, 7].

Different subtypes of the parasite seem to have a different pathogenic behavior; thus, subtypes 1, 2, and 4 have been reported to be more closely linked to symptomatic infections, whereas subtypes 3 and 7 result in clinically silent cases [18]. Mixed infections by different subtypes have been reported; among these, co-infections with subtypes 1 and 3 are the most common. The relationship between the pathogenicity of Blastocystis and large amounts of parasites isolated (regarded as .5 per high-power or oil immersion field) has also been suggested, but the current consensus argues against such a correlation [5, 9].

Four different morphological forms of the parasite have been reported (vacuolar, granular, ameboid, and cystic), with the cystic stage being the most commonly involved in transmission of the infection, on account of its great resistance to environmental conditions, and the ameboid form having a more active role in the development of clinical manifestations of the disease. Still, the vacuolar morphological subtype is the most readily observed in both laboratory culture and stool samples.

CLINICAL ISSUES AND DIAGNOSIS

Diagnosis of Blastocystis infection most commonly involves microscopic observation of the various forms of the parasite in fecal smears stained with Lugol’s iodine or trichrome; vacuolated and cystic forms are found most commonly in these specimens. However, observation of Blastocystis is often laborious, and serial stool analyses might be necessary before a definite exclusion of the diagnosis can be made. Other diagnostic procedures such as stool cultures (more sensitive than microscopy), serological assays (namely enzyme-linked immunosorbent assay or ELISA), or polymerase chain reaction (PCR) tests are also available, but are not used on a routine basis, only for epidemiological and research purposes [9, 19, 20].

Regarding clinical aspects, there has been a long debate in the literature about whether B. hominis is an intestinal commensal or a true pathogen [7, 8], but evidence has accumulated over the last decade from epidemiological and microbiological studies in support of the emerging pathogenic potential of the parasite [1]. Symptoms of the infection are largely non-specific and include acute or chronic diarrhea (often watery in nature), nausea, vomiting, flatulence, abdominal pain, and anorexia. Fever is uncommon in this setting. Intestinal infection by B. hominis can mimic or be associated with other intestinal pathogens (protozoan, bacterial, or viral) or other non-infectious gastrointestinal disorders, so thorough exclusion of such conditions must always be considered in patients presenting with consistent clinical features. Analytical abnormalities such as fecal leukocytosis or peripheral eosinophilia have been reported but are also non-specific [9]. Lower gastrointestinal bleeding as a complication of this infection is uncommon given the generally non-invasive nature of the infection, as shown by endoscopy [10]. Symptoms are most commonly self-limited and may last between 1 and 2 weeks, but administration of some of the antibiotic or antiprotozoal drugs that will be mentioned later have proved to be helpful in reducing the duration of symptoms and increasing the eradication rate of Blastocystis compared with patients given no specific treatment [11].

TREATMENT OF THE INFECTION BY BLASTOCYSTIS HOMINIS Treatment of the infection has also been a source for controversy. There is general agreement that asymptomatic carriers of the infection do not require any treatment. The need for treatment in symptomatic patients after exclusion of other pathogens or etiology appears to be warranted at present despite the often self-limiting nature of the infection; responses may result from eradication of Blastocystis or some other accompanying germs that offer support for parasitic growth. There are several therapeutic options and schemes available, which are summarized in Table 1. Among them, metronidazol can be considered the first choice treatment; a placebo-controlled trial concluded that metronidazol achieved clinical remission in a much higher proportion of cases at 1 (88% vs 14%) and 6 months (75% vs 14%) as well as significantly more common fecal clearance of B. hominis (80% vs 3%) [11]. Cotrimoxazol is another useful second-choice therapeutic tool with eradication rates and resolution or improvement of symptoms in over 90% of patients treated [21]. Nitazoxanide or paromomycin can also be active against the infection; the latter has been shown to be successful in treating Blastocystis infections associated with cutaneous lesions, predominantly urticaria [22].

Associations of Blastocystis infection with other intestinal (irritable bowel syndrome [12]) or nonintestinal (urticaria or other cutaneous rashes [13], iron deficiency [14]) conditions have been suggested but not clearly established. Some clinical features have been reported to contribute to susceptibility to the infection and the development of Blastocystisassociated clinical manifestations; most of such conditions are related to impairment of normal immunity such as HIV infection (especially in cases with very low CD4 counts below 200/mL), patients on immunosuppressive drugs for solid organ or bone marrow JCD 2009; 1:(1). OCTOBER 2009

44

www.slm-hematology.com


Table 1. Drugs and Therapeutic Schemes available for Blastocystis Infections Drug Metronidazol

Dosage Adult

750 mg t.d.s. 6 10 days 500 mg t.d.s. 6 10 days 1.5 g/day single dose 6 10 days

Pediatric

15 mg/kg/day 6 7 days 20–30 mg/kg/day 6 10 days

Trimethoprim–sulfametoxazol (TMP–SMX)

Adult

320 mg TMP and 1600 mg SMX daily in two equal doses 6 7 days

Nitazoxanide

Pediatric Adult

6 mg/kg TMP and 30 mg/kg SMX daily in two equal doses 6 7 days 500 mg b.d. 6 3 days

Pediatric

Paramomycin

1–3 years

100 mg b.d. 6 3 days

4–11 years

200 mg b.d. 6 3 days

25 mg/kg t.d.s. 6 10 days

PUBLISHED EXPERIENCE IN THE HEMOPHILIA SETTING

observed in the stools any more and a fecal occult blood test was negative.

The relevance of symptomatic B. hominis infection in the hemophilic setting remains scarce. Actually only two cases of hemophiliacs diagnosed with Blastocystisrelated enterocolitis have been reported in the international English language literature for the time being; both had different presentation patterns. One of them [23] was a hemophiliac suffering from advanced HIV infection who presented with gastrointestinal symptoms and was found to have a high density of B. hominis in his duodenal secretions and stools. No gastrointestinal bleeding complicated the infection. Furazolidone therapy was successful in achieving clinical improvement and eradication of the organism. The patient died 6 weeks after the infection was treated from AIDS-related complications. The authors raised the issue that B. hominis might behave as an opportunistic pathogen in immunocompromised patients; this issue remains unclear two decades later, but Tan [1] points out that there is an increasing body of evidence supporting this approach.

There are some issues in the case reported that deserve some more detailed discussion. First, bleeding from enteric injuries caused by Blastocystis has not been reported before because of the non-invasive nature of the infection as mentioned above, and the existence of inflammatory diarrhea always makes exclusion of inflammatory bowel disease mandatory even though, for unknown reasons, the association of this condition with congenital bleeding disorders has very rarely been reported; a possible contribution of thrombosis to the pathogenesis of this condition has been claimed as a reason for this somewhat protective role [25]. It has been reported that B. hominis can induce the production of interleukin (IL)-8 and granulocyte–macrophage colony-stimulating factor (GM-CSF) in epithelial cells mediated by the activation of cell surface receptors by cysteine proteases without direct invasion of the intestinal epithelium; this results in an influx of inflammatory cells into the intestinal mucosa leading to tissue damage, barrier compromise, and gastrointestinal disturbances [26]. Surprisingly, our patient presented with endoscopic signs of mucosal disruption, which was most likely a triggering factor for the bleeding; the high density of parasites found in stools or undiagnosed associated infections might have had an influence on clinical presentation.

We also reported [24] a second case of an HIVnegative mild hemophilia A patient who presented with a 6-week history of watery-type diarrhea associated with colic-type central abdominal pain, flatulence, and rectal tenesmus, which became complicated by hemorrhagic rectal discharge. Regular stool cultures and a Clostridium difficile toxin test were negative. A colonoscopy showed generalized edema and hyperemia of the colic mucosa as well as colonic ulcerations. Treatment with recombinant factor VIII concentrate (two doses of 30 IU/kg each) rapidly stopped the bleeding but not the rest of the intestinal symptoms, which were just slightly milder 4 weeks later. Further microbiological investigations revealed the presence of abundant vacuolar forms of B. hominis in stool samples. The patient was started on metronidazol (500 mg t.d.s. p.o. for 10 days) with rapid and full improvement in his symptoms. Four weeks after the antibiotic had been stopped, B. hominis could not be www.slm-hematology.com

The role that hemophilia might play in bleeding in this setting is uncertain but probably limited. There are no reports about specific infection rates by B. hominis in hemophiliacs, but we can easily assume that such rates are similar to those reported for the general population living in the same geographical areas and overall high (especially for hemophilic subjects from developing countries); however, cases of bleeding following this infection have not been reported to date. This means that, should intestinal bleeding arise, local injuries might be responsible, and hemophilia might at most contribute to exacerbating 45

JCD 2009; 1:(1). OCTOBER 2009


Journal of Coagulation Disorders

17. Tasova Y, Sahin B, Koltas S, Paydas S. Clinical significance and frequency of Blastocystis hominis in Turkish patients with haematological malignancy. Acta Med Okayama. 2000;54:133– 136. 18. Kaneda Y, Horiki N, Cheng XJ, Fujita Y, Maruyama M, Tachibana H. Ribodemes of Blastocystis hominis isolated in Japan. Am J Trop Med Hyg. 2001;65:393–396. 19. Zierdt CH, Zierdt WS, Nagy B. Enzyme-linked immunosorbent assay for detection of serum antibody to B. hominis in symptomatic infections. J Parasitol. 1995;81:127–129. 20. Stensvold CR, Traub RJ, von Samson-Himmelstjerna G, Jespersgaard C, Nielsen HV, Thompson RC. Blastocystis: subtyping isolates using pyrosequencing technology. Exp Parasitol. 2007;116:111–119. 21. Ok UZ, Girginkardesler N, Balcioglu C, Ertan P, Pirildar T, Kilimcioglu AA. Effect of trimethoprim–sulfametoxazol in Blastocystis hominis infection. Am J Gastroenterol. 1999;94: 3245–3247. 22. Kick G, Rueff F, Przybilla B. Palmoplantar pruritus subsiding after Blastocystis hominis eradication. Acta Derm Venereol. 2002; 82:60. 23. Narkewicz MR, Janoff EN, Sokol RJ, Levin MJ. Blastocystis hominis gastroenteritis in a haemophiliac with acquired immune deficiency syndrome. J Pediatr Gastroenterol Nutr. 1989;8:125– 128. 24. Lucı´a JF, Aguilar C, Betra´n A. Blastocystis hominis colitis in a haemophilic patient as a cause of lower gastrointestinal bleeding. Haemophilia. 2007;13:224–225. 25. Thompson NP, Wakefield AJ, Pounder RE. Inherited disorders of coagulation appear to protect against inflammatory bowel disease. Gastroenterology. 1995;108:1011–1015. 26. Berkes J, Viswanathan K, Savkovic D, Hecht G. Intestinal epithelial responses to enteric pathogens: effects on the tight junction barrier, ion transport and inflammation. Gut. 2003;52: 439–451.

the severity of the hemorrhage. Given the self-limiting character of B. hominis infection, treatment of the infection as well as administration of a short course of factor VIII/IX infusions or isolated use of DDAVP for mild hemophilia A cases should be enough, as bleeding should not be expected to be severe.

CONCLUSION Thus, even though B. hominis commonly infects hemophilic individuals, it is a rare cause of lower gastrointestinal bleeding in these patients; investigation of parasites in stools from patients with enterocolitis of slow resolution or atypical clinical course compared with a regular enteric infection is warranted. Should bleeding arise, it would be a powerful reason to rule out other potential intestinal disorders. Disclosures: The authors have no financial interests to disclose related to the contents of this article.

REFERENCES 1. Tan KS. New insights on classification, identification and clinical relevance of Blastocystis spp. Clin Microbiol Rev. 2008; 21:639–665. 2. Zierdt CH. Studies of Blastocystis hominis. J Protozool. 1973;20: 114. 3. Silberman JD, Sogin ML, Leipe DD, Clark CG. Human parasite finds taxonomic home. Nature. 1996;380:398. 4. Taamasri P, Mungthin M, Rangsin B, Tongupprakarn B, Areekuf W, Leelayoova S. Transmission of intestinal blastocystosis related to the quality of drinking water. Southeast Asian J Trop Med Public Health. 2000;31:112–117. 5. Senay H, MacPherson D. Blastocystis hominis: epidemiology and natural history. J Infect Dis. 1990;162:987–990. 6. Leelayoova S, Rangsin B, Taamasri P, Naaglor T, Thathaisong U, Mungthin M. Evidence of waterborne transmission of Blastocystis hominis. Am J Trop Med Hyg. 2004;70:658–662. 7. Doyle PW, Helgason MM, Mathias RG, Proctor EM. Epidemiology and pathogenicity of Blastocystis hominis. J Clin Microbiol. 1990;28:116–121. 8. Editorial. Blastocystis hominis: commensal or pathogen? Lancet. 1991;337:521–522. 9. Stenzel DJ, Boreham PFL. Blastocystis hominis revisited. Clin Microbiol Rev. 1996;9:563–584. 10. Zuckerman MJ, Watts MT, Ho H, Meriano FV. Blastocystis hominis infection and intestinal injury. Am J Med Sci. 1994;308: 96–101. 11. Nigro L, Laroca L, Massarelli L, et al. A placebo controlled treatment trial of Blastocystis hominis infection with metronidazol. J Travel Med. 2003;10:128–130. 12. Yakoob J, Jafri W, Jafri M, et al. Irritable bowel syndrome: in search of an etiology. Role of Blastocystis hominis. Am J Trop Med Hyg. 2004;70:383–385. 13. Valsecchi R, Leghissa P, Greco V. Cutaneous lesions in Blastocystis hominis infection. Acta Derm Venereol. 2004;84:322– 323. 14. Yavasoglu I, Kadikoylu G, Uysal H, Ertug S, Bolaman Z. Is Blastocystis hominis a new etiologic factor or coincidence in iron deficiency anemia? Eur J Hematol. 2008;81:47–50. 15. Kurniawan A, Karyadi T, Dwintassari SW, et al. Intestinal parasitic infections in HIV/AIDS patients presenting with diarrhoea in Jakarta, Indonesia. Trans R Soc Trop Med Hyg. 2009;103: 892–898. 16. Ok UZ, Cirit M, Uner A, et al. Cryptosporidiasis and blastocystosis in renal transplant patients. Nephron. 1997;75:171–174.

JCD 2009; 1:(1). OCTOBER 2009

46

www.slm-hematology.com


JOURNAL OF COAGULATION DISORDERS

REVIEW ARTICLE

Rituximab in the Treatment of Postpartum Acquired Hemophilia A Maria Gabriella Mazzucconi, Francesca Biondo and Cristina Santoro Affiliation: Haematology Institute, Dipartimento di Biotecnologie cellulari ed Ematologia, ‘‘Sapienza’’ University of Rome, Rome, Italy Submission date: 10th July 2009, Revision date: 2nd September 2009, Acceptance date: 14th September 2009

A B S T R A C T Acquired hemophilia A is a rare disorder caused by the development of autoantibodies against coagulation factor VIII: it is often characterized by severe bleeding tendency. Its incidence is 0.2–1 per million persons per year. It may be associated with malignancies, autoimmune disorders, or pregnancy but, in about 50% of cases, no cause can be identified. Postpartum acquired hemophilia A represents 7–21% of all cases, with an incidence of about 1/350 000 deliveries. The aim of the management of acquired hemophilia A and, in particular of postpartum acquired hemophilia A, is the control of bleeding, if present, and the eradication of the inhibitor. The latter can be obtained with immunosuppressive drugs. However, cyclophosphamide and other alkylating agents should not be used in young women with postpartum inhibitor because of the risk of infertility. Complete remission rate obtained with these agents is about 80% or more but, in some cases, persistence of the inhibitor despite treatment or relapse after response represent an actual risk of recurrent bleedings. Therefore, alternative therapeutic approaches are advisable. Rituximab has been used in acquired hemophilia A since 2001 either as first-line therapy or in resistant/relapsed patients. Its efficacy is very high (about 90% eradication rate); it is safe and well tolerated, especially in young people. For these reasons, Rituximab can play a role in the therapeutic approach to postpartum acquired hemophilia A. Few cases have been treated up to now, but the results are very promising. Inhibitor eradication was obtained in all cases. No relevant side-effects were noted, nor did infections occur. Cytotoxic immunosuppressive drugs can compromise fertility in young female patients, whereas Rituximab can be a valid and safe alternative treatment. Keywords: acquired hemophilia A, inhibitor to factor VIII, immunosuppressive therapy, postpartum factor VIII inhibitor, Rituximab Correspondence: Maria Gabriella Mazzucconi, Associate Professor, Dipartimento di Biotecnologie cellulari ed Ematologia, ‘‘Sapienza’’ University of Rome, via Benevento n. 6, 00161, Roma, Italy. Tel: +39-06-857951-85795778; fax: +39-0644241984; e-mail: mazzucconi@bce.uniroma1.it

symptoms, in acquired hemophilia, the most common sites of bleeding are skin, mucosa, muscles, and retroperitoneum. The severity of bleeding is not proportional to the inhibitor titer [5].

INTRODUCTION Acquired hemophilia A is a rare disorder caused by the development of autoantibodies directed against coagulation factor VIII (FVIII). Often, it is characterized by a severe bleeding tendency. An estimated incidence of 0.2–1 per million persons per year has been reported. The incidence increases with age: 50– 60% of the cases occur over the age of 50 years, and the incidence is higher in men (about 53%), but in the youngest patients, the incidence is higher in women, because of the relationship with pregnancy [1–3].

In some cases (postpartum, drug-induced inhibitors), the inhibitors tend to disappear spontaneously within a few months after delivery or drug discontinuation [4]. Treatment or cure of the underlying disease causes the disappearance of the inhibitor [6]. The aims of the therapeutic approach are the control of bleeding and the eradication of the inhibitor autoantibodies. With regard to the treatment of bleeding, the main tools are bypassing agents (activated prothrombin complex concentrates (APCC), recombinant activated FVII (rFVIIa), human and recombinant FVIII concentrates and desmopressin (DDAVP)). Although up to 36% of all patients affected by acquired hemophilia A, who do not receive therapy

This disorder may be associated with pregnancy, malignancies, and autoimmune disorders, but in a consistent proportion of cases, about 50%, no cause can be identified at diagnosis [1]. Risk of mortality varies from 8% to 22%, usually from fatal bleedings [4]. At variance from congenital hemophilia, in which joint hemorrhages are the most common bleeding JCD 2009; 1:(1). OCTOBER 2009

47

www.slm-hematology.com


Journal of Coagulation Disorders

for inhibitor eradication, experience a spontaneous disappearance of their autoantibodies, this occurrence is unpredictable, and the patients remain at risk of bleeding as long as autoantibodies persist, especially if very low levels of FVIII are detected [7]. Thus, a therapeutic approach aimed at inhibitor eradication is mandatory, as recommended recently by Huth-Ku¨hne et al [8]. Treatments for inhibitor eradication include immunosuppression with corticosteroids, cyclophosphamide, cyclosporine, and other cytotoxic drugs. Treatment with high-dose immunoglobulin (HDIg), as first-line therapy without concomitant immunosuppressive agents, achieved only a 12% success rate [9], and the recent study by Collins et al [3], comparing patients who either did or did not receive HDIg, showed no benefit for those taking HDIg. So, the use of HDIg, as a single agent is not recommended in acquired hemophilia A [8].

In recent years, many reports have suggested a significant role for Rituximab (RTX) in the treatment of patients with acquired FVIII inhibitors. Rituximab is a chimeric murine/human monoclonal antibody of immunoglobulin G1 kappa type directed against CD20, a transmembrane protein expressed on the surface of premature and mature B lymphocytes. It depletes B cells from the blood, lymph nodes, and bone marrow and has demonstrated efficacy in the treatment of CD20-positive lymphoproliferative diseases. It blocks the proliferation of normal B cells, thus interfering with antibody production of the IgG type. In fact, several studies have reported therapeutic efficacy of RTX in a number of autoimmune diseases, such as systemic lupus erythematosus, rheumatoid arthritis, autoimmune hemolytic anemia, autoimmune thrombocytopenic purpura, thrombotic thrombocytopenic purpura, and acquired hemophilia A [14–16]; RTX may induce immune tolerance in congenital hemophilia with inhibitor alloantibodies to FVIII or FIX [17].

Corticosteroids and cyclophosphamide are regarded as the most effective therapeutic approach. Corticosteroids given at standard doses (e.g., prednisone 1 mg/ kg daily for 3–6 weeks) eradicate the inhibitor in about 30–50% of cases [10]. In general, responder patients have low inhibitor titers and no underlying diseases [11]. Cyclophosphamide (at doses of 1–2 mg/kg daily) alone or in combination with steroids may eradicate the inhibitor in patients resistant to corticosteroid [10]. Numerous authors also consider this drug as initial treatment in combination with prednisone in patients with very high inhibitor titers [12]. Moreover, a meta-analysis by Delgado et al [4] concerning a review of 20 reports showed that the higher response rate obtained with cyclophosphamide/prednisone combination therapy did not translate to lower mortality. This is probably due to the toxicity of the combination therapy in elderly patients, specifically for neutropeniarelated infections [13]. Recently, in a non-randomized study, Collins et al [3] compared patients treated with corticosteroids alone (n534) vs corticosteroids plus cytotoxic drugs (n545): response rates were 76% and 78% respectively. However, about 20–30% of the patients are refractory to these treatments, and up to 20% of patients responding to immunosuppressive treatments relapse and require additional treatment [13]. In order to eradicate the inhibitor, recently published international recommendations on the diagnosis and treatment of patients with acquired hemophilia A suggested corticosteroids as first-line therapy, at a dose of 1 mg/kg/day p.o. (prednisone) for 4–6 weeks, either alone or in combination with cyclophosphamide at a dose of 1.5–2 mg/kg/day for a maximum of 6 weeks. However, the potential risks and benefits of immunosuppressive therapy should be considered in individual patients: women with postpartum inhibitor should not be treated with cyclophosphamide or other alkylating drugs because of the risk of infertility [8]. JCD 2009; 1:(1). OCTOBER 2009

Figure 1 reports the characteristics and structure of the RTX molecule and its target (CD20 protein). Since the description of the first three cases treated in 2001 by Karwal et al [18], RTX has been used in adult patients (aged 18–94 years) with acquired hemophilia A, both as first-line therapy and as salvage treatment in cases refractory to conventional immunosuppressive drugs. In most studies, the therapy schedule was 375 mg/m2 per dose, once a week for 4 weeks. Rituximab has been used either as a single agent or in association with corticosteroids or other immunosuppressive drugs; the latter course is used in patients with very high inhibitor titer in whom combined therapy improved the response [19]. In some relapsed patients, retreatment with RTX was also effective [7]. The response rate was very high: in the review by Franchini [20] involving 65 treated patients, complete

Figure 1. Characteristics and structure of Rituximab chimeric antibody molecule and of CD20 transmembrane protein of B lymphocyte surface. Mouse (V): murine anti-CD20 variable sequence regions (in red wine); Human (C): human constant sequence regions (in blue)

48

www.slm-hematology.com


response was achieved in 57, partial response in 2, minimal response in 2, and no response in 5 only. Similar results were reported in the review by Garvey [19], who described 47 patients (44 of them were the same as Franchini reported): complete response was reached in 37, partial response in 8, complete/partial response in 1, no response in 1. Time to achieve a response ranged from 1 to 106 weeks (median 11.5 weeks), in the review by Franchini [20], and from 1 to 65 weeks (median 10 weeks) in that of Garvey [19], in comparison with the time taken to reach a response with corticosteroids (median 8 weeks), with corticosteroids plus cyclophosphamide (median 7 weeks), and with cyclophosphamide alone (median 30 weeks) [19]. Sperr et al [21] compared the efficacy of RTX in 43 patients with that of cyclophosphamide/prednisone in 44 patients. Complete remission with RTX was obtained in 78.6% of treated cases, median time to response was 8.3 weeks and, during follow-up, 66% of responders were still in complete remission after 2 years. Among the 44 patients treated with immunosuppressive drugs, complete remission was reached in 84.1% of cases, median time to response was 6.3 weeks, and the probability of continuous complete remission was 94% [21]. Rituximab treatment was safe in the majority of cases, but Onitilo et al [22] reported that two patients (75 and 73 years old, respectively, both with cancer) died from sepsis in remission. Moreover, they described a rebound elevation of FVIII activity (FVIII:C) after successful therapy with RTX [22]. This elevation may be associated with increased risk of venous thrombosis [23], and therefore periodic monitoring of FVIII:C levels is mandatory in the postremission follow-up. Furthermore, RTX has been rarely associated with cases of progressive multifocal encephalopathy, hepatitis B reactivation, and other viral infection in non-Hodgkin’s lymphoma (NHL) and in autoimmune diseases [19].

particular for patients for whom prednisone or cytotoxic drugs are unsuitable [19]. In the recently published international recommendations, RTX has been suggested as second-line therapy if first-line immunosuppressive therapy fails or is contraindicated [8]. Definition of the response to immune therapy, according to Baudo and de Cataldo [26] is as follows. Complete remission (CR) is defined as normal level of FVIII and no detectable inhibitor; partial remission (PR) is defined as an inhibitor titer , 10 BU/mL or a decrease of 50% if the baseline titer was ,10 BU/mL ; failure is defined as no change, or an inhibitor titer .10 BU/mL, or a decrease less than 50% of the baseline values. Complete and partial response must be combined with clinical improvement and disappearance of the bleeding tendency.

SEARCH STRATEGY We performed a search on MEDLINE without temporal limits, but with special attention to the last 10 years. The keywords we used were: acquired hemophilia, postpartum inhibitors, postpartum acquired hemophilia, Rituximab, immunosuppression, immunosuppressive therapy, FVIII inhibitors, and autoantibodies.

POSTPARTUM ACQUIRED HEMOPHILIA A With regard to postpartum FVIII inhibitors, they represent 7–21% of all cases of acquired hemophilia A, for an estimated incidence of about 1/350 000 deliveries [3]. Its pathogenesis remains unclear but, because autoantibodies develop more often in the postpartum period, one hypothesis may be that the mother was exposed to fetal FVIII during delivery. However, in subsequent pregnancies, a reappearance of the inhibitor is very rare, which would exclude an anamnestic response [1]. Until now, more than 100 cases of postpartum acquired hemophilia A have been described. Such a disease may occur after any pregnancy, but it is reported more frequently in primigravidas and most commonly between 1 and 4 months after delivery. Delgado et al [4], in their meta-analysis, reported that, in the majority of cases (about 60%) of pregnancyassociated inhibitors, these inhibitors spontaneously disappear after a mean time of 30 months, but in this review, the median inhibitor titer was not so high, that is 20 BU/mL. Moreover, they recorded only one death out of 34 reported cases [4]. In the analysis of Hauser et al [27] concerning 51 cases of postpartum acquired hemophilia A, a considerable variability emerged in the titer of inhibitors (range between 5 and 200 BU/mL) and the severity of bleeding symptoms. They reported three deaths but, in 76.5% of cases, the inhibitor disappeared spontaneously [27]. Solymoss also described a variable

In summary, the efficacy rate of RTX is very high: in fact, more than 90% of treated patients responded positively. However, patients with very high inhibitor titer may require additional courses of RTX and/or concomitant administration of other immunosuppressive drugs, such as cyclophosphamide. With these considerations in mind, Aggarwal et al [24] proposed an algorithm incorporating RTX, based on inhibitor titer. In patients with low inhibitor titer ,5 Bethesda units (BU)/mL) and minimal or absent bleeding, prednisone is the first-line treatment; in patients with inhibitor titer .5,30 BU/mL and serious bleeding, or without response to prednisone, RTX is added; in patients with inhibitor titer >30 BU/mL, combined therapy with prednisone, cyclophosphamide, and RTX has been proposed [24]. On the contrary, some authors considered that RTX may be a useful salvage treatment in patients who have failed first-line therapy [21, 25] or a useful alternative to existing treatments, in www.slm-hematology.com

49

JCD 2009; 1:(1). OCTOBER 2009


Journal of Coagulation Disorders

not appear to be correlated with the inhibitor titer or the bleeding severity. Presently, however, international recommendations advise against the use of cyclophosphamide and other alkylating agents in women with postpartum inhibitor because of the risk of infertility [8].

severity of bleeding symptoms: indeed, no fatal episodes occurred over a total of 80 bleeding episodes [28].

THERAPEUTIC APPROACH TO POSTPARTUM ACQUIRED HEMOPHILIA A The aim of management of postpartum acquired hemophilia A is the control of bleeding and the eradication of the inhibitors. In some cases, a ‘‘wait and see’’ approach can be justified by the absence of hemorrhagic symptoms and low inhibitor titer: in fact, a spontaneous disappearance of inhibitor is possible. Mannucci and Peyvandi [30] have recently suggested that, even though it may take several weeks after delivery for inhibitor disappearance, a conservative approach based upon the prompt treatment of bleeding is warranted in these women, whereas inhibitor eradication with immunosuppressive agents should be deferred and implemented only in the rare cases of persisting antibodies [30]. On the contrary, the recent international recommendations state that ‘‘all patients diagnosed with acquired hemophilia A receive immunosuppressive therapy immediately following diagnosis’’ [8].

Mazzucconi et al [31] used combined therapy with dexamethasone and HDIg in four cases of postpartum inhibitor: in three cases, a rapid decrease and subsequent eradication of inhibitor was obtained, without relevant side-effects. However, the question arises whether the immunosuppressive therapy can really change the natural course of the postpartum acquired hemophilia A: in fact, the clinical course of this disease is mostly benign, and a spontaneous disappearance of inhibitor is possible. Hauser et al [27] reported that the probability of remission is about 100%, while immunosuppressive therapy may induce only faster eradication of the inhibitor. On the contrary, in some cases, persistence of hightiter inhibitor, despite several lines of therapy, represents an actual risk of recurrent bleedings, so other therapeutic approaches are advisable that could possibly eradicate the autoantibodies without further toxicity.

However, treatment is mandatory when bleeding occurs, which can become a life-threatening event in the postpartum period. Treatment of bleeding episodes depends on the inhibitor titer, whether low, ,5 BU/ mL, or high, .5 BU/mL, and on the level of residual FVIII:C (undetectable or detectable). In the presence of low inhibitor titer and detectable residual FVIII:C, desmopressin (DDAVP) or human or recombinant FVIII concentrates can be used, whereas if FVIII:C is undetectable, human or recombinant FVIII concentrates at adequate dosages must be used. In case of high inhibitor levels, bypassing agents, such as APCC and rFVIIa, constitute the treatment of choice [1, 8, 16].

ROLE OF RITUXIMAB (RTX) IN POSTPARTUM ACQUIRED HEMOPHILIA A Rituximab is considered a safe therapeutic approach in acquired hemophilia A and can be a valid alternative for patients unable to receive cytotoxic immunosuppressive drugs: in particular, in young female patients, treatment with cytotoxic drugs can compromise fertility [19, 32]. On the contrary, RTX is well tolerated with few side-effects, especially in young people with acquired hemophilia A. For these reasons, RTX can also have a role in the therapeutic approach to postpartum acquired hemophilia A, especially when first-line therapy has failed, according to the recent international recommendations [8]. Until now, only a few cases of postpartum acquired hemophilia A treated with RTX have been reported, with promising results.

In rare cases that do not respond to these therapeutic agents, a reduction in inhibitor titer can be obtained by plasmapheresis with or without extracorporeal immunoadsorption [29]. The eradication of persistent autoantibodies can be obtained by the immunosuppressive therapeutic approach. Corticosteroids represent first-line therapy, while other immunosuppressant drugs such as cyclophosphamide or, less frequently, azathioprine, cyclosporine, and 6-mercaptopurine have been used as a second-line therapeutic approach or in combination with corticosteroids (mainly cyclophosphamide). Solymoss reported that, with steroids alone or in combination with cyclophosphamide, it is possible to obtain inhibitor eradication in 86% of cases [28]. Similar results were also observed by Baudo and de Cataldo [26]: 78% successful eradication when steroids with or without cyclophosphamide, azathioprine, or HDIg were given. The response to treatment did JCD 2009; 1:(1). OCTOBER 2009

From 2004 to the present, eight cases have been completely described in the literature, either as single case reports or as case/cases among series regarding patients with acquired inhibitors to FVIII associated with various conditions. In Table 1, the main characteristics of patients are described. Age at diagnosis ranged between 18 and 40 years (mean 29 years, median 27.5 years); three women were primigravidas, two have had previous pregnancies (one of them a first trimester spontaneous abortion), and in the other three, the number of pregnancies was not specified. In two patients, lupus anticoagulant (LA) and multiple sclerosis were associated conditions. The 50

www.slm-hematology.com


www.slm-hematology.com

51

25 Dedeken [35]

*From the last RTX infusion. {Associated condition: lupus anticoagulant. {Associated condition: multiple sclerosis. AZA, azathioprine; CR, complete remission; CYCLO, cyclophosphamide; HDIg, high-dose immunoglobulin; PDN, prednisone; n.d., not done; VCR, vincristine.

None 8+ 2 CR PDN 1 None 0.8 4

34

yes

None

None 24+ 2.6 CR PDN 1 None 0.5 4

n.d.

Dedeken{ [35]

no

24+

24+ n.d.

3 CR

CR CYCLO AZA

PDN HDIg 1

4 PDN CYCLO HDIg AZA

None 0.3

13.5 1.7

10 36 Dedeken [35]

no

40 Machado [34]

n.d.

Post-infusion headache, chills 25 Santoro [33]

yes

621

,1

PDN DXM HDIg CYCLO

4

PDN

CR

39

29+

None Bilateral limb amputation for compartment syndrome 12+ 36+ 48 43 CR CR None CYCLO 4 9 PDN HDIg AZA VCR CYCLO PDN HDIg ,1 ,1 470 3075 18 24 Maillard [32] Onitilo{ [22]

yes n.d.

n.d n.d. CR None ,1 16 n.d. 30

Inhibitor FVIII:C titer (BU/mL) (%) Case

Among the above-described cases of postpartum acquired hemophilia, four were treated with RTX as first-line therapy, associated with prednisone in three. In the other four cases, more than two previous therapeutic

Stasi [7]

COMMENTS AND DISCUSSION

Primi gravida (yes/no)

Behavior of CD19/20 peripheral positive B cells during RTX treatment was reported in the case described by Santoro et al [33]: the levels were detected at baseline (205 6 106/L), at the first week after stopping therapy (level 0/L), at the fifth month (0/L), at the ninth month (56 6 106/L) and at the twelfth month, when a complete recovery was recorded (212 6 106/L). There were no changes in the serum immunoglobulin.

Age (years)

Table 1. Features of Postpartum Acquired Hemophilia A patients

As far as bleeding symptoms at presentation or during follow-up were concerned, these varied from ecchymoses, easy bruising, and epistaxis to large muscle hematomas, abdominal shedding of blood, and hemorrhagic shock post delivery. In three cases, bleeding symptoms were successfully treated with rFVIIa, followed in one case by infusion of porcine FVIII.

None

Previous treatments

RTX doses (375 mg/m2/ week)

Complete remission, that is disappearance of inhibitor and complete recovery of FVIII:C, was reached in all cases, with variable times to response: in six patients, in whom these data were reported, the time to response ranged from 2 to 48 weeks (mean 23 weeks, median 21 weeks). No patient relapsed: duration of persistent CR (reported in seven patients) ranged from 8 to 36 months (mean 22.4 months, median 24 months). No relevant side-effects related to the drug were recorded, but in one patient [22] with LA, bilateral limb amputation, due to compartment syndrome, was performed. No infections occurred in any case.

4

Concurrent treatments

Time* to CR Response (weeks)

Duration of CR (months)

n.d

SE/AE

interval from delivery to diagnosis ranged between 0.6 and 34.4 weeks (mean 14.1 weeks, median 8.6 weeks); inhibitor titer (BU/mL) ranged from 1.7 to 3075 ( mean 525.2, median 13); in three patients, it was very high, whereas in the other five it was ,5 BU/mL in three and .5 and ,20 BU/mL in two; FVIII:C levels were ,1% in all but one patient, in whom it was detectable (13.5%). In four patients, RTX was given after the failure of two or more immunosuppressive treatments, whereas in the other four, no previous therapy had been given. Rituximab was administered according to a standard single weekly dose for 4 weeks in four patients, for 9 weeks in one patient, and only once in three patients [7, 22, 32–35]. In the patients reported by Stasi et al [7] and by Maillard et al [32], no concurrent therapy was given, whereas in those described by Onitilo et al [22], by Santoro et al [33], and by Machado et al [34], cyclophosphamide, prednisone, and azathioprine plus cyclophosphamide were associated with RTX, respectively; in the remaining three patients reported by Dedeken et al [35], in whom RTX was given as a single dose, at the onset of the disease, prednisone was concurrently administered and in one patient HDIgs were also given.

JCD 2009; 1:(1). OCTOBER 2009


Journal of Coagulation Disorders

the recovery of peripheral CD20-positive B cells began months after stopping therapy: in the case described by Santoro et al [34], it was completed by the 12th month. It seems to be advisable to monitor the behavior of peripheral CD20-positive B cells at baseline and then serially until a complete recovery is reached.

approaches had been given without success. In four patients, four doses of RTX were administered, while in the others, nine doses were given in one patient and only one dose in three. Inhibitor titer was very high in three patients and at low–intermediate levels in the other five. Concurrent treatments were given in all but two patients. All patients achieved a CR; the time to response was very short in the cases described by Dedeken et al [35] (2–3 weeks) and very long (39–48 weeks) in the others, who were also characterized by highest inhibitor titers. No relapses occurred during follow-up, which ranged from 8 to 36 months (Table 1).

In conclusion, RTX is a new drug in acquired hemophilia and, in particular, in the setting of postpartum cases. Available data are thus limited and may carry a bias in reporting only positive outcomes. However, it may be a useful alternative or addition to existing therapies. While bleeding symptoms usually disappeared early during the course of therapy with RTX, a CR occurred later after the beginning of therapy. Nevertheless, RTX can be considered a valid secondline therapy after the failure of corticosteroids, sparing the use of cytotoxic drugs.

In these cases, RTX seems to be very effective, but the response delay seems to be related to the highest inhibitor titer. Moreover, no relevant side-effects were recorded and, in spite of the suppression of CD20positive B lymphocytes, no infections occurred. It is interesting that the patient described by Santoro et al [33] experienced a clinical improvement just after the first infusion of RTX, as she never presented with further hemorrhagic events, while the inhibitor titer began a slow, but continuous decrease.

Disclosure: The authors have no funding disclosure to declare.

REFERENCES 1. Franchini M. Postpartum acquired factor VIII inhibitors. Am J Hematol. 2006;81:768–773. 2. Green D, Lechner K. A survey of 215 non-hemophilic patients with inhibitors to Factor VIII. Thromb Haemost. 1981;30;45:200–203. 3. Collins PW, Hirsch S, Baglin TP, et al. UK Haemophilia Centre Doctors’ Organisation. Acquired haemophilia A in the United Kingdom: a 2-year national surveillance study by the United Kingdom Haemophilia Centre Doctors’ Organisation. Blood. 2007;109:1870–1877. 4. Delgado J, Jimenez-Yuste V, Hernandez-Navarro F, Villar A. Acquired haemophilia: review and meta-analysis focused on therapy and prognostic factors. Br J Haematol. 2003;121:21–35. 5. Bossi P, Cabane J, Ninet J, et al. Acquired hemophilia due to factor VIII inhibitors in 34 patients. Am J Med. 1998;105:400–408. 6. Sallah S, Wan JY. Inhibitors against factor VIII in patients with cancer. Cancer. 2001;38:26–34. 7. Stasi R, Brunetti M, Stipa E, Amadori S. Selective B-cell depletion with rituximab for the treatment of patients with acquired hemophilia. Blood. 2004;103:4424–4428. 8. Huth-Ku¨hne A, Baudo F, Collins P, et al. International recommendations on the diagnosis and treatment of patients with acquired hemophilia A. Haematologica. 2009;94:566–575. 9. Crenier L, Ducobu J, des Grottes JM, Cerny J, Delaunoit C, Capel P. Low response to high-dose intravenous immunoglobulin in the treatment factor VIII inhibitor. Br J Haematol. 1996;95: 750–753. 10. Green D, Rademaker AW, Brie¨t E. A prospective, randomized trial of prednisone and cyclophosphamide in the treatment of patients with factor VIII autoantibodies. Thromb Haemost. 1993; 70:753–757. 11. Boggio LN, Green D. Acquired hemophilia. Rev Clin Exp Hematol. 2001;5:389–404. 12. Zakarija A, Green D. Acquired hemophilia: diagnosis and management. Curr Hematol Rep. 2002;1:27–33. 13. Barnett B, Kruse-Jarres R, Leissinger CA. Current management of acquired factor VIII inhibitors. Curr Opin Hematol. 2008;15: 451–455. 14. Abdallah A, Coghlan DW, Duncan EM, Chunilal SD, Lloyd JV. Rituximab-induced long-term remission in patients with refractory acquired hemophilia. J Thromb Haemost. 2005; 3:2589–2590. 15. Franchini M, Veneri D, Lippi G, Stenner R. The efficacy of rituximab in the treatment of inhibitor-associated hemostatic disorders. Thromb Haemost. 2006;96:119–125.

Although spontaneous disappearance of the autoantibody may occur in a large number of postpartum inhibitor cases, the risk of hemorrhagic events persists as long as inhibitor and very low/undetectable levels of FVIII:C are present. Therefore, in these cases, eradication of inhibitor is mandatory. However a ‘‘wait and see’’ strategy may be considered before starting eradication therapy in cases without significant bleeding symptoms and low inhibitor titer [20, 30]. When inhibitor eradication is decided upon, prednisone at standard dose (1 mg/kg/day for 4–6 weeks) should be considered as the first-line therapeutic approach. This treatment is effective and side-effects are very few. However, in patients resistant to corticosteroids who need further treatment, immunosuppressive cytotoxic drugs, such as cyclosphosphamide, are not advisable because of the risk of fertility failure [8, 19, 32, 34]. So, when the first-line approach with prednisone has failed, RTX should be suggested as effective and safe second-line therapy for postpartum inhibitor eradication [8]. In the eight patients described above, the response was reached in all cases, the time to response was largely variable (from 2 to 48 weeks, median 21 weeks), but the median time of persistent CR was 24 months, and no relapses occurred. As suggested by the international recommendations, a close follow-up after a sustained CR is mandatory, ‘‘monitoring of aPTT and FVIII:C levels monthly during the first 6 months, every 2–3 months up to 12 months and every 6 months during the second year and beyond, if possible’’ [8]. Moreover, account must be taken of the rebound rise in FVIII after therapy with RTX, which can occur months after inhibitor eradication, with a greater risk of thrombosis rather than of bleeding [22]. After treatment with RTX, JCD 2009; 1:(1). OCTOBER 2009

52

www.slm-hematology.com


16. Franchini M, Lippi G. Acquired factor VIII inhibitors. Blood. 2008;112:250–255. 17. Mathias M, Khair K, Hann I, Liesner R. Rituximab in the treatment of alloimmune factor VIII and IX antibodies in two children with severe haemophilia. Br J Haematol. 2004;125:366– 368. 18. Karwal MW, Schlueter AJ, Zenk DW, Davis R. Treatment of acquired factor VIII deficiency with rituximab (abstract). Blood. 2001;98:533a. 19. Garvey B. Rituximab in the treatment of autoimmune haematological disorders. Br J Haematol. 2008;141:149–169. 20. Franchini M. Rituximab in the treatment of adult acquired hemophilia A: a systematic review. Crit Rev Oncol Hematol. 2007; 63:47–52. 21. Sperr WR, Lechner K, Pabinger I. Rituximab for the treatment of acquired antibodies to factor VIII. Haematologica. 2007;92:66–71. 22. Onitilo AA, Skorupa A, Lal A, et al. Rituximab in the treatment of acquired factor VIII inhibitors. Thromb Haemost. 2006;96:84– 87. 23. Rosendaal FR. High levels of factor VIII and venous thrombosis. Thromb Haemost. 2000;83:1–2. 24. Aggarwal A, Grewal R, Green RJ, et al. Rituximab for autoimmune haemophilia: a proposed treatment algorithm. Haemophilia. 2005;11:13–19. 25. Collins PW. Treatment of acquired hemophilia A. J Thromb Haemost. 2007;5:893–900. 26. Baudo F, de Cataldo F; Italian Association of Haemophilia Centres (AICE). Register of acquired factor VIII inhibitors (RIIA). Acquired factor VIII inhibitors in pregnancy: data from the Italian Haemophilia Register relevant to clinical practice. Br J Obstet Gynaecol. 2003;110:311–314. 27. Hauser I, Schneider B, Lechner K. Post-partum factor VIII inhibitors. A review of the literature with special reference to the value of steroids and immunosuppressive treatment. Thromb Haemost. 1995;73:1–5. 28. Solymoss S. Postpartum acquired factor VIII inhibitors: results of a survey. Am J Hematol. 1998;59:1–4. 29. Jansen M, Schmaldienst S, Banyai S, et al. Treatment of coagulation inhibitors with extracorporeal immunoadsorption (Ig-Therasorb). Br J Haematol. 2001;112:91–97. 30. Mannucci PM, Peyvandi F. Autoimmune hemophilia at rescue. Haematologica. 2009;94:459–461. 31. Mazzucconi MG, Bizzoni L, Giorgi A, et al. Postpartum inhibitor to factor VIII: treatment with high-dose immunoglobulin and dexamethasone. Haemophilia. 2001;7:422–427. 32. Maillard H, Launay D, Hachulla E, et al. Rituximab in post partum acquired hemophilia. Am J Med. 2006;119:86–88. 33. Santoro C, Rago A, Biondo F, et al. Efficacy of rituximab treatment in postpartum acquired haemophilia A. Haemophilia. 2008;14:147–149. 34. Machado P, Raya JM, Martı´n T, Morabito L, Brito ML, Rodrı´guez-Martı´n JM. Successful response to rituximab in two cases of acquired haemophilia refractory to standard-therapy. Int J Hematol. 2008;87:545–549. 35. Dedeken L, St-Louis J, Demers C, Meilleur C, Rivard GE. Postpartum acquired haemophilia: a single centre experience with rituximab. Haemophilia. 2009; 15:1166–1168.

www.slm-hematology.com

53

JCD 2009; 1:(1). OCTOBER 2009


JOURNAL OF COAGULATION DISORDERS

REVIEW ARTICLE

Acquired von Willebrand Syndrome in Childhood and Adolescence Andrew M Will Affiliation: Royal Manchester Children’s Hospital, Manchester, UK Submission date: 2nd July 2009, Revision date: 29th August 2009, Acceptance date: 5th September 2009

A B S T R A C T Acquired von Willebrand syndrome (AVWS) is a rare condition usually seen in middle-aged and elderly patients. This review summarizes the clinical features, etiology, pathophysiology, classification, and management of 113 children and adolescents with AVWS. There were nine underlying conditions associated with the occurrence of AVWS in younger patients: patients taking valproic acid, congenital heart disease, renal tumors, glycogen storage disease type 1a, hypothyroidism, systemic lupus erythematosus, E/beta0 thalassemia, acute lymphoblastic leukemia, and one case following an Epstein–Barr infection. Diagnostic criteria included recent onset bleeding or bruising in a patient with no previous personal history of bleeding or family history of a bleeding disorder with laboratory tests suggestive of von Willebrand disease (VWD) or the presence of abnormal von Willebrand multimers. The pathophysiology and management of acute bleeding was dependent on the primary underlying disease. Successful treatment of the underlying disease was almost always associated with sustained correction of the AVWS. Keywords: acquired von Willebrand syndrome, children, adolescents, valproate, congenital heart disease, Wilms, hypothyroidism, systemic lupus erythematosus Correspondence: Ward 84, Royal Manchester Children’s Hospital, Oxford Road, Manchester M13 9WL, UK. e-mail: andrew.will@cmft.nhs.uk

(SLE), hypothyroidism, and with epileptics taking valproic acid (Table 1). The pathophysiology of the different types of AVWS in children and adolescents has a similar spectrum to that seen in older patients [3– 5] and is directly related to the underlying diagnosis: N Reduced synthesis of von Willebrand proteins (hypothyroidism); N Antibodies to von Willebrand proteins (SLE); N Mechanical loss of high-molecular-weight von Willebrand multimers under conditions of abnormal blood flow with high shear stress (CHD, HbE/ Beta0Thal); N Drug related (valproic acid); N Uncertain mechanism (Wilms’ tumor, GSD-1a).

INTRODUCTION Acquired von Willebrand syndrome (AVWS) is well described in adults [1], but literature concerning children and adolescents is rare. Registry data confirm that the majority of patients with AVWS have underlying lymphoproliferative and myeloproliferative diseases that are almost exclusively seen in the middleaged and elderly [2]. However, AVWS does affect younger patients, and this review includes over 100 cases of AVWS occurring in children and adolescents. The aim of this review is to define the epidemiology, pathophysiology, and management of AVWS in children and adolescents and to raise awareness of AVWS in younger patients. It is likely that AVWS is underdiagnosed in the younger age groups and, unless the possibility of AVWS is considered, the diagnosis may be delayed or missed altogether. If abnormal bleeding occurs in a child or adolescent with one of the at-risk, underlying diseases (Table 1), investigations for AVWS should be undertaken (see below). These should include VWF multimer analysis.

METHODS A search of the medical literature from 1968 to 2009 was carried out by interrogation of MEDLINE, Google, and Google Scholar in order to identify cases of AVWS occurring in children and adolescents under 18 years of age. In total, 113 cases of AVWS affecting children and adolescents are included in this review. Among this group of 113 individuals were a small number of patients with the inherited disorder GSD-1a, who were over

In younger patients, AVWS is most often associated with congenital heart disease, renal tumors, glycogen storage disease type 1a, systemic lupus erythematosus JCD 2009; 1:(1). OCTOBER 2009

55

www.slm-hematology.com


Journal of Coagulation Disorders

presentation and only had abnormal VWF multimer analyses. Densitometric analysis of von Willebrand multimers is therefore an essential part of the work-up of suspected AVWS, not just for subtyping of the VWD but also because, in some cases, this may be the only abnormality found in the patient.

Table 1. AVWS in Children and Adolescents grouped by Primary Diagnosis Primary diagnosis

Number of patients with AVWS

1. Taking valproic acid

38

2. Congenital heart disease

24

16 Septal defects (ASD, VSD or combined) 5 Patent ductus arteriosus

It is likely that the true incidence of AVWS in both children and adults is much higher than reported and, unless multimer analysis is routinely carried out in appropriate patients, AVWS will remain underdiagnosed.

2 Aortic stenosis (1/2 supravalvular) 1 Pulmonary stenosis 3. Renal tumors

15

4. Glycogen storage disease 1a

13*

5. Hypothyroidism

12

6. Systemic lupus erythematosus

5

7. HbE/beta0 thalassemia

4{

8. Acute lymphoblastic leukemia 9. Post-EBV infection

1

Total

13 Wilms’ tumor 2 Embryonal renal adenoma

In children and adolescents, the AVWS seen in association with SLE is immune mediated and, in many cases, it is possible to demonstrate a functional antibody against VWF [4, 6, 7]. In immune-mediated AVWS, the half-life (T1/2) of VWF is reduced, and the response to therapy with FVIII/VWF concentrates and DDAVP is variable and often short-lived. This suboptimal response to concentrates and DDAVP compared with that seen in the inherited forms of VWD is not unique to immune-mediated AVWS and is also seen in many of the other AVWS subtypes. In some, the T1/2 of VWF is reduced due to increased proteolysis, as occurs in cardiac structural abnormalities; in others, interference results from secreted proteins that significantly reduce the VWF T1/2, e.g., hyaluronic acid in Wilms’ tumor. In others, the mechanism causing the reduction in the normal lifespan of VWF is unknown. So in AVWS, where there is active bleeding, or where there is a significant risk of bleeding, and particularly where surgical interventions are being considered, extended pharmacokinetic studies of the response to different treatment modalities should be performed to assess the effectiveness of any proposed therapy and the likely duration of its response.

1 113

*At the time of diagnosis, some of the GSD-1a patients were over 17 years of age, but bleeding symptoms in these patients had been more marked during their childhood (see text). {Two other HbE/beta0 thalassemia patients aged 18 years and 21 years were also described but have not been included in the total above (see text).

17 years of age when investigated for AVWS. However, they had been symptomatic from a younger age and, in fact, had experienced more bleeding problems during childhood than after adolescence (see below).

DIAGNOSIS The diagnosis of AVWS should be considered in any young patient who presents with abnormal bleeding or bruising of recent onset in the absence of previous personal history or family history of a bleeding disorder, particularly in patients with one of the known underlying associated primary diseases listed in Table 1.

ASSOCIATED PRIMARY DISEASES Acquired von Willebrand Syndrome and Valproic Acid The effect of valproic acid on the coagulation system is complex. Epileptic patients taking valproic acid can have a variety of coagulation defects that sometimes coexist in the same patient; thrombocytopenia, platelet dysfunction, hypofibrinogenemia, reduced vitaminK dependent factors, factor XIII deficiency, and AVWS [8–10]. The cause of AVWS in patients taking valproic acid is unknown.

Diagnosis is not always straightforward. Although some cases can be diagnosed using routine coagulation testing and are found to have a prolonged activated partial thromboplastin time (APTT), reduced von Willebrand factor antigen (VWF:Ag), reduced von Willebrand activity in response to ristocetin (VWF:RCo), and a low factor VIII (FVIII), many cases have a normal APTT with a normal FVIII and, in some, even the VWF assays are within normal limits.

The reported incidence of AVWS in patients taking valproic acid is uncertain with the published studies reporting very different results. An early study found an incidence of 67% in a cohort of 30 epileptic children [11]. As pointed out in a later study that found no cases of AVWS in a group of 40 similar patients [10], the earlier study [11] had used an unusually high lower limit of normal for factor FVIII, VWF:Ag, and VWF:RCo at 70% for all three parameters, which had significantly

Tiede et al [6] retrospectively reviewed the diagnostic work-up of a cohort of 35 adult patients with AVWS. Using multiple testing procedures, they found that a combination of VWF:Ag ,50 IU/dL, VWF:RCo/Ag ratio ,0.7, and collagen binding (VWF:CB:Ag) ratio ,0.8 yielded a sensitivity of 86%. However, a substantial number of patients have normal routine test results at JCD 2009; 1:(1). OCTOBER 2009

56

www.slm-hematology.com


increased the number of apparently affected patients. However, three other reports are broadly in agreement as to the incidence of children taking valproate with laboratory tests compatible with a diagnosis of AVWS. Serdaroglu et al [12] found 6/29 children on valproate had AVWS with VWF:Ag ranging from 53% to 98% but with significantly reduced VWF:RCo varying from 11.5% to 39.5%. This is in keeping with two recent studies that reported AVWS in epileptic children on valproic acid as 6/23 [9] and 6/15 [8] respectively.

in VWF itself. The absorption by activated platelets and accelerated plasmin degradation of the structurally altered VWF multimers are the likely causes of the AVWS seen in association with some cases of CHD [14, 15]. It is probable that the conformational change in the VWF caused by its exposure to increased shear stress exposes an ADAMTS13 cleavage site, allowing ADAMTS13 to selectively remove high-molecularweight multimers [16]. The underlying cardiac lesions are listed in Table 1 [15–19]. Unlike in adult patients, where most cases of AVWS are associated with aortic stenosis, in younger patients, most (16/24) are associated with septal defects (ventral septal defect (VSD), atrial septal defect (ASD) or combined ASD/VSD) and with patent ductus arteriosus (PDA) (5/24). Only 2/24 had aortic stenosis, one of which was supravalvular. The other patient had pulmonary stenosis.

In none of these studies was the presence of AVWS associated with any bleeding tendency in the epileptic children. There was no relationship between valproate dosage or duration of therapy and the incidence of AVWS. Valproic acid therapy was not withdrawn from any of the patients because of the presence of AVWS. Although there was no correlation between the presence of AVWS and spontaneous bleeding, patients on long-term valproate therapy do bleed more during operative procedures [8]. So any operative procedure in a patient on valproic acid has to be carefully managed with a full preoperative assessment for all of the valproate-associated coagulation disorders listed above including AVWS. When AVWS is present, most authorities would advise against the use of DDAVP to cover invasive procedures because desmopressin can induce seizures and, instead, would administer VWF concentrates preoperatively [11, 12]. However, DDAVP is not inherently eleptogenic and is thought to cause fits secondary to hyponatremia. So, as long as these patients are managed appropriately to avoid significant hyponatremia, DDAVP could probably be used safely in these patients, as an alternative to the more expensive and potentially infectious plasma-derived concentrates [8].

The incidence of AVWS in CHD is uncertain but, in one prospective study, 6/49 (12.2%) children with CHD had laboratory tests compatible with AVWS [15] and, in a study of 12 infants with PDA, four patients had loss of high-molecular-weight multimers [17]. Gill et al [18] reported 12 consecutive patients with CHD who were referred preoperatively for assessment of abnormal coagulation tests. Seven out of 12 gave a history of mild bleeding problems. All 12 had loss of highmolecular-weight multimers and variable minor abnormalities of VWF antigen and function. In four patients reinvestigated postoperatively, the VWF multimers had returned to normal. However, despite successful corrective surgery and reversal of the VWF abnormalities, they still all had prolonged bleeding times, suggesting that the preoperative coagulation results and mild bleeding symptoms may not have been related to the reduction in high-molecular-weight VWF multimers The same abnormal VWF multimer pattern with loss of high-molecular-weight multimers was also reported in a 6-year-old child and a 17-year-old adolescent in a recent German paper [19].

Other Drugs associated with AVWS There have been reports of AVWS associated with other drugs including hydroxyethyl starch, ciprofloxacin, and griseofulvin [2]. As yet, these drugs have only been described as causing AVWS in adult patients and so are not directly relevant to this review. However, younger patients administered these products might similarly be at risk of developing AVWS. Withdrawal of the relevant drug reverses the bleeding diathesis and the abnormal laboratory findings of AVWS [13].

With an apparently significant number of children with CHD having AVWS, it is surprising that clinical bleeding problems related to AVWS in CHD patients are not seen more often in clinical practice. However, in none of the above studies was there any definite relationship between the abnormal laboratory findings and clinical bleeding. It would appear therefore that the loss of high-molecular-weight multimers in these patients does not in itself cause a clinically significant hemorrhagic diathesis

ACQUIRED VON WILLEBRAND SYNDROME IN CONGENITAL HEART DISEASE Some patients with congenital heart disease (CHD) have cardiac lesions that are associated with abnormal vasculature and localized high pressure flow, which together create the conditions necessary for high intraluminal shear stress. The localized high shear stress activates platelets and, when activated, these platelets selectively adsorb high-molecular-weight von Willebrand multimers and cause structural alterations www.slm-hematology.com

ACQUIRED VON WILLEBRAND SYNDROME IN RENAL TUMORS Wilms’ Tumor The cause of the AVWS seen in a small percentage of patients with Wilms’ tumor remains uncertain. No 57

JCD 2009; 1:(1). OCTOBER 2009


Journal of Coagulation Disorders

evidence for an autoantibody has been demonstrated in any of the six children investigated for the presence of an inhibitor [20–24]. Similarly, there has been no evidence to support adsorption onto tumor cells [21, 23, 24]. What is certain is that the coagulopathy disappears after chemotherapy or resection of the tumor [5]. This point was most dramatically demonstrated by the two patients described by Baxter et al [24] with severe intraoperative bleeding that had not been prevented by fresh frozen plasma (FFP), Humate-P, factor VIII concentrate, cryoprecipitate and, in one case, preoperative plasmapheresis, but which stopped immediately in both children as soon as the renal vessels were ligated.

Bracey et al [22] postulated that the high levels of hyaluronic acid secreted by some Wilms’ tumors may be the cause of the abnormal von Willebrand assays seen in these patients. The presence of high concentrations of hyaluronic acid can directly interfere with laboratory testing of VWF, producing falsely low VWF assay results [3]. This might explain why the very significantly reduced VWF results seen in the Wilms’ patients with AVWS are not often associated with a severe bleeding diathesis at presentation (see Table 2). However, interference with laboratory assays by high levels of hyaluronic acid is clearly not the only explanation for the low VWF levels. Six out of 13 of the patients with Wilms’ tumor AVWS had some evidence of abnormal bleeding at presentation, including Bracey’s own patient [22]. The two patients described by Baxter et al [24] had abnormal bleeding at presentation, and both experienced serious intraoperative hemorrhage. Both these patients were shown to have high levels of plasma hyaluronic acid at presentation.

The resolution of AVWS seen in all the reported cases following chemotherapy or surgical removal of the Wilms’ tumor suggests that some characteristic of the tumor itself is causing the AVWS. It is possible that abnormal vasculature and high blood flow through the tumor vessels could produce conditions of high shear stress with physical disruption of von Willebrand multimers, as is postulated to be the cause of AVWS seen in some cardiovascular diseases (see above). In these structural cardiac defects associated with AVWS, there is preferential loss of high-molecular-weight multimers producing a pattern of type II VWD [4]. However, multimer analysis carried out in five of the cases of Wilms’ tumor-associated AVWS demonstrated either type I or type III patterns, i.e., not type II, making mechanical destruction of VWF less likely as the cause of AVWS in Wilms’ patients [5].

It seems likely that Wilms’ tumors secrete a substance that interferes with the clinical efficacy of VWF. This might be excess hyaluronic acid. AVWS in Wilms’ tumor patients is uncommon and has been reported in 13 children with an age range of 4 months to 9 years (see Table 2). The incidence of AVWS in Wilms’ tumor patients has been studied in one prospective study in which 4/50 (8%) patients had laboratory evidence of AVWS [25] and one retro-

Table 2. Age, Factor VIII, and von Willebrand Antigen and Activity Testing in Patients with AVWS and Wilms’ Tumour (adapted from Michiels et al [5] and updated) Reference

Age

VIIIc

VWF:Ag

VWF:RCo

7 months

5%

,2%

NT

Bruising and gum bleeding

9 years

18%

,6%

28%

Hematuria

3 years

37%

,1%

16%

Epistaxis

Han [23]

4 months

6%

4%

18%

None

Coppes

4 years

10%

4%

8%

None

Norhona

Bleeding at presentation

[20] Scott [21] Bracey [22]

[25]

Jonge

2 years

10%

,1%

5%

None

4 years

27%

4%

46%

None

1 year

27%

76%

54%

None

NS

13%

25%

,12%

None

[26] Will

NS

47%

,12,5%

27%

3 years

17%

,20%

,20%

Bruising

None

2 years

13%

11%

28%

Epistaxis

7 months

14%

28%

12%

[30] Baxter [24] Hematuria Bleeding from venepuncture sites VIIIC, factor VIII activity in plasma; VWF:Ag, von Willebrand antigen; VWF:RCo, von Willebrand ristocetin cofactor activity.

JCD 2009; 1:(1). OCTOBER 2009

58

www.slm-hematology.com


spective study with 2/47 (4%) patients fulfilling the criteria for AVWS [26]. In view of the potential seriousness of bleeding from surgical procedures in these patients, suspected Wilms’ tumor patients should be screened for the presence of abnormal coagulation [27]; this is perhaps even more relevant in view of the recent report of an acquired factor VII deficiency in a Wilms’ tumor patient [28].

at the time of testing and diagnosis of AVWS, their bleeding symptoms had been present since early childhood [33, 35]. Investigations revealed a prolonged bleeding time in most patients [33] with normal APTT, FVIII but low levels of plasma VWF:Ag and VWF:RCo [33, 35]. Up to 60% of GSD-1a patients may have abnormal VWF on testing, and it is postulated that the accumulating glucose phosphates seen in GSD-1a might alter the glycosylation of glycoproteins such as VWF [35]. However, in the patients tested, there was no correlation between control of their underlying GSD-1a and the VWF abnormalities.

The most difficult practical problem is how best to manage a child presenting with the clinical and radiological features of a Wilms’ tumor who has laboratory evidence of AVWS. Patients with Wilms’ tumors have an excellent prognosis with a 90% survival for patients with localized disease and 70% survival for those presenting with metastatic disease [29]. On the other hand, in Wilms’ tumors complicated by AVWS, tumor resection [21, 24] and even biopsy [25, 30] are not infrequently associated with serious, potentially life-threatening, perioperative hemorrhage despite active preoperative intervention aimed at reducing the risk of bleeding. It is therefore essential to assess the improvement in laboratory parameters and the duration of this response following administration of FVIII/VWF concentrates and/or DDAVP prior to surgery. In patients who demonstrate a good response to concentrates or DDAVP, untoward bleeding during surgery is unlikely. However, a poor response makes the decision to operate more difficult and, knowing that the AVWS will respond to chemotherapy within a few weeks [22, 23, 25, 26, 30], it may be better to administer chemotherapy for some weeks prior to any operative intervention as long as the patient is monitored closely for tumor progression.

GSD-1a patients respond well to DDAVP at least initially [33, 34], but the VWF levels return to baseline after 4 h [33].

ACQUIRED VON WILLEBRAND SYNDROME IN HYPOTHYROIDISM The AVWS associated with hypothyroidism differs from the other forms AVWS in that there is a global reduction in the synthesis and release of vWF that is not associated with a reduced half-life because of either autoantibodies or secondary structural changes in VWF multimers [36]. This has been confirmed by studies of the VWF propeptide (VWF:AgII). During normal VWF:Ag production, VWF:AgII is synthesized in a 1:1 ratio with VWF:Ag. In AVWS associated with hypothyroidism, this ratio remains normal with a parallel reduction in the plasma levels of both VWF:AgII and VWF:Ag, thus confirming reduced synthesis and/or release of VWF antigens rather than the accelerated removal of VWF:Ag from circulation, which would in contrast greatly increase the VWF:AgII to VWF:Ag ratio [4]. This reduced production of VWF antigens may be caused by the non-specific action of thyroxine on protein synthesis via the stimulation of mRNA by triiodothyronine [3, 37].

Embryonal Adenoma AVWS has also been described in two pediatric patients with embryonal adenomas of the kidney. Both patients also had polycythemia at presentation. The AVWS and polycythemia disappeared after nephrectomy in one of the two patients and after removal of the adenoma from the affected kidney in the other [31].

However, this non-specific effect on mRNA protein synthesis cannot be the only factor involved. A subgroup of 11 patients with AVWS and hypothyroidism was investigated for evidence of reduction in other coagulation factors. Six out of 11 had normal levels of all other coagulation factors. Of the other five patients, two had minor reductions in factor V, three had low levels of factor IX, and one a reduced factor XI. Nevertheless, the only consistent abnormality was the effect on the synthesis of VWF which was present in all 11 patients [36]. As far as treatment is concerned, DDAVP causes an immediate correction of bleeding time and factor levels [36] and, in one case, led to an eightfold increase in FVIII, VWF:Ag, and VWF:RCo [38]. These observations suggest that, in part at least, the AVWS in hypothyroidism is due to reduced release of stored VWF.

ACQUIRED VON WILLEBRAND SYNDROME AND GLYCOGEN STORAGE DISEASE (GSD) AVWS associated with GSD was first reported in a family with GSD-1b [32]. However, as pointed out by the authors, this might have been due to coexistence and co-inheritance of GSD-1b and inheritable VWD. However, subsequent reports [33–35] have confirmed that AVWS can be associated with GSD-1a. Abnormal bleeding in GSD-1a usually presents in late infancy or early childhood as easy bruising and troublesome epistaxis, symptoms that generally improve with the onset of adolescence. Although most of the GSD-1a patients were more than 17 years of age www.slm-hematology.com

59

JCD 2009; 1:(1). OCTOBER 2009


Journal of Coagulation Disorders

Although a relatively common cause of AVWS in younger patients, hypothyroidism-associated AVWS only represented 2% of all patients registered in a combined registry of AVWS containing information on both adult and pediatric patients [2].

of VWF:Ag to platelets. However, in the majority of cases, the autoantibody is less specific, binding to VWF:Ag to produce large immune complexes that are rapidly cleared by the reticuloendothelial system [4]. Affected patients present with the typical bleeding symptoms seen in inherited VWD. Of the affected children and adolescents, four out of five had troublesome epistaxis, four out of five gum bleeding, and two out of three females had menorrhagia [48, 53]. In two out of five cases, the diagnosis of VWD predated the diagnosis of SLE by 5 months [53] and 12 months [49] respectively.

Laboratory diagnosis is usually straightforward. A recent review article [36] stated median values of VWF:Ag of 28 U/dL (range 4–45 U/dL), VWF:Ag activity 28.5 U/dL (range ,3–55 U/dL), and factor VIII 47 U/dL (range 9–74 U/dL). The laboratory results and patient characteristics of the affected children and adolescents are listed in Table 3.

Diagnosis of AVWS was straightforward with all five having prolonged APTT and very low levels of factor VIII and von Willebrand antigen and ristocetin cofactor activity (see Table 4). The subclassification of AVWS has been less well investigated. Most cases that have undergone multimer analysis in the literature have been type II with one case described as type I, one type 3, and one indeterminate type II/III [48].

The subtype of VWD seen in affected patients closely resembles type I VWD [3, 37]. Bleeding symptoms were as expected for type I VWD. Eight out of the 12 had significant bleeding; four menorrhagia, three epistaxis, two post-procedural hemorrhages, and one oral bleeding (Table 3). All patients responded to thyroxine replacement, although the plasma levels of FVIII and VWF:Ag may take 4 months to increase significantly and up to 7 months to correct the bleeding time. For patients with acute bleeding symptoms, DDAVP has an almost immediate effect in correcting the bleeding time and significantly increasing FVIII and VWF:Ag [47].

Subtyping of the AVWS in SLE may be important, particularly in patients presenting with acute bleeding. In the whole group of 13 adults and teenagers described in the literature, DDAVP was moderately effective in two patients with subtypes I and III but not in a patient with type II. A further five type II patients received VWF/FVIII concentrates but had poor levels of factor recovery post infusion. One type I patient did achieve a short-lived, low-level recovery following administration of concentrate. Two type II patients received high-dose intravenous immunoglobulin (Ig) with complete correction of their FVIII and VWF:Ag that lasted for a few weeks before returning to pretreatment levels [48]. In all cases, treatment with prednisolone at doses ranging from 1 to 2 mg/kg was curative, although one adult case required one dose of intravenous cyclophosphamide at 700 mg/m2 a month

ACQUIRED VON WILLEBRAND SYNDROME IN SYSTEMIC LUPUS ERYTHEMATOSUS AVWS in patients diagnosed with SLE makes up only 2% of all published cases of AVWS [2]. However, younger patients form a significant proportion of these with five out of 13 published cases occurring in patients aged between 13 and 17 years (Table 4) [48]. AVWS in SLE is caused by the production of autoantibodies to vWF. In some instances, these are specific autoantibodies that interfere with the binding

Table 3. Characteristics of Children and Adolescents with AVWS and Hypothyroidism with Response to Treatment (adapted from Michiels et al [5] and updated)

Reference

Sex

Age (years)

Pre-thyroxine Bleeding symptoms

VIIIc

Post-thyroxine

VWF:Ag

VWF:RCo

VIIIc

VWF:Ag

VWF:RCo

Dalton [39]

F

17

Oral

67%

45%

27%

122%

144%

120%

Smith [40] Coccia [41]

F F

13 15

None Menorrhagia

36% 70%

30% 32%

– 44%

100% 79%

86% 105%

144%

Blesing [42]

F

17

Menorrhagia

36%

36%

31%

120%

79%

71%

Bruggers (twins) [38]

F

13

Menorrhagia

41%

39%

30%

71%

51%

160%

90%

110%

F

13

None

33%

37%

36%

Viswanath [43]

F

12

Menorrhagia Epistaxis

39%

33%

27%

Concha [44]

F

13

Epistaxis Postop bleed

50%

37%

37%

80%

95%

F

11

Epistaxis

27%

86%

Galli-Tsinopoulou [45]

F F

7 10

None None

58% 69%

34% 42%

– –

76% 122%

92% 72%

– –

Setian [46]

M

9

Bled after tooth extraction

,1%

28%

6.25%

68%

All became normal

VIIIC, factor VIII activity in plasma; VWF:Ag, von Willebrand antigen; VWF:RCo, von Willebrand ristocetin cofactor activity.

JCD 2009; 1:(1). OCTOBER 2009

60

www.slm-hematology.com


Table 4. Characteristics and Steroid Response of Teenagers with AVWS and SLE (adapted from Michiels et al [5] and updated) Reference

Age (years)

APTT

VIIIc

VWF:Ag

Simone [49]

13 +

4%

NT

NT

VWF:RCo

Steroid response Cured

Gazengel [50]

16 +

8%

,5%

,3%

Cured

Yoshida [51]

12 +

11%

,10%

1.6%

Cured

Soff [52]

17 +

10%

,10%

,1%

Cured

Niiya [53]

16 +

8%

10%

10%

Cured

APTT, activated partial thromboplastin time; VIIIC, factor VIII activity in plasma; VWF:Ag, von Willebrand antigen; VWF:RCo, von Willebrand ristocetin cofactor activity; +, increased; NT, not tested.

after starting prednisolone before full correction of the laboratory tests was achieved [54].

The patient did not respond to DDAVP or FVIII/ VWF concentrates. However, despite the failure to detect any inhibitory antibodies against VWF, there was a sustained response to a 5-day course of intravenous immunoglobulins at a dose 0.44 g/kg/day and a year later the child remains symptom free.

ACQUIRED VON WILLEBRAND SYNDROME IN HEMOGLOBIN E/BETA0 THALASSEMIA Benson et al [55] described six patients aged from 14 to 21 years (four of whom were under 18 years of age) with hemoglobin E/beta0 thalassemia with reduced highmolecular-weight von Willebrand multimers. Three had prolonged bleeding times, two had mildly reduced VWF:RCo, and two mildly reduced VWF:Ag. All had normal FVIII levels. In four, the APTT was prolonged by 3.5–14.1 s, but all four also had low levels of FXII, which could have accounted for the prolongation of the APTT in these patients. The patients had a severe anemia with evidence of extramedullary hematopoiesis. Two out of six patients had epistaxis; the others had no symptoms of abnormal bleeding or bruising.

ACQUIRED VON WILLEBRAND SYNDROME FOLLOWING EPSTEIN–BARR VIRUS INFECTION Kinoshita and colleagues [57] described a 6-year-old girl who developed de novo petechiae and bruising 2 weeks after an Epstein–Barr virus (EBV) infection. She had a prolonged bleeding time, reduced FVIII, VWF:Ag, and VWF:RCo with a selective reduction in high-molecular-weight multimers. DDAVP produced a transient response, returning all parameters to normal after 30 min until 2 h after the infusion. Her AVWS resolved after 2 weeks and had not reoccurred 2 years later at the time of the report. The relationship to the prior EBV infection remains uncertain.

These patients appear to have a similar type of AVWS to that seen in patients with CHD, who also have a qualitative reduction in high-molecular-weight multimers but with little evidence of abnormal bleeding. It has been suggested that the AVWS in these hemoglobinopathy patients has a similar etiology and is caused by high shear stress related to a high cardiac output, which induces platelet activation with preferential adsorption of high-molecular-weight multimers and excess proteolysis of VWF [3, 5].

MANAGEMENT OF ACQUIRED VON WILLEBRAND SYNDROME IN CHILDREN AND ADOLESCENTS Acute Bleeding Other than immune-mediated AVWS, in which immunoglobulins are the initial treatment of choice, FVIII/VWF:Ag concentrates and DDAVP are the main treatment options for acute bleeding in AVWS [4, 13].

ACQUIRED VON WILLEBRAND SYNDROME AND ACUTE LYMPHOBLASTIC LEUKEMIA

Modern FVIII/VWF:Ag concentrates appear to carry little risk of transmission of infectious disease and are the first-choice treatment for most patients with AVWS. In one study, 80% of patients responded to concentrates using a mean starting dose of 43 IU/kg FVIII [6]. However, in many patients, the duration of response is short-lived. So it is important to monitor the response to therapy and anticipate the need to repeat the administration of concentrates more frequently than would be the case for patients with the inherited forms of VWD.

In a recent case report, Dorn et al [56] described a 10year-old girl in full remission and 3 months into her chemotherapy schedule for acute lymphoblastic leukemia (ALL), who developed AVWS associated with epistaxis, prolonged bleeding after venepunctures and marrow aspirations, and bleeding from her central venous line site. Investigation at presentation of ALL had demonstrated normal APTT, FVIII, VWF:Ag, and VWF:RCo but, at the time of the symptomatic bleeding, repeat investigation revealed a prolonged APTT, FVIII of 27%, VWF:Ag of 13%, and VWF:RCo of 13%. Multimer analysis demonstrated loss of large-molecular-weight multimers with loss of triplet structure. These findings were associated with a biclonal IgM gammopathy. www.slm-hematology.com

DDAVP is the first choice for acute bleeding in hypothyroidism with AVWS, and it can be expected to be effective in all cases. However, in AVWS complicating other conditions, only about 30% of patients 61

JCD 2009; 1:(1). OCTOBER 2009


Journal of Coagulation Disorders

respond adequately [4]. Kumar et al [13] suggest using the standard dose of 0.3 mg/kg. As with concentrates, the duration of response is often short-lived. DDAVP and concentrates can be used in combination.

valproate with another alternative antiepileptic would effectively reverse the AVWS seen in these patients.

Other treatments have been used successfully in small numbers of patients. These include plasma exchange, extracorporeal immunoadsorption [13], and recombinant factor VIIa (rVIIa) [58, 59]. The first two pose practical difficulties in pediatric patients, particularly in younger children; rVIIa can be readily administered to children of all ages.

Acute bleeding requiring immediate intervention has not been described in this patient group. In a prospective study that involved no specific therapeutic intervention, the presence of AVWS did not appear to increase the likelihood of operative bleeding [15]. The qualitative reduction in high-molecular-weight multimers was reversed by correction of the underlying cardiac defect. However, patients with bleeding symptoms or abnormal coagulation tests need to be fully assessed preoperatively for the presence of other, coexisting bleeding disorders.

AVWS and CHD

Recombinant Factor VIIa Use in AVWS Although an off-license use and so only really a ‘treatment of last resort’ and then only after assessment of any underlying thrombotic risk, recombinant factor VIIa (rVIIa) has been used successfully in two adult patients with AVWS. An elderly man with a monoclonal gammopathy of undetermined significance with AVWS developed life-threatening gastrointestinal hemorrhage and hematuria that failed to respond to DDAVP, von Willebrand factor-containing factor VIII concentrate, tranexamic acid, and immunoglobulin. Treatment with rVIIa as a 90 mg/kg bolus followed by a continuous infusion at 17.5 mg/kg/h in combination with tranexamic acid 4 g/day arrested the bleeding completely [58]. The other patient had AVWS associated with an IgG kappa immunoglobulin. He required thyroidectomy for a thyroid mass. Preoperative assessment included the evaluation of various possible interventions to prevent operative bleeding. These included DDAVP, immunoglobulins, and von Willebrand factor-containing factor VIII concentrate, none of which resulted in an adequate response. He received rVIIa at a dose of 90 mg/kg starting 30 min preoperatively with at total of 32 doses over 72 h. There was no excess operative bleeding [59]. rVIIa may be considered an option to control or avoid serious hemorrhage in patients unresponsive to other therapies.

AVWS and Wilms’ Tumor Although acute spontaneous bleeding was not a significant clinical problem in these patients, operative intervention by either biopsy or excision was associated with severe perioperative bleeding in a significant minority of patients; therefore, all Wilms’ tumor patients should be screened for the presence of AVWS at presentation before any operative procedures are undertaken. As discussed in the ‘Diagnosis’ section, it is important to assess the response to FVIII/VWF:Ag concentrates and DDAVP prior to surgery over an extended period of at least 4 h. However, none of the preparative regimens described in the literature gave reliable control of surgical bleeding in these patients. Furthermore, all the patients who received preoperative chemotherapy responded after an interval of 4– 6 weeks with normalization of coagulation tests and subsequently went on to have uncomplicated surgery. So, in those patients who do not demonstrate an adequate response to proposed preoperative, antiAVWS therapy, the option of treating preoperatively with a preliminary course of chemotherapy prior to surgery may be considered a safer approach than immediate excision or biopsy

Disease-specific Management The approach to the management of AVWS in children and adolescents is dependent on the associated underlying disease process and is summarized below.

In all cases of AVWS associated with Wilms’ tumor, surgical excision and chemotherapy reversed the coagulation abnormalities.

AVWS in Patients taking Valproic Acid

AVWS and Hypothyroidism

Acute bleeding related solely to the presence of AVWS does not seem to be a frequent occurrence. However, children on valproate may bleed excessively during invasive procedures. The effect of valproic acid on the coagulation system is complex, and patients may have several, sometimes coexisting, different abnormalities of their coagulation system. For patients with AVWS with bleeding or undergoing surgical procedures, the treatment of choice is probably VWF concentrates, as DDAVP may induce seizures. If considered appropriate, substitution of JCD 2009; 1:(1). OCTOBER 2009

Acute bleeding should be managed with DDAVP. In all patients, the AVWS can be expected to be reversed after several months of thyroxine therapy.

AVWS and SLE. Acute bleeds in the type II patients with SLE should respond to intravenous immunoglobulins (iv Ig). Type I and III patients may respond to iv Ig and, in some of these patients, DDAVP may also be partially effective. All the patients described in the literature were cured 62

www.slm-hematology.com


14. Mohri H. Acquired von Willebrand syndrome: features and management. Am J Hematol. 2006;81:616–623. 15. Arslan MT, Ozyurek R, Kavakli K, et al. Frequency of acquired von Willebrand disease in children with congenital heart disease. Acta Cardiol. 2007;62:403-408. 16. Tsai H-M. Deficiency of ADAMTS13 causes thrombotic thrombocytopenic purpura. Arterioscler Thromb Vasc Biol. 2003;23:388396. 17. Rauch R, Budde U, Koch A, Girisch M, Hofbeck M. Acquired von Willebrand syndrome in children with patent ductus arteriosus. Heart. 2002;88:87-88. 18. Gill LC, Wilson AD, Endres-Brooks J, Montgomery RR. Loss of the largest von Willebrand factor multimers from the plasma of patients with con genital cardiac defects. Blood. 1986;67:758–761. 19. Voges I, Burstein C, Budde U, Lenschow U. Acquired von Willebrand syndrome in two children with congenital heart defects and abnormal haemodynamics. Hamostaseologie. 2006;26: 345–348. 20. Norhona PA, Hruby MA, Maurer HS. Acquired von Willebrand disease in a patient with Wilms tumor. J Pediatr. 1979;95:997–999. 21. Scott JP, Montgomery RR, Tubergen DG, Hays T. Acquired von Willebrand’s disease in association with Wilms’ tumor: regression following treatment. Blood. 1981;58:665–669. 22. Bracey AW, Wu AHB, Aceves J, et al. Platelet dysfunction with Wilms’ tumor and hyaluronic acid. Am J Hematol. 1987;24:247– 257. 23. Han P, Lou J, Wong HB. Wilms’ tumour with acquired von Willebrand’s disease. Aust Paediatr J. 1987;23:253–255. 24. Baxter PA, Nuchtern JG, Guillerman RP, Mahoney DH. Acquired von Willebrand syndrome and Wilms tumor: not always benign. Pediatr Blood Cancer. 2009;52:392–394. 25. Coppes MJ, Zandvoort SWH, Sparling CR, Poon AO, Weitzman S, Blanchette VS. Acquired von Willebrand disease in Wilms’ tumour patients. J Clin Oncol. 1992;10:422–427. 26. Jonge Paerink-Stockshader AB, Dekker AB, Riseeuw-Appel I, Hahlen K. Acquired von Willebrand’s disease in children with Wilms tumour. Med Paediatr Oncol. 1992;10:422–427. 27. Blanchette V. Routine bleeding history and laboratory tests in children presenting with a renal mass. Paediatr Blood Cancer. 2009;52:392. 28. Granger J, Gidvani VK. Acquired factor VII deficiency associated with Wilms tumor. Pediatr Blood Cancer. 2009;52:314–315. 29. Pritchard-Jones K. Controversies and advances in the management of Wilms’ tumour. Arch Dis Child. 2002;87:241–244. 30. Will A. Paediatric acquired von Willebrand syndrome. Haemophilia. 2006;12:287–288. 31. Konety BR, Hord JD, Weiner ES, Schneck FX. Embryonal adenoma of the kidney associated with polycythaemia and von Willebrand disease. J Urol. 1998;160:2171–2174. 32. Heyne K, Hosenfeld D, Grote W, Schaub J. Glycogen storage disease type 1b: familial bleeding tendency. Eur J Pediatr. 1984; 143:7–9. 33. Marti GE, Rick ME, Sidbury J, Gralnick HR. DDACP infusion in five patients with type 1a glycogen storage disease and associated correction of prolonged bleeding times. Blood. 1986; 68:180–184. 34. Dellinger TM, Livingston HM, Holder R, Streckfus CF. Glycogen storage disease and von Willebrand’s disease implications for dental treatment: dental management of a pediatric patient. Special Care Dent. 1998;18:243–245. 35. Muhlhausen C, Schneppenheim R, Budde U, et al. Decreased plasma concentration of von Willlebrand factor antigen (VWF:Ag) in patients with glycogen storage disease type 1a. J Inherit Metab Dis. 2005;28:945–950. 36. Manfredi E, van Zaane B, Gerdes VEA, Brandjes DPM, Squizzato A. Hypothyroidism and acquired von Willebrand’s syndrome: a systematic review. Haemophilia. 2008;14:423–433. 37. Michiels JJ, Schroyens W, Bememan Z, van der Planken M. Acquired von Willebrand syndrome type I in hypothyroidism. Clin Appl Thromb Hemost. 2001;7:113–115.

with prednisolone alone except for one adult patient who needed a single dose of adjuvant cyclophosphamide as well as steroids [54].

CONCLUSION AVWS in children and adolescents is a rare clinical problem but is probably underdiagnosed. The spectrum of underlying conditions associated with AVWS in the pediatric population is different from that seen in adults. Full investigation of suspected cases should include assessment of von Willebrand multimers or the diagnosis may be missed. The approach to managing AVWS is dependent on the underlying disease process. Acute bleeding episodes usually respond to FVIII/ VWF:Ag concentrates, DDAVP or, in immunemediated AVWS, to infusion of immunoglobulins. In most cases, AVWS is cured by successful treatment of the underlying disorder. Abnormal perioperative hemorrhage is a particular risk in patients with Wilms’ tumor, and these patients should be fully screened for AVWS at presentation. Disclosure: The author has no financial interests to disclose related to the contents of this article.

REFERENCES 1. Maddox JM, Anderson JAM, Plews D, Ludlum CA. Management of acquired von Willebrand’s syndrome in a patient requiring major surgery. Haemophilia. 2005;11:633–637. 2. Federici A, Rand J, Bucciarelli P, et al. Acquired von Willebrand syndrome: data from the International Registry. Thromb Haemost. 2000;84:345–349. 3. Jakway JJ. Acquired von Willebrand’s disease. Hematol/Oncol Clinics North Am. 1992;6:1409–1419. 4. Mohri H. Acquired von Willebrand syndrome: features and management. Am J Hematol. 2006;81:616–623. 5. Michiels JJ, Budde U, van der Planken M, van Vliet HHDM, Schroyens W, Berneman Z. Acquired von Willebrand syndromes: clinical features, aetiology, pathophysiology, classification and management. Best Pract Res Clin Haematol. 2001;14:401–436. 6. Tiede A, Priesack J, Werwitzke S, et al. Diagnostic workup of patients with acquired von Willebrand syndrome: a retrospective single-centre cohort study. J Thromb Haemost. 2007;6:569–576. 7. Michiels J, Schroyens W, van der Planken M, Bememan Z. Acquired von Willebrand syndrome in systemic lupus erythematodes. Clin Appl Thromb Hemost. 2001;7:106–112. 8. Gerstner T, Teich M, Bell N, et al. Valproate-associated coagulopathies are frequent and variable in children. Epilepsia. 2006;47:1136–1143. 9. Koeniga S, Gerstnera T, Kellera A, Teicha M, Longina E, Dempfleb C-E. High incidence of valproate-induced coagulation disorders in children receiving valproic acid: a prospective study. Blood Coagul Fibrinolysis. 2008;19:375–382. 10. Eberl W, Budde U, Bentele K, et al. Acquired von Willebrand syndrome as side effect of valproic acid therapy in children is rare. Hamostaseologie. 2009;29:137–142. 11. Kreuz W, Linde R, Funk M, et al. Valproate therapy induces von Willebrand disease type I. Epilepsia. 1992;33:178–184. 12. Serdaroglu G, Tutuncuoglu S, Kavakli K, Tekgiil H. Coagulation abnormalities and acquired von Willebrand’s disease type I in children receiving valproic acid. J Child Neurol. 2002;17:41–43. 13. Kumar S, Pruthi RK, Nichols WL. Acquired von Willebrand disease. Mayo Clin Proc. 2002;77:181–187.

www.slm-hematology.com

63

JCD 2009; 1:(1). OCTOBER 2009


Journal of Coagulation Disorders

59. Smaradottir A, Bona R. A case of acquired von Willebrand syndrome successfully treated with recombinant factor VIIa during thyroidectomy. Thromb Haemost. 2004;92:666–667.

38. Bruggers CS, McElligot K, Rasllison ML. Acquired von Willebrand disease in twins with autoimmune hypothyroidism: response to desmopressin and l-thyroxine therapy. J Pediatr. 1994;1125:911–913. 39. Dalton RG, Savidge GF, Matthew KB, et al. Hypothyroidism as a cause of acquired von Willebrand disease. Lancet. 1987;1:1007– 1009. 40. Smith SR, Auger MJ. Hypothyroidism and von Willebrand’s disease. Lancet. 1987;1:1314–1317. 41. Coccia MR, Barnes HV. Hypothyroidism and von Willebrand’s disease. J Adolesc Health. 1991;12:152–154. 42. Blesing NE, Hambley H, McDonald GA. Acquired von Willebrand’s disease and hypothyroidism: report of a case presenting with menorrhagia. Postgrad Med J. 1990;66:474–476. 43. Viswanath AK, Mayo A, King D, Bevan JS. Willebrand’s disease secondary to autoimmune hypothyroidism. 23rd Joint Meeting of the British Endocrine Societies, Brighton, UK, 22–24 March 2004. Endocr Abstr. 2004;7:262. 44. Concha RN, Borzone MA, Castillo MN, Rossle AS, Quevedo IL. Willebrand disease as an unusual manifestation of primary hypothyroidism. Report of two cases. Rev Med Chile. 2005;133: 813–816. 45. Galli-Tsinopoulou A, Stylianou C, Papaioannou G, NousiaArvanitakis S. Acquired von Willebrand’s syndrome resulting from untreated hypothyroidism in two prepubertal girls. Haemophilia. 2006;12:687–689. 46. Setian N, Tanaka CM, Damiani D, Dichtchekenian V, Carneiro JD, D’Amico EA. Hypopituitarism, deficiency of factors V and VIII and von Willebrand factor: an uncommon association. J Pediatr Endocrinol Metab. 2002;15:331–333. 47. Erfurth EM, Ericsson UB, Egervall K, Lethagen SR. Effect of acute desmopressin and of long-term thyroxine replacement on haemostasis in hypothyroidism. Clin Endocrinol (Oxford). 1995; 42:373–378. 48. Michiels J, Berneman Z, Gadisseur A, et al. Immune-mediated etiology of acquired von Willebrand syndrome in systemic lupus erythematosus and in benign monoclonal gammopathy: therapeutic implications. Semin Thromb Hemost. 2006;32:577–588. 49. Simone JV, Comet JA. Abildgaard CF. Acquired von Willebrand syndrome in systemic lupus erythematosus. Blood. 1968;31:806– 811. 50. Gazengel G, Prieur AM, Buriot D, et al. Antibody-induced von Willebrand Syndrome. Am J Hematol. 1978;5:355–363. 51. Yoshida H, Aria K, Wakashin M. Development of acquired von Willebrand disease after mixed connective tissue disease. Am J Med. 1988;85:445–446. 52. Soff GA, Green D. Autoantibody to von Willebrand factor in systemic lupus erythematosus. J Lab Clin Med. 1993;121:424–430. 53. Niiya M, Niiya K, Takazawa Y, et al. Acquired type 3-like von Willebrand syndrome preceded by full-blown systemic lupus erythematodes. Blood Coagul Fibrinolysis. 2002;13:361–363. 54. Viallard J-F, Pelligrin JL, Vergnes C, et al. Three cases of acquired von Willebrand disease associated with systemic lupus erythematosus. Br J Haematol. 1999;105:532–537. 55. Benson PJ, Peterson LC, Hasgawa DK, Smith CM II. Abnormality of von Willebrand factor in patients with haemoglobin E-Beta0 thalassaemia. Am J Clin Pathol. 1990;93:395–399. 56. Dorn I, Budde U, Fruhwald MC, Poppelmann M, Nowak-Gottl U. Acquired von Willebrand syndrome in a 10-year-old girl with acute lymphoblastic leukaemia. BMJ Case Reports. 2009 (doi:10.1136/bcr.04.2009.1816). 57. Kinoshita S, Yoshioka K, Kasahara M, Takamiya O. Acquired von Willebrand’s disease after Epstein–Barr virus infection. J Pediatr. 1991;119:595–598. 58. Friederich PW, Wever PC, Brie E, Doorenbos CJ, Levi M. Successful treatment with recombinant factor VIIa of therapyresistant severe bleeding in a patient with acquired von Willebrand disease. Am J Hematol. 2001;66:292–294.

JCD 2009; 1:(1). OCTOBER 2009

64

www.slm-hematology.com


JOURNAL OF COAGULATION DISORDERS

REVIEW ARTICLE

Cardiac Surgery and Percutaneous Coronary Interventions in Patients with Hemophilia B: an Overview of Published Reported Cases Elizabeth F Krakow1,2, Ruslan Ganchev3 and Julia AM Anderson1,3 Affiliations: Departments of 1Medicine and 2Pathology and Molecular Medicine, Faculty of Health Sciences, McMaster University, Hamilton, Ontario, Canada and 3Department of Haematology, Royal Infirmary of Edinburgh, Edinburgh, UK Submission date: 27th July 2009, Revision date: 1st September 2009, Acceptance date: 7th September 2009

A B S T R A C T With an ageing hemophilia population and decreasing mortality from transfusion-transmitted viruses, it is likely that percutaneous coronary interventions and surgery for coronary artery disease and valvular heart disease will become more commonplace. Eighteen reports of cardiac surgery and coronary interventions in patients with hemophilia B are published, and although limited in number and heterogeneous in nature, they represent the available published experience of this complex and evolving field. Cardiac surgery appears to be feasible and safe in individuals with all severities of hemophilia B, but the optimal factor replacement schedule remains uncertain, and there is a lack of consensus on the minimal target factor IX (FIX) level to attain peri- and postoperatively, in addition to the duration of replacement therapy, the use of antifibrinolytic agents, and the use of antithrombotic therapies following surgery. Close cooperation and good communication between all members of a multidisciplinary team are necessary for successful outcome, and an individualized written protocol is highly recommended. For individuals with hemophilia B undergoing coronary intervention, the bleeding risk for patients with different severities of hemophilia B requiring dual antiplatelet therapies after angioplasty and stent insertion is unclear, and the use of antithrombotic therapies pre-, peri- and postprocedure remains unknown. The aim of this overview is to detail key aspects of the management of patients with hemophilia B undergoing cardiac surgery and percutaneous coronary interventions to enable informed decision-making for individual patients in the future and to assist in the development of future management recommendations. Keywords: percutaneous coronary intervention, cardiac surgery, hemophilia B Correspondence: Julia AM Anderson, Department of Clinical and Laboratory Haematology, Royal Infirmary of Edinburgh, Little France Crescent, Edinburgh EH16 4SA, UK. Tel: +44-(0)131-242-6866; fax: +44-(0)131-242-6812; e-mail: jander@mcmaster.ca

severity. Practical issues arise regarding the choice of arterial approach, in addition to the optimal type of stent, bare-metal or drug-eluting, to employ in patients with hemophilia [2, 3].

INTRODUCTION Over the past decades, there have been several important advances in hemophilia care, including the improved safety of plasma-derived factor concentrates, treatment of human immunodeficiency virus (HIV) and hepatitis C, and the use of prophylaxis regimens to limit the development of joint arthropathy. For many men with hemophilia, there is now an opportunity for normal life expectancy [1]. Consequently, symptomatic coronary artery disease and consequences of rheumatic heart disease are likely to become more prevalent in this patient population. There have also been many advances in interventional cardiology and cardiac surgery that introduce new dilemmas for clinical decision-making. Such issues include the optimal antithrombotic regimen prior to, during, and following percutaneous coronary intervention (PCI) in patients with congenital bleeding diatheses of varying JCD 2009; 1:(1). OCTOBER 2009

For individuals with hemophilia undergoing cardiac surgery, the optimal factor levels to attain pre-, periand postoperatively are unclear, and the duration of factor replacement therapy is uncertain. In addition, the relative advantages of ‘off-pump’ surgery for cardiac revascularization in patients with hemophilia are unclear, and the optimal antithrombotic regimens immediately following cardiac surgery, and longer term, in persons with varying severity of factor deficiency are unknown. As it is anticipated that hemophilia carers will have more patients undergoing invasive cardiac procedures and cardiac surgery in the near future, this review aims to provide an overview of the current literature 65

www.slm-hematology.com


Journal of Coagulation Disorders

and recombinant activated factor VII, most of whom had inhibitors, appeared to be particularly susceptible to ischemic arterial events [14]. Similarly, in a review of 14 patients with acute coronary ischemic events and ischemic strokes in patients with hemophilia B, 50% of arterial thrombotic events occurred during or after infusion with PCCs, FEIBA, or cryoprecipitate supernatant [15]. In addition, the atherogenic effects of highly active antiretroviral HIV therapy might partially reduce the protective effect of hemophilia on ischemic heart disease [10].

regarding the management of patients with hemophilia B undergoing such operations and procedures to enable informed decision-making for individual patients in the future.

THE PREVALENCE OF ISCHEMIC HEART DISEASE IN PATIENTS WITH HEMOPHILIA B Studies of age-adjusted and age-matched mortality rates in hemophilia populations from Europe [4–6] and the United States (US) [7] have demonstrated lower rates of death from ischemic heart disease in individuals with hemophilia A and B compared with the general male population. In the United Kingdom Haemophilia Doctors’ Organisation (UKHCDO) study, spanning 23 years, of 6018 individuals who were not infected with HIV, the age-specific death rate from ischemic heart disease was 62% of that in the general population with similar mortality rates in severe and moderate/mild hemophilia groups [6]. Although the number of ischemic heart disease deaths in the severe hemophilia group was small and not statistically significant, this rate is comparable to rates of mortality from ischemic heart disease among carriers of hemophilia [8]. Although carriers of hemophilia represent a different population from hemophilia patients, these results might suggest that the protective effect against ischemic heart disease mortality is not associated with absolute factor levels. A detailed study of hemophilia patients in the Netherlands demonstrated that the reduction in ischemic heart disease was not attributable to differences in cardiovascular risk factors such as hypertension and hyperlipidemia, and therefore the explanation for the protective effect remains unclear but might relate to differences in the development of atherosclerosis or arterial thrombosis in hemophilia patients [9].

It is therefore likely that individuals with hemophilia have risk factors such as infusions of factor concentrates, HIV infection, and treatment with highly active antiretroviral therapy (HAART) that may predispose them to ischemic heart disease as life expectancy increases, although highly purified plasma-derived FIX concentrates and recombinant FIX concentrates are less thrombogenic than previously used PCCs.

SEARCH STRATEGY MEDLINE was searched (1966 to June 2009, Week 1) using the following MeSH terms: haemophilia, hemophilia, hemophilia B, hemorrhagic disorders, bleeding diathesis, cardiac surgery, thoracic surgery, coronary restenosis, myocardial infarction, angioplasty, transluminal, percutaneous coronary intervention, or coronary disease. Of 228 articles, 214 were excluded based on title and abstract review, and 14 articles were retrieved for full text review. Reference lists of relevant articles were manually searched and yielded four additional articles. The articles included 13 case reports, two case series, and three letters. Case reports written in non-English languages were kindly translated with the assistance of colleagues within our departments.

Despite the observed protective effect, the age-specific prevalence of ischemic heart disease is not trivial and was estimated at 15.2% in those 60 years or older among a US cohort from the mid-1990s [10]. This study additionally demonstrated that people diagnosed with ischemic heart disease were more likely to have hemophilia B, with a prevalence of 2.4% compared with a prevalence of 1.1% in those with hemophilia A (P,0.05). The explanation for this observed difference may be the increased thrombogenicity of factor IX (FIX) concentrates, but could also relate to differences in the development of thrombosis in hemophilia B patients.

PERCUTANEOUS CORONARY INTERVENTIONS IN PATIENTS WITH HEMOPHILIA B There are three reports of PCI in patients with hemophilia B [16–18]. Helft and colleagues [16] outline the management of a 58-year-old patient with mild hemophilia B (basal FIX level 9%) diagnosed with an acute coronary syndrome (ACS) with angiographic evidence of a single severe stenosis of a right dominant coronary artery. The patient had known risk factors for hypercholesterolemia and a smoking habit. FIX was not stated as being given prior to the onset of this individual’s ACS. The patient was treated with 5000 IU plasma-derived FIX (Mononine) 4 h in advance of angioplasty (femoral route) and stent insertion (micro A.V.E.) under 10 000 IU heparin cover [16]. Aiming for a pre-procedure FIX level of 70%, the procedure was uncomplicated in terms of bleeding and thrombosis. A further 3000 IU bolus infusion of plasma-derived FIX was given prior to the sheath removal. No antiplatelet agents were administered. No bleeding and no acute or

There are numerous case reports of acute coronary syndromes in patients with hemophilia and von Willebrand’s disease possibly related to high levels of infused factors or the use of activated factor concentrates [11–13]. A review of 42 reported cases of myocardial infarcts and ischemic strokes in patients with hemophilia A suggested that patients treated with prothrombotic activated bypassing agents such as FEIBA, prothrombin complex concentrates (PCCs), JCD 2009; 1:(1). OCTOBER 2009

66

www.slm-hematology.com


FIX levels varying from 17% to 24% [33]. The type of cardiac surgery (valve surgery, coronary revascularization, or repair of congenital defect), hemophilia B phenotype, FIX replacement strategy (bolus, continuous infusion, or both) and product used, actual and target FIX levels, and documentation of the use of antifibrinolytic therapy are summarized in Table 1.

subacute thrombosis were reported 3 months after the procedure. The other reports relate to patients with severe hemophilia B. The report by Bovenzi and colleagues [17] describes a 73-year-old patient with a FIX level of less than 1% requiring PCI with stent insertion following a diagnosis of unstable angina. FIX was not stated as being given prior to the onset of this individual’s ACS. The individual had type 2 diabetes mellitus and hypertension. The authors state that ‘the standard pre-treatment antithrombotic regimen with ticlopidine, acetylsalicylic acid and heparin was avoided’, and the patient was administered high-purity FIX (AIMAFIX, Castelvecchio Pascoli-LU, Italy) less than 1 h prior to the procedure. The authors state that, following the administration of FIX, 5000 IU heparin was given prior to proceeding with the catheterization procedure. The dose of FIX administered was 30 IU/kg, but FIX levels pre- and post-procedure were not stated in the report. The stented vessel became occluded immediately after stent deployment and required treatment with a glycoprotein IIb/IIIa inhibitor, abciximab at a standard dosage of 0.25 mg/kg bolus followed by 0.125 mg/kg/min infusion in addition to a further 2000 IU heparin, with successful recanalization and without any clinically relevant bleeding. Clopidogrel was prescribed for 1 month post-procedure and the patient was well at his 6-month follow-up.

The reports date from 1979 until 2009 and represent a highly heterogeneous group of case reports, providing details of patients with all severities of hemophilia B, including one description of an atrial septal defect repair in a carrier [32].

Operative Outcome and Complications In general, outcomes across reported cases and series were generally good, although this may reflect reporting bias. The majority of reports documented the extubation of the patient later on the day of operation or on the first postoperative day (POD), with removal of chest tubes and drains on or around the first to third PODs. Regarding reported hemorrhagic complications, one hemophilia B carrier underwent atrial septal defect closure with ImmunineH by bolus dosing over 11 days but required drainage of a pericardial effusion later [32]. The authors do not suggest that the pericardial effusion was a surgical bleeding complication. Another man underwent coronary artery bypass grafting (CABG) with bolus dosing of BenefixH. He suffered an aortic dissection with hemorrhage into the intima on the second POD, at a time when his FIX level was 37% [26].

The third report details the management of a 31-yearold individual with severe hemophilia B, factor IX level less than 1%, who developed interferon-induced thyrotoxicosis during treatment with interferon therapy, with subsequent chest pains and palpitations. No risk factors for ischemic heart disease were present. The report mentions the ‘use of Konyne FIX intermittently’, but it is unclear whether FIX replacement therapy was administered around the time of development of the chest pains. The patient underwent coronary angiography confirming near complete occlusion of the left anterior descending coronary artery with widespread vessel irregularities. He underwent PCI and stent insertion with ‘full factor replacement and heparinization’ (no further details provided), and was given aspirin 100 mg daily and ticlopidine 100 mg twice daily for 1 month post-procedure with no reported bleeding or thrombotic complications [18].

Perioperative Hemostatic Strategies Some of the earlier reports use perioperative hemostatic strategies that reflect attempts to minimize the side-effects of factor concentrates of lesser purity such as disseminated intravascular coagulation and thromboembolism. These strategies include the use of plasmapheresis with fresh frozen plasma (FFP) replacement preoperatively to increment the FIX level prior to cardiac surgery and avoid volume overload [20]. Other therapeutic strategies include the use of FFP alone to increment FIX levels [32], and also the combined use of FFP and PCCs [19, 30, 31]. PCCs have also been used alone to increment FIX prior to cardiac surgery [22, 23]. There are three reports of cardiac surgery using continuous infusion (CI) of plasma-derived FIX [25, 27, 28], one report of cardiac surgery using bolus infusions of recombinant FIX [26], and one report of the successful use of CI of recombinant FIX [29].

CARDIAC SURGERY IN PATIENTS WITH HEMOPHILIA B Nine case reports describe individual revascularization procedures [19–27], two case reports describe the management of individuals with rheumatic or congenital valvular heart disease undergoing valve replacement [28, 29], and three case reports describe the repair of congenital defects [30–32]; a single case series does not specify the nature of the cardiac surgery in three individuals with hemophilia B with www.slm-hematology.com

Pre- and Postoperative FIX Levels In this group of letters, case reports, and case series, the collective authors variably stated the target FIX levels and actual FIX levels attained pre- and postoperatively (see Table 1). 67

JCD 2009; 1:(1). OCTOBER 2009


JCD 2009; 1:(1). OCTOBER 2009

Congenital

bicuspid valve

Krakow, 2009

valve

AVR, tissue

68

smoker NIDDM

CABG

pd-FIX

Bolus

Bolus

Bolus

Bolus

Bolus

Bolus + CI

Target not stated

Stepwise reduction POD 3–7

Actual 100% POD2

Target not stated

Target not stated

Actual 60%

Target not stated

Actual not stated

Actual not stated Target not stated Actual 90%

Target 50% for 7 days

Target 20–40% Actual 44% POD1

Actual 12–26%

Target not stated

,40–50% POD2–3 .20% second postoperative week

Target 100%

Target 60% Actual 84%

Actual 51%

Target not given

Actual 84%

Target 80–90% Actual ,80–90% POD1

Actual 92%

Target 100%

Target 90–100% POD1 Actual 72–88% POD1

Actual 82%

Target 50% POD1–3

Aprotinin

Not stated

Not stated

Not stated

Not stated

Not stated

TXA

Not stated

Postoperative antithrombotic therapy N/S

Hematoma right thigh at saphenous vein graft site, settled conservatively

Discharge on POD13

Factor IX doses documented through POD10

Postoperative antithrombotic therapy N/S

Discharge on POD7 Postoperative antithrombotic therapy N/S

Postoperative antithrombotic therapy N/S

Discharge POD13

Good hemostasis documented

Postoperative bilateral pneumothoraces

Postoperative antithrombotic therapy N/S

FFP intraoperatively

Discharge POD7

ASA 81 mg daily

Heparin 5000 IU POD1 until POD7 FIX replacement until POD10

INR 1.5–2.0

acenocoumarol postoperative

cardiac tamponade POD4, ARF

Low-dose LMWH initially postoperative

Comments

Discharge on POD6

2.4%

Bebulin-TIM4IMMUNO

pd-FIX

PCC Konyne

PCC Konyne

FFP, PEx

PCC Konyne

FFP

BeneFix

recFIX

Actual 88%

Target 80%

Bolus + CI

Antifibrinolytic therapy

TRANSIENT episode atrial fibrillation POD3

71 years

2%

4.9%

pd-FIX

Alphanine

Post-FIX (%)

CABG 6 6

58 years

52 years

12%

5%

2%

8%

14%

Pre-FIX (%)

Bolus/CI

Palanzo, 1995

CABG 6 3

Dimitrova, 1995

CABG 6 4

BP

74 years

Wilson, 1991

Scharfman, 1993

59 years

smoker

CABG 6 4

Raish, 1985

CABG 6 4

40 years

61 years

AVR, MVR, mechanical

Tourbaf, 1979

25 years

Rheumatic

valve

Age Risk Basal FIX FIX factors level (%) replacement

Type of cardiac surgery Thankachen, 2007

Table 1. Summary of Data abstracted from Published Case Reports and Case Series relating to Patients with Hemophilia B undergoing Cardiac Surgery

Journal of Coagulation Disorders

www.slm-hematology.com


www.slm-hematology.com

69

JCD 2009; 1:(1). OCTOBER 2009

18 years F

7 years

16 years

FIX infusions

IDDM

52 years

80 years

lipids

BP

64 years

NIDDM

PCC preoperatively, FFP, PCC postoperatively

PCC Proplex POD2

FFP pre and POD1

BETAFACT

pd-FIX

BeneFix

recFIX

pd-FIX

32% pd-FIX (carrier) Immunine

,1%

2%

,1%

8%

6%

Mononine

Age Risk Basal FIX FIX factors level (%) replacement

Bolus

Bolus

Bolus

Bolus + CI

Bolus

Bolus + CI

Bolus/CI

Target .25% for 7 days

Actual not stated

POD7–14, 50–60%

Target POD1–7, 100–120%

Actual 37–120%

Target 50–150%

Target and actual not stated

Actual 38%

Post-FIX (%)

Target 100% Actual not stated

Actual 98%

Target 100%

Target not stated Actual not stated

Actual ,60–70% (graph)

Target .50% for 20 days

‘Estimated’ 64% Actual ,40% (graph)

Target not stated

Actual not stated

Target 100–120%

Actual 110%

Target 100%

Actual 71%

Target not stated

Actual not stated

Pre-FIX (%)

Not stated

Not stated

Not stated

Not stated

1 g/h infusion

10 g load

e-ACA

Not stated

(total 600 000 KIU)

Antifibrinolytic therapy Comments

Duration of therapy 11+4 days Postoperative antithrombotic therapy N/S

Postoperative antithrombotic therapy N/S

Postoperative antithrombotic therapy N/S

Discharged POD11

enoxaparin 4000 IU daily until discharge POD14

Off-pump, ASA 160 mg daily preoperatively,

emergency reoperation, death

Type 1 aortic dissection POD2, level 37–50%,

FIX given to POD19

ASA 100 mg daily

Postoperative antithrombotic therapy N/S

AVR, aortic valve replacement; MVR, mitral valve replacement; pd-FIX, plasma-derived factor IX; CI, continuous infusion; POD, postoperative day; LMWH, low-molecular-weight heparin; ARF, acute renal failure; INR, international normalized ratio; recFIX, recombinant factor IX; TXA, tranexamic acid; ASA, aspirin; FFP, fresh frozen plasma; PCC, prothrombin complex concentrate; PEx, plasma exchange; IDDM, insulin-dependent diabetes; BP, hypertension; Lipids, hyperlipidemia; e-ACA, e-aminocaproic acid.

closure ASD; drainage of pericardial effusion

MacKinlay 2000

repair VSD, aortic regurgitation

Mazzucco, 1986

correction transposition of arteries, VSD repair, PS correction

Roskos, 1983

CABG

Grandmougin, 2003

CABG 6 3

Donahue, 1999

CABG 6 4

Bukowski, 1996

Type of cardiac surgery

Table 1. Continued


Journal of Coagulation Disorders

Preoperative target FIX levels or actual FIX levels were clearly stated in 13 out of 14 reports, although only six of the articles stated both FIX levels. One group aimed for preoperative target FIX levels of 100– 120%, six groups chose a target FIX level of 100%, and two groups chose target FIX levels of 80–100%. One group chose a target FIX level of 60%, although the actual FIX level attained preoperatively was 84%. Five reports did not state the target preoperative FIX levels.

with aspirin 160 mg preoperatively, and this was recommenced postoperatively in addition to lowmolecular-weight heparin, enoxaparin 4000 IU, daily until discharge on POD14. Both reports of valve replacement described individuals with a mild hemophilia B phenotype. As cardioembolic complications are feared in the immediate postoperative period, it was of interest to note the antithrombotic strategies used by the authors. One group [29] chose to use a bioprosthetic valve to avoid the need for anticoagulation with a vitamin K antagonist long term postoperatively and chose to administer low-dose unfractionated heparin 5000 IU twice daily from POD1–9 (for venous thromboprophylaxis), and low-dose aspirin (81 mg daily) was commenced on POD5 and continued long term. This group had administered a CI of recombinant FIX postoperatively until day 5 and then bolus infusions of FIX until POD14.

In terms of the actual preoperative FIX levels, six of the groups stated FIX levels of between 80% and 99%, and one group with a stated target of 100% achieved a higher preoperative FIX level of 110%. The lowest FIX level preoperatively was 51% in the report by Raish and colleagues who used FFP as the means to replace FIX and reported ‘‘good hemostasis throughout the patient’s operation’’ [20]. Postoperatively, groups chose highly variable target FIX levels for POD1 (see Table 1) with the lowest actual FIX levels of 12–26% reported by Raish and colleagues, with comments of good hemostasis reported in the postoperative period [20]. Other groups chose higher target FIX levels for POD1, aiming for 80–100% FIX levels but achieving lower actual levels of between 72% and 88%. There was no uniformity in the setting of target levels following POD2. Of interest, earlier reports that used FFP quoted FIX levels (actual) of 12–26% for revascularization procedures and FIX levels (target) .25% for complex cardiac reconstruction, throughout the postoperative period [20, 30]. Later reports chose considerably higher FIX levels, although this may reflect differences in approach and ‘mindset’ to achieving higher FIX levels when using high purity factor concentrates and recombinant FIX concentrates.

The other report of valve surgery is one of double mechanical valve replacement (Starr Edwards, Medtronic) in a patient with mild hemophilia B, managed using a combination of bolus infusions and CIs of plasma-derived FIX; despite the complication of pericardial tamponade requiring pericardiostomy on POD3, the patient otherwise recovered well and received prophylactic doses of low-molecular-weight heparin in the immediate postoperative period, stopping on POD11, and was later commenced on acenocoumarol aiming for a target international normalized ratio (INR) of 1.5–2.0. The authors reported the patient to be well 9 months after the operation [28].

DISCUSSION Currently, there are no evidence-based guidelines for the management of hemophilia patients with ischemic heart disease undergoing coronary interventions. Attempts have been made at one institution to develop local guidelines to ensure uniformity of care [2]. The three reports described in this overview of the literature represent such diversity of cases that no firm conclusions can be reached except that PCI can be performed safely in individual cases if a multidisciplinary approach is used.

The duration of factor replacement differed in each report, but no group administered FIX for more than 15 days. From our literature search, none of the case reports attaining FIX levels of 100% or more in the first few days postoperatively documented complications of thromboembolism.

Bypass and Off-pump Techniques

The case reports and letters that describe cardiac surgery in patients with hemophilia B are also highly heterogeneous and only limited conclusions can be drawn. At present, it seems that the factor replacement schedule and antithrombotic strategies for patients with hemophilia B undergoing PCIs and cardiac surgery require to be made on an individual basis.

One group used off-pump revascularization on the basis that this might give less coagulopathy than traditional bypass techniques [27]. All other cases used bypass with heparinization and reversal with protamine.

Thromboprophylaxis Strategies In the case reports of patients undergoing coronary revascularization procedures, only two state the postoperative antithrombotic strategies used. In the report by Bukowski and colleagues [25], aspirin at a dose of 100 mg was given postoperatively, whereas in the report by Grandmougin [27], the patient was treated JCD 2009; 1:(1). OCTOBER 2009

Regarding target FIX levels, following major surgery, the World Federation of Hemophilia recommends achieving a preoperative FIX level of 0.60–0.80 IU/mL in situations of no significant resource constraint. Postoperatively, a trough FIX level of 0.40–0.60 IU/mL 70

www.slm-hematology.com


www.slm-hematology.com

71

JCD 2009; 1:(1). OCTOBER 2009

smoker

CABG 64

CABG 63

2% Bebulin-TIM4IMMUNO

pd-FIX

bolus

bolus

target 100%

actual 90%

target not stated

58yrs

PCC Konyne

Dimitrova, 1995

4.9%

bolus

actual not stated

52yrs

PCC Konyne

CABG64

Scharfman, 1993

BP11

NIDDM10

12%

target 60% actual 84%

74yrs smoker

Wilson, 1991 CABG

target not given

FFP, PEx9

target 80–90%

actual 92%

target 100%

actual 84%

bolus

bolus + CI

PCC8 Konyne

FFP7

BeneFix

recFIX

4

actual 88%

target 80%

pd-FIX{ Alphanine

pre FIX %

bolus/CI bolus + CI1

FIX replacement

actual 51%

5%

2%

8%

14%

basal FIX level %

CABG64

59yrs

40yrs

Tourbaf, 1979

Raish, 1985

congenital

bicuspid valve

valve

61yrs

AVR, tissue

Krakow, 2009

rheumatic

25 yrs

Thankachen, 2007

AVR*, MVR1, mechanical valve

age and risk factors

Author and type of cardiac surgery

Table 2

post FIX %

actual 60%

target not stated

actual not stated

target 50% for 7 days

target 20–40% actual 44% POD1

actual 12–26%

target not stated

.20% 2nd post-op wk

,40–50% POD2–3

actual ,80–90% POD1

target not stated

actual 100% POD2 stepwise reduction POD 3–7

actual 72–88% POD1

target 90–100% POD1

actual 82%

target 50% POD{1–3

not stated

not stated

not stated

not stated

not stated

TXA

5

not stated

antifibrinolytic therapy comments

post-op antithrombotic therapy N/S

Hematoma right thigh at saphenous vein graft site, settled conservatively

discharge on POD13

factor IX doses documented through POD10

post-op antithrombotic therapy N/S

discharge on POD7 post-op antithrombotic therapy N/S

post-op antithrombotic therapy N/S

discharge POD13

good hemostasis documented

post-op bilateral pneumothoraces

post-op antithrombotic therapy N/S

FFP intra-op

ASA6 81mg daily discharge POD7

FIX replacement until POD10

heparin 5000 IU POD1 until POD 7

INR3 1.5–2.0

cardiac tamponade POD4, ARF2, acenocoumarol post-op

low dose LMWH1 initially post-op


JCD 2009; 1:(1). OCTOBER 2009

71yrs

NIDDM

64yrs

BP lipids12

80yrs

Palanzo, 1995

CABG 66

Bukowski, 1996

CABG64

Donahue, 1999

72

drainage of pericardial effusion

Mackinlay 2000 closure ASD;

repair VSD, aortic regurgitation

18yrs F

32% (carrier)

,1%

pd-FIX Immunine

PCC pre-op, FFP, PCC post-op

7yrs

Mazzucco, 1986

FFP pre and POD1

BETAFACT

pd-FIX

BeneFix

recFIX

pd-FIX

Mononine

pd-FIX

FIX replacement

PCC14 Proplex POD2

2%

,1%

8%

6%

2.4%

basal FIX level %

correction transposition of arteries, VSD repair, PS correction

16yrs

FIX infusions

IDDM14

CABG

Roskos, 1983

52yrs

Grandmougin, 2003

CABG 63

age and risk factors

Author and type of cardiac surgery

Table 2. Continued

bolus

bolus

bolus

+CI

bolus

bolus

bolus +CI

bolus

bolus/CI

pre FIX %

post FIX %

actual not stated

POD7–14 50–60%

target POD1–7, 100–120%

actual 37–120%

target 50 – 150%

target and actual not stated

actual 38%

target not stated

target 100% actual not stated

actual 98%

target 100%

‘estimated’ 64%

target not stated actual not stated

actual ,60–70% (graph)

target .50% for 20 days

actual ,40% (graph)

target not stated target .25% for 7 days

actual not stated

target 100–120%

actual 110%

target 100%

actual 71%

target not stated

actual not stated

target not stated

not stated

not stated

not stated

not stated

1g/hr infusion

10g load

e-ACA13

not stated

(total 600 000 KIU)

aprotinin

antifibrinolytic therapy comments

duration of therapy 11+4 days post-op antithrombotic therapy N/S

post-op antithrombotic therapy N/S

post-op antithrombotic therapy N/S

Discharged POD11

enoxaparin 4000IU daily until discharge POD14

off-pump, ASA 160mg daily pre-op,

emergency re-operation, death

type 1 aortic dissection POD2, level 37–50%,

FIX given to POD19

ASA 100mg daily

post-op antithrombotic therapy N/S

discharge on POD6

transient episode atrial fibrillation POD3

Journal of Coagulation Disorders

www.slm-hematology.com


is recommended for the first 1–3 days, then 0.30– 0.50 IU/mL for days 4–6, and 0.20–0.40 IU/mL for days 7–14. There are, however, no specific recommendations regarding the target levels for cardiac surgery, and there may be a risk of thrombosis at supranormal FIX levels [12, 34]. From our overview, it is clear that there is no consensus on the safe target FIX level to attain on POD1 or thereafter in an individual’s management to avoid hemorrhagic and thrombotic complications.

coronary heart disease and valvular heart disease requiring intervention and operation in the future. We are aware that the level of evidence available in the literature search is low and only cautious conclusions can be drawn. Nevertheless, it appears that cardiac surgery is feasible and safe in individuals with all severities of hemophilia B, and to ensure optimal outcome, close cooperation and good communication between all members of the multidisciplinary team is necessary and an individualized written protocol is highly recommended. There are no data from prospective randomized trials to inform the decision to use bolus or CI of FIX. There is also no consensus on the minimal target FIX level to attain peri- and postoperatively, the duration of replacement therapy, or the use of antifibrinolytic agents.

Turning to the mode of FIX administration, the halflife of FIX concentrate is approximately 18 h, and therefore FIX concentrate is usually administered by bolus infusion once daily. There are case reports and case series of the successful use of CI of plasma-derived and recombinant FIX concentrates in patients undergoing different types of general, orthopedic, and dental surgeries [35–38], with theoretical advantages of reduction in total quantity of factor leading to overall reduced cost of a procedure, in addition to avoidance of wide fluctuations in factor levels [39, 40]. This is of potential advantage in patients undergoing cardiac surgery in whom there are several indwelling central catheters and drains that require removal on different days during the postoperative period. Variable clearance of FIX as a result of individual patient characteristics and the postoperative hemostatic changes following major surgery may mean that infusion rates require adjustment on a daily basis [36, 37]. This is likely to be an issue in patients post-bypass, and one group commented that, in the immediate postoperative period, factor levels may need to be checked frequently because of increased consumption of FIX and dilution due to replacement with crystalloid [29].

The bleeding risk for patients with different severities of hemophilia B requiring dual antiplatelet therapies or vitamin K antagonists after angioplasty and stent insertion, or after valve surgery, is unclear. In view of this uncertainty, avoidance of the need for long-term use of such drugs by employing particular types of stent, or choice of valve prosthesis, should be carefully considered in the preoperative discussion and planning of an individual’s operation. Disclosure: The authors have no financial interests to disclose related to the contents of this article.

REFERENCES 1. Street A, Hill K, Sussex B, Warner M, Scully M-F. Haemophilia and ageing. Haemophilia. 2006;12(Suppl 3):8–12. 2. Schutgens REG, Tuinenburg A, Roosendaal G, Hoseyni Guyomi S, Mauser-Bunschoten EP. Treatment of ischaemic heart disease in haemophilia patients: an institutional guideline. Haemophilia. 2009;15:952–958. 3. Mafrici A, Baudo F. Hemophilia and percutaneous coronary interventions. Ital Heart J. 2003;4:731–733. 4. Rosendaal FR, Varekamp I, Smit C, et al. Mortality and cause of death in Dutch haemophiliacs, 1973–86. Br J Haematol. 1989;71: 71–76. 5. Koumbarelis E, Rosendaal FR, Gialeraki A, et al. Epidemiology of haemophilia in Greece; an overview. Thromb Haemost. 1994;72: 808–813. 6. Darby SC, Kan SW, Spooner RJ, et al for the UK Haemophilia Centre Doctors Organisation. Mortality rates, life expectancy, and causes of death in people with hemophilia A or B in the United Kingdom who were not infected with HIV. Blood. 2007; 110:815–825. 7. Aronson DL. Cause of death in haemophilia A patients in the US from 1968–1979. Am J Hematol. 1988;27:7–12. 8. Sramek A, Kriek M, Rosendaal FR. Decreased mortality of ischaemic heart disease among carriers of haemophilia. Lancet. 2003;362:351–354. 9. Rosendaal FR, Briet E, Stibbe J, et al. Haemophilia protects against ischaemic heart disease: a study of risk factors. Br J Haematol. 1990;75:525–530. 10. Kulkarni R, Soucie JM, Evatt BL. Hemophilia Surveillance System Project Investigators. Prevalence and risk factors for heart disease among males with hemophilia. Am J Hematol. 2005; 79:36–42. 11. Sullivan DW, Purdy LJ, Billingham M, Glader BE. Fatal myocardial infarction after therapy with prothrombin complex

Several reports emphasize the importance of multidisciplinary coordination of care to ensure an uncomplicated peri- and postoperative course around the time of cardiac surgery [19, 26, 27, 29, 31, 32]. For cardiac surgery, the need for a written care plan involving interventional cardiologists, cardiac surgeons, and anesthesiologists, in addition to the regional hemophilia service, the coagulation laboratory, and blood bank or pharmacy is viewed as essential. In view of the rarity of hemophilia, specialty and ward nurses, physiotherapists, and pharmacists in intensive care, high-dependency units, and postoperative ward settings must all receive education about the specific issues for a patient with hemophilia undergoing cardiac surgery. This is time-consuming and may be made more complex if the hospital sites involved in coordinating patient care are at different geographical locations.

CONCLUSIONS With improvements in the safety of plasma-derived products, advances in the treatment of HIV and chronic hepatitis, and prophylaxis regimes, life expectancy in hemophilia has increased, and it is likely that there will be an increasing number of patients with www.slm-hematology.com

73

JCD 2009; 1:(1). OCTOBER 2009


Journal of Coagulation Disorders

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24. 25.

26.

27.

28.

29.

30.

31.

concentrate in a young man with haemophilia A. Pediatrics. 1984;74:279–281. Alsolaiman MM, Chang K, Arjomand H, Oza R, Costacurta G. Acute left anterior descending artery occlusion in a hemophiliac A patient during recombinant factor VIII infusion: treatment with coronary angioplasty. Catheterization Cardiovasc Intervent. 2000;50:468–472. Basso IN, Keeling D. Myocardial infarction following recombinant activated factor VII in a patient with type 2A von Willebrand disease. Blood Coag Fibrinol. 2004;15:503–504. Girolami A, Ruzzon E, Fabris F, Varvarikis C, Sartori R, Girolami B. Myocardial infarction and other arterial occlusions in hemophilia a patients. A cardiological evaluation of all 42 cases reported in the literature. Acta Haematol. 2006;116:120–125. Girolami A, Randi ML, Ruzzon E, Zanon E, Girolami B. Myocardial infarction, other arterial thrombosis, and invasive coronary procedures, hemophilia B: a critical evaluation of reported cases. J Thromb Thrombolysis. 2005;20:43–46. Helft G, Metzger JP, Samama MM, Rothschild C, Batisse JP, Vacheron A. Coronary stenting in a hemophilic patient. Thromb Haemost. 1997;77:1044–1045. Bovenzi F, De Luca L, Signore N, Fusco F, de Luca I. Abciximab for the treatment of an acute thrombotic coronary occlusion during stent implantation in a patient with severe hemophilia B. Ital Heart J. 2003;4:728–730. Au WY, Jim MH, Lam CC. Unusual case of coronary artery disease in a patient with severe hemophilia B. Am J Hematol. 2002;69:152–153. Tourbaf KD, Bettigole RE, Zizzi JA, Subramanian S, Andersen MN. Coronary bypass in a patient with hemophilia B, or Christmas disease. Case report. J Thorac Cardiovasc Surg. 1979;77:562–569. Raish RJ, Witte DL, Goldsmith JC. Successful cardiac surgery following plasmapheresis in a patient with hemophilia B. Transfusion. 1985;25:128–130. Wilson CJ, Frankville D, Robinson B, Koch F, Lake CL. Perioperative management of coronary artery bypass surgery in a patient with factor IX deficiency. J Cardiothorac Vasc Anesth. 1991;5:160–162. Scharfman WB, Rauch AE, Ferraris V, Burkart PT. Treatment of a patient with factor IX deficiency (hemophilia B) with coronary bypass surgery. J Thorac Cardiovasc Surg. 1993;105:765–766. Dimitrova A, Tsarianski G, Aleksandrov V, Kereshka P, Nachev G, Chirkov A. Hemophilia B and aortocoronary bypass. Khirurgiia (Sofia). 1995;48:76–79. Palanzo DA, Sadr FS. Coronary artery bypass grafting in a patient with haemophilia B. Perfusion. 1995;10:265–270. Bukowski JG, De Brux JL, Ganascia B, Cottineau C, Jacob JP. Coronary artery bypass with extracorporeal circulation in a patient with hemophilia B. Ann Fr Anesth Reanim. 1996;15:304– 306. Donahue BS, Emerson CW, Slaughter TF. Case 1–1999. Elective and emergency cardiac surgery on a patient with hemophilia B. J Cardiothorac Vasc Anesth. 1999;13:92–97. Grandmougin D, Delolme MC, Reynaud J, Barral X. Off-pump myocardial revascularization in a diabetic patient with severe hemophilia B and impaired left ventricular function: hematological and operative strategies. J Card Surg. 2005;20:366–369. Thankachen R, George B, Shukla V, Korula RJ. Aortic and mitral valve replacement in a patient with hemophilia B. Asian Cardiovasc Thorac Ann. 2007;15:526–527. Krakow EF, Walker I, Lamy A, Anderson JAM. Cardiac surgery in patients with haemophilia B: a case report and review of the literature. Haemophilia. 2009; 15:108–113. Roskos RR, Gilchrist GS, Kazmier FJ, Feldt RH, Danielson GK. Management of hemophilia A and B during surgical correction of transposition of the great arteries. Mayo Clin Proc. 1983;58: 182–186. Mazzucco A, Stellin G, Cantele P, Boschello M, Traldi A, Gallucci V. Repair of ventricular septal defect and aortic

JCD 2009; 1:(1). OCTOBER 2009

32.

33.

34.

35.

36.

37.

38.

39. 40.

74

regurgitation associated with severe hemophilia B. Ann Thorac Surg. 1986;42:97–99. MacKinlay N, Taper J, Renisson F, Rickard K. Cardiac surgery and catheterization in patients with haemophilia. Haemophilia. 2000;6:84–88. Vander Woude JC, Milam JD, Walker WE, Houchin DP, Weiland AP, Cooley DA. Cardiovascular surgery in patients with congenital plasma coagulopathies. Ann Thorac Surg. 1988;46: 283–288. van Hylckama Vlieg A, van der Linden IK, Bertina RM, Rosendaal FR. High levels of factor IX increase the risk of venous thrombosis. Blood. 2000;95:3678–3682. Chowdary P, Dasani H, Jones JA, et al. Recombinant factor IX (BeneFix) by adjusted continuous infusion: a study of stability, sterility and clinical experience. Haemophilia. 2001;7:140–145. Ragni MV, Pasi KJ, White GC, Giangrande PL, Courter SG, Tubridy KL. Use of recombinant factor IX in subjects with haemophilia B undergoing surgery. Haemophilia. 2002;8:91–97. Hoots WK, Leissinger C, Stabler S, et al. Continuous intravenous infusion of a plasma-derived factor IX concentrate (Mononine) in haemophilia B. Haemophilia. 2003;9:164–172. Tagariello G, Davoli PG, Gajo GB, et al. Safety and efficacy of high-purity concentrates in haemophiliac patients undergoing surgery by continuous infusion. Haemophilia. 1999;5:426–430. Schulman S. Continuous infusion. Haemophilia. 2003;9:368–375. Schulman S, Wallensten R, White B, Smith OP. Efficacy of a high purity, chemically treated and nanofiltered factor IX concentrate for continuous infusion in haemophilia patients undergoing surgery. Haemophilia. 1999;5:96–100.

www.slm-hematology.com


JOURNAL OF COAGULATION DISORDERS

REVEW ARTICLE

von Willebrand Factor Proteolysis by ADAMTS13 Derrick J Bowen Affiliation: Department of Haematology, School of Medicine, Cardiff University, Cardiff, UK Submission date: 27th July 2009, Revision date: 1st September 2009, Acceptance date: 5th September 2009

A B S T R A C T The plasma glycoprotein von Willebrand factor (VWF) is essential for effective hemostasis. It mediates platelet adhesion and aggregation at a site of vascular damage. Additionally, it transports coagulation factor VIII (an important protein of the clotting cascade) in the blood, shielding the coagulation factor from inactivation. VWF is found in plasma, platelets, vascular subendothelium and endothelial cells; its presence throughout the vasculature reflects its fundamental role in blood clot formation. Plasma VWF circulates as homopolymers (multimers) of different lengths, with the size differences arising principally through cleavage by the metalloprotease ADAMTS13 (a disintegrin and metalloprotease with thrombospondin repeats). ADAMTS13-mediated VWF proteolysis contributes to the regulation of VWF bioactivity: newly synthesized ultralarge multimers are highly thrombogenic in comparison to smaller proteolysed forms. The regulation of VWF multimer size by ADAMTS13 is essential for normal hemostatic function, as evidenced by the pathological states that occur when proteolysis is deficient or excessive: ADAMTS13 deficiency is the basis for the life-threatening disorder thrombotic thrombocytopenic purpura (TTP), in which ultralarge VWF multimers in the circulation predispose to the spontaneous formation of platelet aggregates, with potentially fatal consequences; conversely, too much proteolysis of VWF underlies one form of the hemorrhagic disorder von Willebrand disease (VWD): hemostasis is severely compromised by the absence of large (and sometimes intermediate) multimers in the circulation due to their rapid breakdown by ADAMTS13. This review aims to provide the clinician or scientist new to this subject area with an overview of ADAMTS13-mediated VWF proteolysis and the known key factors that can affect this important biochemical process in health and disease. Keywords: von Willebrand factor, von Willebrand disease, ADAMTS13, proteolysis, thrombotic thrombocytopenic purpura (TTP) Correspondence: Derrick J Bowen, Department of Haematology, School of Medicine, Cardiff University, Heath Park, Cardiff CF14 4XN, UK. E-mail: bowendj1@cf.ac.uk

The formation of a blood clot at a site of vascular injury is a complex physiological process in which the glycoprotein von Willebrand factor (VWF) plays a fundamental role. Clot formation initiates when blood contacts subendothelial structures, and circulating platelets adhere, triggering further platelet aggregation to produce the primary hemostatic plug or soft clot. During this process, platelet activation provides a procoagulant surface that supports the activity of key protein complexes in the coagulation cascade. The latter produces fibrin polymers that become covalently cross-linked into the soft clot, thereby reinforcing and stabilizing it. Formation of the platelet-rich primary hemostatic plug (primary hemostasis) and clot stabilization (secondary hemostasis) are tightly regulated processes, defects in which give rise to hemorrhagic or thrombotic disorders.

including those involving VWF. However, at high shear rates, there appears to be an absolute dependency on VWF. VWF interacts with platelets via two platelet receptors: glycoprotein Iba (GPIba) [1] and integrin aIIbb3 (previously referred to as platelet glycoprotein IIb/IIIa or GPIIb/IIIa) [2]. GPIba is unique among platelet receptors in that it does not require prior platelet activation in order to bind its ligand (VWF); in contrast, aIIbb3 does. VWF interaction with GPIba is rapid and reversible, whereas interaction with aIIbb3 is slow and irreversible [3, 4]. At high shear rates, the VWF–GPIba interaction tethers platelets long enough for activation and subsequent irreversible binding via aIIbb3. Although GPIba is competent to bind VWF without platelet activation, there is apparently no significant interaction between the two proteins in circulating blood. This is explained, at least in part, by the requirement for a critical level of shear stress [4].

Platelet arrest, adhesion, and aggregation are central to primary hemostasis. At low shear rates, multiple ligand–receptor interactions may participate,

Following platelet arrest and adhesion, aggregation of further platelets involves a number of ligand– receptor interactions, among which VWF–aIIbb3 and

INTRODUCTION

JCD 2009; 1:(1). OCTOBER 2009

75

www.slm-hematology.com


Journal of Coagulation Disorders

fibrinogen–aIIbb3 are important. VWF and fibrinogen act in concert, serving distinct but synergistic roles in promoting platelet aggregation [5]. At high shear rates, surface-bound VWF appears to be capable of reversible association with circulating VWF, and the resulting homotypic multimer assemblies may provide a further contribution to the arrest of passing platelets [6]. At pathophysiological shear rates, transient VWFmediated platelet aggregation can occur that is dependent upon GPIba but independent of aIIbb3 and fibrinogen [7]. The latter may be highly relevant as a pathological mechanism leading to vessel occlusion at a site of stenosis, where vessel narrowing may vastly increase shear stress. However, it may also be relevant in a normal setting: as the soft clot is laid down, the lesion aperture narrows while blood pressure remains unchanged. Exiting blood may therefore be exposed to ever-increasing shear stress, possibly permitting activation-independent, VWF–GPIba-mediated platelet aggregation, and this may facilitate lesion closure.

of dimers [14], is removed leaving multimers composed entirely of mature VWF subunits (Figure 1). Posttranslational processing additionally includes extensive glycosylation involving the addition of 12 Nlinked and 10 O-linked carbohydrate moieties (Figure 1) [15] and sulfation of specific N-linked carbohydrates [16]. Internal homologies within the mature subunit give rise to repeated structural domains [17] (Figure 1). The A1 domain contains the binding site for GPIba (and also for heparin and sulfated glycolipids); the C1 domain contains the ligand motif (Arg-Gly-Asp, RGD) for integrin aIIbb3; the A3 domain (and possibly also the A1 domain) is considered to be the binding site for collagen; and the D9–D3 domains bind FVIII (Figure 1). VWF produced in endothelial cells can be stored in Weibel–Palade bodies [18] or secreted. Secretion can occur by a constitutive pathway that does not require stimulation by secretagogues, or by a regulated pathway that does [19]. The former releases multimers directly upon synthesis; the latter releases the Weibel– Palade stores. Endothelial VWF can be secreted apically into the bloodstream or basolaterally into the subendothelial matrix. Only arteries, arterioles, and large veins have subendothelial deposits of VWF [20]; in capillaries, VWF may be deposited basolaterally in response to endothelial cell activation [21]. Plasma VWF appears to be derived principally from endothelial cells [22]. Platelet VWF is stored in agranules from which it is released upon platelet activation [23]. As much as 15% of the total VWF in blood is contained within the platelet compartment [24]. VWF is therefore found in four physiological compartments—plasma, platelets, endothelial cells, and the subendothelium—and this no doubt reflects its important role in primary hemostasis.

Collagen contributes to the overall process in at least two ways: first, it may potentiate the ability of VWF that is bound to it to arrest platelets; second, it can interact directly with any of several receptors on the platelet surface, thereby adding to platelet–subendothelium interactions. While the intricacies and timeline of events in primary hemostasis may not yet be fully delineated, the detailed studies to date have identified key features of VWF involvement and emphasize the central role that this protein plays at high shear stress. VWF has evidently evolved to ‘‘catch’’ platelets from the passing blood under the adverse circumstance of high shear. In addition to its role in primary hemostasis, VWF contributes towards secondary hemostasis: it transports and protects coagulation factor VIII (FVIII) in the circulation [8]. FVIII is an essential protein in the intrinsic coagulation cascade and is necessary for efficient fibrin production. FVIII deficiency is characterized by a breakdown of secondary hemostasis, resulting in clinically significant bleeding (hemophilia A). Thus, VWF facilitates secondary hemostasis by contributing to the maintenance of circulating FVIII levels.

Stimulated release of VWF from its storage organelles places ultralarge multimers, which are highly thrombogenic, directly at the site of need (Figure 3). In the case of endothelial cells, these acutely released ultralarge multimers rapidly form strings and networks that are anchored on to the cell surface via integrin aVb3 [25]. Such immobilization would expose them to shear stress in the passing blood flow. This may potentiate platelet binding; however, it would also promote VWF proteolysis by the metalloprotease ADAMTS13 (a disintegrin and metalloprotease with thrombospondin repeats) [26].

von WILLEBRAND FACTOR VWF is synthesized by endothelial cells [9] and megakaryocytes [10]. It is produced as a pre-proprotein comprising 2813 amino acids. As part of posttranslational processing, the pre- and pro-sequences (22 and 741 amino acids respectively) are removed, leaving the mature VWF subunit (2050 amino acids) (Figure 1) [11, 12]. Prior to removal of the prosequence, pro-VWF forms dimers via C-terminal disulfide bonds, and these dimeric units subsequently join via their N-termini to give polymers (multimers) [13]. The pro-peptide, which is needed for multimerization JCD 2009; 1:(1). OCTOBER 2009

ADAMTS13 ADAMTS13 is synthesized predominantly in the liver [27]; however, other tissues and cells (including endothelial cells and platelets) also produce the enzyme [28, 29]. The primary translation product is 1427 amino acids long and comprises a signal peptide (33 amino acids), a propeptide (41 amino acids), and a multidomain mature protein (1353 amino acids) [30] 76

www.slm-hematology.com


Figure 1. von Willebrand factor structure. (A) Relative extent of pre-, pro-, and mature VWF. (B) Pre-pro-VWF domain structure and position of carbohydrate moieties in the mature subunit. Internal homologies within the protein give rise to repeated domains (A1–D4); D9 is a partial homolog of the D domains; CK is a cysteine-rich region (cysteine-knot); H and D represent N-linked and O-linked glycosylation sites respectively. The A2 domain contains the ADAMTS13 cleavage site at Tyr1605–Met1606, the Tyr1584Cys polymorphic residue (*) that affects proteolysis, two N-linked carbohydrates (Asn1515 and Asn1574, of which Asn1574 may influence proteolysis), and five O-linked carbohydrates. The numeric scale indicates amino acid residue number. (C) Location of VWF functional domains involved in ligand binding and multimer formation (RGD represents the arg-gly-asp motif, which is the ligand for integrin aIIbb3). (D) Formation of VWF multimers. Mature VWF dimerizes via C-termini and dimers form multimers via N-termini. ADAMTS13 proteolysis within any of the constituent A2 domains yields multimeric forms found in plasma

(Figure 2). The latter includes a metalloprotease domain, disintegrin domain, cysteine-rich and spacer region, and several thrombospondin type 1 motifs, all of which characterize the enzyme as a member of the ADAMTS family [30]. The metalloprotease domain has the consensus sequence HEXXHXXGXXHD characteristic of certain zinc-dependent metalloendopeptidases

[31], a predicted calcium ion binding site (Glu83, Asp173, Cys281), and a ‘‘met turn’’ in which Met249 supports the active site zinc ion. These features are characteristic of the ‘‘metzincin’’ family of metalloendopeptidases [31], which achieve optimal activity with both zinc and calcium ions. Cooperative activation of ADAMTS13 has been demonstrated by zinc and

Figure 2. Structure of ADAMTS13. Pre- and pro-sequences are followed by a metalloprotease domain containing the active site, a disintegrin domain, a TSP-1 (thrombospondin-1) motif, a Cys-rich region, a spacer domain, seven additional TSP-1 motifs, and two CUB domains

www.slm-hematology.com

77

JCD 2009; 1:(1). OCTOBER 2009


Journal of Coagulation Disorders

Figure 3. A model of events in primary hemostasis, illustrated by a single site of vascular damage at which blood exits the vessel. (A) In an intact blood vessel, VWF is compartmentalized: ultralarge, highly thrombogenic multimers are localized to the a-granules of platelets and Weibel–Palade bodies of endothelial cells; plasma VWF (which comprises smaller, less thrombogenic forms) is prevented from immobilization by the endothelial layer, which also masks matrix-bound subendothelial VWF (if present). (B) Vessel damage allows exiting blood to contact the subendothelial matrix (containing collagen to which plasma VWF can bind) and matrix-bound VWF (if present). (C) In the earliest stages, platelets arrest and adhere to immobilized VWF. (D) Platelet adhesion is followed by activation and release of agranule contents including ultralarge VWF multimers. Local agonists may stimulate endothelial cells to release ultralarge VWF multimers from Weibel–Palade bodies. (E) VWF-mediated platelet recruitment and ADAMTS13-mediated VWF proteolysis may take place concurrently during subsequent stages. The former would contribute to clot growth; the latter would limit it. Both are favored by high shear, which may itself be facilitated by the narrowing of the lesion aperture as the soft clot evolves. Factors that enhance proteolysis (such as the cysteine 1584 variant of VWF or, to a lesser extent, blood group O) or inhibit proteolysis (such as the serine 475 variant of ADAMTS13) may imbalance VWF-mediated platelet recruitment and predispose to a hemorrhagic or thrombotic tendency respectively. For clarity, peripheral constituents of the blood have been omitted from (D) and (E)

calcium together [32]. The C-terminus of ADAMTS13 contains two CUB domains that are common to many proteins involved in developmental processes, but whose function in ADAMTS13 is uncertain (CUB abbreviates: complement component Clr/Cls, Uegf and bone morphogenic protein 1 domain). Circulating ADAMTS13 is highly glycosylated [33, 34] and is primarily full length (pre-pro-sequences attached) [35].

phenomena could contribute to the control of clot growth (Figure 3), particularly when ADAMTS13 is available from the various local sources (plasma, endothelial cells, and activated platelets). Evidence that excess cleavage has a negative impact on VWF hemostatic function is provided by certain forms of the bleeding disorder von Willebrand disease (VWD) that result from a mutant VWF that is highly susceptible to ADAMTS13 proteolysis [36]. Conversely, a deficiency or dysfunction of ADAMTS13 leads to the survival of uncleaved, or partially cleaved, ultralarge VWF in plasma, which predisposes to the spontaneous formation of intravascular platelet aggregates and is the biochemical basis of the life-threatening disorder thrombotic thrombocytopenic purpura (TTP) [37, 38, 41].

ADAMTS13 cleaves VWF at the peptide bond between Tyr1605 and Met1606, located in the VWF A2 domain (Figure 1) [36–38]. This proteolysis gives rise to multimers of different lengths (Figure 1) and underlies the characteristic triplet structure of plasma VWF [39]. The proteolysis is important because it appears to be the predominant physiological mechanism through which multimer size is regulated. Although ADAMTS13 and VWF both circulate in plasma, their interaction leading to proteolysis is limited by a requirement for VWF to undergo shear stress, the basis for which may be stretching of the protein and exposure of the otherwise buried A2 domain [26, 40].

It is presently unclear which sites on ADAMTS13 and VWF contribute towards the interaction of the two proteins under physiologic conditions. Data to date support the involvement of several sites of interaction on both proteins, at least under static/ denaturing conditions. The isolated metalloprotease domain of ADAMTS13 is ineffective in cleaving VWF [42, 43]; efficient proteolysis requires the presence of the non-catalytic domains. The disintegrin-like domain

Therefore, shear stress, on the one hand, promotes VWF interaction with platelets, while on the other, it enhances cleavage by ADAMTS13. These two opposing JCD 2009; 1:(1). OCTOBER 2009

78

www.slm-hematology.com


recruit platelets and, as noted earlier, this may provide a mechanism contributing to the control of clot growth. ADAMTS13 can limit the activation-independent, VWF– GPIba-mediated aggregation of platelets that takes place at pathologically elevated shear stress [7] and may thereby provide a control mechanism against thrombogenesis at sites experiencing such shear.

appears to interact with VWF via Arg349 and Leu350 in ADAMTS13 and Asp1614 and Ala1612 in VWF [44]. The ADAMTS13 TSP-1 (first motif) and Cys-rich domains, respectively, appear to bind VWF regions Gln1624–Val1630 and Ile1642–Gln1652 [43, 45]. The spacer domain binds a C-terminal portion of the VWF A2 domain (residues 1653–1668) [46]. The carboxyterminal TSP-1 repeats and CUB domains may modulate ADAMTS13–VWF interaction [46, 47]. Data from mutagenesis experiments suggest that ADAMTS13 is relatively flexible and does not require a fixed spacing between its catalytic site and the distal VWF binding sites in the Cys-rich and spacer domains [45]. This ‘‘elasticity’’ may facilitate interaction between the two proteins in flow conditions.

VWF Glycosylation Studies prior to the discovery of ADAMTS13 provided evidence that the carbohydrate component of VWF may protect the protein from degradation [58]. Subsequent studies have indicated that VWF glycosylation can affect ADAMTS13-mediated proteolysis. ABO blood group sugars—A, B, and H—are attached to the N-linked carbohydrates of VWF [59, 60] (Figure 1) and appear to alter the rate of ADAMTS13 cleavage to a small extent: proteolysis was faster for VWF of blood group O compared with non-O (O > B . A > AB); however, the differences between blood groups were not considerable [61]. VWF from the plasma of individuals with the Bombay phenotype (characterized by a failure to attach A, B, and H sugars to carbohydrate structures) was proteolysed more rapidly than that of both blood group O and blood group AB (Bombay . O . AB) [62].

A model has been proposed in which ADAMTS13– VWF interaction comprises two components acting in concert: tight binding is provided by the Cys-rich and spacer domain interactions, whereas the weaker interactions (in particular of the disintegrin domain) may assist in positioning the VWF scissile bond into the active site cleft of the metalloprotease domain [44].

FACTORS INFLUENCING VWF PROTEOLYSIS BY ADAMTS13 Various factors can affect the efficiency of VWF cleavage by ADAMTS13. Those that contribute towards the normal spectrum of proteolysis include shear stress, VWF glycosylation, amino acid variants in both proteins, and ligand–VWF interaction; these are discussed in detail below. Additionally, there are factors that underlie the extremes of proteolysis found in the pathologies of TTP and VWD. These are summarized briefly below, but are reviewed in detail elsewhere [48–52].

The physiological significance of these findings is uncertain. Plasma VWF level correlates with ABO blood group (O , A , B , AB) and is approximately 25% less in blood group O than in non-O [63, 64]. The rank order and magnitude of proteolysis in the ABO blood groups do not correlate with those of VWF level; therefore, proteolysis is unlikely to explain the differences in the amount of VWF between these blood groups. The increased VWF proteolysis associated with blood group O is small; however, it could have a subtle deleterious effect on primary hemostasis. In conjunction with a low plasma VWF level and/or other deleterious factors, this may compromise clot formation (Figure 3) [48].

Shear Stress The velocity of blood flow is dependent upon several parameters including blood pressure, vessel diameter, and fluid viscosity. Within a given vessel, flow differs according to the axial location within the lumen: it is zero at the vessel wall and increases towards the lumen centre [53]. This gives rise to shear stress, produced by the faster movement of one layer over an adjacent layer. In the circulatory system, shear stress is highest in the microvasculature [54, 55]. In vessels occluded by an atherosclerotic plaque, shear stress at the site of stenosis can greatly exceed even the highest normal levels [56].

Global removal of VWF N-linked carbohydrates has been shown to increase affinity for, and proteolysis by, ADAMTS13, and to enable proteolysis in the absence of denaturant [65]. Mutation of the two asparagine residues vicinal to the Tyr1605–Met1606 cleavage site demonstrated that loss of the carbohydrate moiety at Asn1574, but not Asn1515, increased the susceptibility of VWF to proteolysis and allowed cleavage in the absence of denaturant [65]. Thus, VWF N-linked carbohydrates collectively may help to support a globular conformation, and Asn1574 may additionally influence accessibility of the scissile Tyr1605–Met1606 bond to the ADAMTS13 active site.

ADAMTS13 does not proteolyse VWF significantly under static conditions, but does cleave it slowly in vitro in the presence of denaturing agents [37]. However, at shear stress levels equal to those found in the microvasculature, proteolysis is extremely rapid and does not require denaturants [26, 38, 57]. In a physiological setting, ADAMTS13 therefore functions efficiently under the very conditions in which VWF functions to www.slm-hematology.com

In contrast, N-linked carbohydrates on ADAMTS13 do not appear to modulate its activity: they can be enzymatically removed without affecting proteolysis. However, they do appear to be necessary for efficient 79

JCD 2009; 1:(1). OCTOBER 2009


Journal of Coagulation Disorders

secretion of the metalloprotease and may promote correct conformational maturation of the enzyme during its biosynthesis [34].

[71], possibly due to Cys1584 inducing a conformational change in the A2 domain that abolishes the blood group effect. As discussed above, increased proteolysis of VWF may be detrimental to hemostatic function (Figure 3). Cys1584 is enriched in type 1 VWD [72, 75]; however, whether this relates to its effect on VWF proteolysis is uncertain. There is adequate evidence showing that, in addition to increased proteolysis, the variant is associated with decreased VWF level, increased clearance of the protein, and decreased VWF functionality [64, 76]. It is likely that the prevalence of Cys1584 in type 1 VWD reflects all of these deleterious effects [77].

Amino Acid Variants Both VWF and ADAMTS13 show amino acid variation in the normal population. Several coding sequence variants have been described in ADAMTS13, including Arg7Trp, Gln448Glu, Pro475Ser, Pro618Ala, Arg625His, Ala732Val, and Ala900Val [41]. The Pro475Ser variant, which is prevalent in the Japanese population [66] but absent or rare in other populations (Caucasians, AfroAmericans, Chinese) [67, 68], has a significant effect on proteolytic activity. The serine allele is associated with very low ADAMTS13 activity despite normal secretion. Based upon allele frequencies at the genetic level, as many as 10% of the Japanese population may be heterozygous for Ser475 and consequently may have decreased ADAMTS13 levels [66].

The amino acid variation Val1565Leu located in the VWF A2 domain has also been shown to affect proteolysis, however to a much smaller extent than Tyr1584Cys [71]. The rarer leucine allele correlates with a minor increase in proteolytic susceptibility; however, the magnitude of the effect is so small that it may have no physiological relevance.

The polymorphism Pro618Ala in ADAMTS13 also has a significant effect on ADAMTS13 level, mediated by a deleterious effect of the Ala618 allele on secretion and proteolytic activity [69]. Ala732Val has a small but demonstrable effect, the valine allele correlating with impaired catalytic activity [69].

Interestingly, both Tyr1584 and Val1565 are highly conserved in VWF from different species, they each occur in regions that are homologous between species, and they are near to the ADAMTS13 scissile bond [71]. These features provide an argument for a direct effect of the variant residues Cys1584 and Leu1565 on VWF proteolysis.

Studies on the interplay between ADAMTS13 amino acid variants have indicated that the same polymorphisms can be either positive or negative modifiers depending on the allelic combination present. Additionally, polymorphisms may interact synergistically and not just additively [69]. These observations may be relevant to normal variations in hemostatic function (although this has not been formally investigated), and there is evidence supporting their importance in the phenotypic expression of TTP [69].

The possibility that the amino acid polymorphisms Asp1472His, Gln1571His, Pro1601Thr, and Gly1643Ser in the VWF A domains influence ADAMTS13-mediated proteolysis has been investigated using recombinant expressed proteins. The advantage of this approach is that potential confounding variables are standardized, thereby allowing comparison solely of the effect of each amino acid allele. Using recombinant VWF and recombinant ADAMTS13, His1472, His1571, and Thr1601 in VWF all conferred mild resistance to proteolysis, whereas Ser1643 potentiated proteolysis [78]. The disadvantage of recombinant studies is that VWF and ADAMTS13 expressed in vitro may differ from their in vivo counterparts in a number of ways (for example in terms of glycosylation) that could influence the results. In this context, it is interesting to note that, in contrast to the above result, His1472 appeared to have no effect on proteolysis when plasma VWF, rather than the recombinant protein, was used [71].

Amino acid variation in VWF can also affect ADAMTS13-mediated proteolysis. The Tyr1584Cys variation in the VWF A2 domain has been extensively studied. The cysteine allele is associated with mildly enhanced proteolysis [70, 71], the basis for which may be the formation of new inter- or intramolecular disulfide linkages that alter the conformation of the A2 domain and enhance access to the cleavage site [40, 72, 73]. That Cys1584 is necessary for enhanced proteolysis and not simply linked to a causative change elsewhere in VWF has been shown by comparing the coding sequences of the protein in related individuals whose plasma VWF showed either normal or increased proteolysis. Two sequence variants were unique to VWF that showed increased proteolysis: Arg484 and Cys1584. Arg484, in the absence of Cys1584, did not influence VWF proteolysis, indicating that Cys1584 is necessary for the effect [74].

In combination, these detailed studies indicate that ADAMTS13-mediated VWF proteolysis appears to be highly dependent upon the amino acid variants present at certain influential positions in both proteins. One important inference from this is that the rate of proteolysis could be quite different between two individuals who have similar VWF protein levels and also similar ADAMTS13 protein levels, potentially leading to differences in hemostatic efficiency despite the similar phenotypic values. This may contribute to

The mildly enhanced proteolysis associated with Cys1584 is considerably greater than that of blood group O. It is not additive with that of blood group O JCD 2009; 1:(1). OCTOBER 2009

80

www.slm-hematology.com


normal variation in healthy individuals, and it may be relevant to the phenotypic expression of TTP and VWD.

characteristic feature of a subset of type 2A VWD (‘‘group 2’’ type 2A) [86]. In this subset, missense point mutations in the region of the VWF gene (VWF, located at 12p12–13) encoding the A2 domain bring about amino acid changes that cause the protein to be proteolysed without the need for shear stress. These amino acid substitutions may induce conformational changes that alter the accessibility of the cleavage site within the VWF A2 domain, or they may directly alter ADAMTS13–VWF interaction without structural perturbation [40, 87, 88]. In the circulation, the mutant VWF is spontaneously and rapidly cleaved by ADAMTS13, resulting in the loss of large (and sometimes intermediate) multimer forms [86, 87, 89]. This, together with the possibility that ultralarge multimers secreted at a time of hemostatic challenge may be cleaved far more rapidly than normal, severely compromises hemostatic function. Characteristic clinical features include mucocutaneous bleeding, epistaxis, easy bruising, bleeding after dental extraction, heavy menstrual periods, and excessive blood loss during childbirth (for a review, see http://www.nhlbi. nih.gov/guidelines/vwd/index.htm).

VWF–Ligand Binding X-ray crystallography studies indicate that the VWF A1 domain undergoes large structural changes on binding with GPIba [79]. Additionally, proteolysis of a recombinant peptide containing the VWF A1A2A3 region was increased when a mutant fragment of GPIba bound [80]. Together, these observations suggest that binding of GPIba to VWF increases the susceptibility of the latter to proteolysis by ADAMTS13 via a conformational change. Heparin, a glycosaminoglycan stored in the secretory granules of mast cells and released into the blood at a site of vascular damage, also increases VWF proteolysis [80]. The obsolete antibiotic ristocetin, which is used in the laboratory to trigger platelet–VWF aggregation, causes increased proteolysis [61]. For both heparin and ristocetin, the mechanism of the effect may be through the induction of conformational changes upon binding to VWF.

CLINICAL SIGNIFICANCE

For both TTP and VWD, the parameters that influence VWF proteolysis by ADAMTS13 may contribute towards differences in penetrance and severity. This is likely to be most important in the case of mutations in either protein that cause a mild deficiency: the additive or synergistic effect of negative modifiers acting on such mutations could tip the balance in favor of clinical presentation. This is certainly believed to be the case for the Tyr1584Cys polymorphism in VWF: the cysteine allele is found in association with phenotypes ranging from asymptomatic to overt type 1 VWD.

ADAMTS13-mediated proteolysis is a natural part of VWF processing and, as is the case for all other biochemical processes, displays a normal range in the population [81]. At the extremes of the normal range, the level of proteolysis may well predispose to thrombotic (low level) or hemorrhagic (high level) tendencies. Beyond theses extremes, the overt pathological disorders of TTP and a subset of VWD arise. TTP is caused by a marked deficiency of ADAMTS13 activity, brought about either by mutation of the relevant gene (ADAMTS13, located at 9q34) [41] or by acquired antibodies arising from an autoimmune response to the protein [81, 82]. In the case of the former (hereditary TTP), heterogeneous mutations have been characterized in ADAMTS13, including missense and nonsense point mutations, splice site mutations, and deletions [51]. Characterization of antibodies in acquired TTP has shown that diverse ADAMTS13 epitopes may be targeted [83]. Irrespective of the root cause of ADAMTS13 deficiency, the outcome is the survival of ultralarge VWF multimers in the circulation, leading to spontaneous platelet aggregation with potentially disastrous consequences [84]. These include damage or failure of the major organs fed by the circulatory system (heart, kidneys, brain, eyes). Five classical symptoms are indicative of TTP—neurologic disturbance, kidney failure, fever, thrombocytopenia, and microangiopathic hemolytic anemia. The last is believed to arise from shearinduced damage of red blood cells passing the platelet aggregates (for a review, see [85]).

SUMMARY In an undamaged, healthy vessel, blood is maintained in a liquid state by several key phenomena: the endothelial surface lining the lumen is anticoagulant, endothelial cells themselves secrete anticoagulant molecules, hemostatic proteins in the blood circulate in inactive forms, and platelets, although primed ready for clot formation, do not do so because the appropriate triggers are unavailable to them. All this changes when vascular damage occurs and the various signals that activate clot formation come into play. Amid the complex array of biochemical activities that occur, VWF capture of platelets at the site of damage is fundamental. The efficiency of the entire process of clot formation reflects a considerable number of variables. VWF proteolysis by ADAMTS13 is just one biochemical process in the complex milieu, but it is of pivotal importance, as evidenced by the severe pathologies arising from its perturbation. Proteolysis is itself influenced by many variables, including the amount of ADAMTS13 and VWF in the

In contrast to the above, excessive VWF proteolysis by ADAMTS13 may predispose to bleeding. This is the www.slm-hematology.com

81

JCD 2009; 1:(1). OCTOBER 2009


Journal of Coagulation Disorders

19. Sporn LA, Marder VJ, Wagner DD. Differing polarity of the constitutive and regulated secretory pathways for von Willebrand factor in endothelial cells. J Cell Biol. 1989;108(4): 1283–1289. 20. van der Kwast TH, Stel HV, Cristen E, Bertina RM, Veerman EC. Localization of factor VIII-procoagulant antigen: an immunohistological survey of the human body using monoclonal antibodies. Blood. 1986;67(1):222–227. 21. Takeuchi M, Nagura H, Kaneda T. DDAVP and epinephrineinduced changes in the localization of von Willebrand factor antigen in endothelial cells of human oral mucosa. Blood. 1988; 72(3):850–854. 22. Nichols TC, Samama CM, Bellinger DA, et al. Function of von Willebrand factor after crossed bone marrow transplantation between normal and von Willebrand disease pigs: effect on arterial thrombosis in chimeras. Proc Natl Acad Sci USA. 1995; 92(7):2455–2459. Comment in Proc Natl Acad Sci USA. 1995;92(7):2428–2432. 23. Koutts J, Walsh PN, Plow EF, Fenton JW 2nd, Bouma BN, Zimmerman TS. Active release of human platelet factor VIIIrelated antigen by adenosine diphosphate, collagen, and thrombin. J Clin Invest. 1978;62(6):1255–1263. 24. Howard MA, Montgomery DC, Hardisty RM. Factor-VIII-related antigen in platelets. Thromb Res. 1974;4(5):617–624. 25. Huang J, Roth R, Heuser JE, Sadler JE. Integrin aVb3 on human umbilical vein endothelial cells binds von Willebrand factor strings under fluid shear stress. Blood. 2009;113(7):1589–1597. 26. Dong JF, Moake JL, Nolasco L, et al. ADAMTS-13 rapidly cleaves newly secreted ultralarge von Willebrand factor multimers on the endothelial surface under flowing conditions. Blood. 2002;100(12):4033–4039. 27. Uemura M, Tatsumi K, Matsumoto M, et al. Localization of ADAMTS13 to the stellate cells of human liver. Blood. 2005; 106(3):922–924. 28. Liu L, Choi H, Bernardo A, et al. Platelet-derived VWF-cleaving metalloprotease ADAMTS-13. J Thromb Haemost. 2005;3(11): 2536–2544. 29. Shang D, Zheng XW, Niiya M, Zheng XL. Apical sorting of ADAMTS13 in vascular endothelial cells and Madin–Darby canine kidney cells depends on the CUB domains and their association with lipid rafts. Blood. 2006;108(7):2207–2215. 30. Zheng X, Chung D, Takayama TK, Majerus EM, Sadler JE, Fujikawa K. Structure of von Willebrand factor-cleaving protease (ADAMTS13), a metalloprotease involved in thrombotic thrombocytopenic purpura. J Biol Chem. 2001;276(44):41059– 41063. 31. Gomis-Ru¨th FX. Structural aspects of the metzincin clan of metalloendopeptidases. Mol Biotechnol. 2003;24(2):157–202. 32. Anderson PJ, Kokame K, Sadler JE. Zinc and calcium ions cooperatively modulate ADAMTS13 activity. J Biol Chem. 2006; 281(2):850–857. 33. Ricketts LM, Dlugosz M, Luther KB, Haltiwanger RS, Majerus EM. O-fucosylation is required for ADAMTS13 secretion. J Biol Chem. 2007;282:17014–17023. 34. Zhou W, Tsai HM. N-Glycans of ADAMTS13 modulate its secretion and von Willebrand factor cleaving activity. Blood. 2009;113(4):929–935. 35. Soejima K, Nakamura H, Hirashima M, Morikawa W, Nozaki C, Nakagaki T. Analysis on the molecular species and concentration of circulating ADAMTS13 in blood. J Biochem. 2006;139(1): 147–154. 36. Dent JA, Berkowitz SD, Ware J, Kasper CK, Ruggeri ZM. Identification of a cleavage site directing the immunochemical detection of molecular abnormalities in type IIA von Willebrand factor. Proc Natl Acad Sci USA. 1990;87(16):6306–6310. Erratum in Proc Natl Acad Sci USA. 1990;87(23):9508. 37. Furlan M, Robles R, La¨mmle B. Partial purification and characterization of aprotease from human plasma cleaving von Willebrand factor to fragments produced by in vivo proteolysis. Blood. 1996;87(10):4223–4234.

blood, shear stress, amino acid polymorphisms within each protein, VWF–GPIba binding, and ABO blood group. These provide for a large number of possible permutations, some of which may be important to the phenotypic expression of TTP or VWD. Disclosure: The author has no financial interests to disclose related to the contents of this article.

REFERENCES 1. Martin SE, Marder VJ, Francis CW, Loftus LS, Barlow GH. Enzymatic degradation of the factor-VIII-von-Willebrand protein: a unique tryptic fragment with ristocetin cofactor activity. Blood. 1980;55(5):848–858. 2. Girma JP, Kalafatis M, Pie´tu G, et al. Mapping of distinct von Willebrand factor domains interacting with platelet GPIb and GPIIb/IIIa and with collagen using monoclonal antibodies. Blood. 1986;67(5):1356–1366. 3. Savage B, Saldivar E, Ruggeri ZM. Initiation of platelet adhesion by arrest onto fibrinogen or translocation on von Willebrand factor. Cell. 1996;84(2):289–297. 4. Doggett TA, Girdhar G, Lawshe´ A, et al. Selectin-like kinetics and biomechanics promote rapid platelet adhesion in flow: the GPIb(alpha)-vWF tether bond. Biophys J. 2002;83(1):194–205. 5. Ruggeri ZM, Dent JA, Saldivar E. Contribution of distinct adhesive interactions to platelet aggregation in flowing blood. Blood. 1999;94(1):172–178. 6. Savage B, Sixma JJ, Ruggeri ZM. Functional self-association of plasma von Willebrand factor during platelet adhesion under flow. Proc Natl Acad Sci USA. 2002;99(1):425–430. 7. Donadelli R, Orje JN, Capoferri C, Remuzzi G, Ruggeri ZM. Size regulation of von Willebrand factor-mediated platelet thrombi by ADAMTS13 in flowing blood. Blood. 2006;107(5):1943–1950. 8. Foster PA, Fulcher CA, Marti T, Titani K, Zimmerman TS. A major factor VIII binding domain resides within the aminoterminal 272 amino acid residues of von Willebrand factor. J Biol Chem. 1987;262(18):8443–8446. 9. Jaffe EA, Hoyer LW, Nachman RL. Synthesis of von Willebrand factor by cultured human endothelial cells. Proc Natl Acad Sci USA. 1974;71(5):1906–1909. 10. Nachman R, Levine R, Jaffe EA. Synthesis of FVIII antigen by cultured guinea-pig megakaryocytes. J Clin Invest. 1977;60(4): 914–921. 11. Verweij CL, Diergaarde PJ, Hart M, Pannekoek H. Full-length von Willebrand factor (vWF) cDNA encodes a highly repetitive protein considerably larger than the mature vWF subunit. EMBO J. 1986;5(8):1839–1847. Erratum in EMBO J. 1986;5(11):3074. 12. Bonthron DT, Handin RI, Kaufman RJ, et al. Structure of prepro-von Willebrand factor and its expression in heterologous cells. Nature. 1986;324(6094):270–273. 13. Wagner DD, Marder VJ. Biosynthesis of von Willebrand protein by human endothelial cells: processing steps and their intracellular localization. J Cell Biol. 1984;99(6):2123–2130. 14. Verweij CL, Hart M, Pannekoek H. Expression of variant von Willebrand factor (vWF) cDNA in heterologous cells: requirement of the pro-polypeptide in vWF multimer formation. EMBO J. 1987;6(10):2885–2890. 15. Titani K, Kumar S, Takio K, et al. Amino acid sequence of human von Willebrand factor. Biochemistry. 1986;25(11):3171–3184. 16. Carew JA, Browning PJ, Lynch DC. Sulfation of von Willebrand factor. Blood. 1990;76(12):2530–2539. 17. Mancuso DJ, Tuley EA, Westfield LA, et al. Structure of the gene for human von Willebrand factor. J Biol Chem. 1989;264(33): 19514–19527. 18. Wagner DD, Olmsted JB, Marder VJ. Immunolocalization of von Willebrand protein in Weibel–Palade bodies of human endothelial cells. J Cell Biol. 1982;95(1):355–360.

JCD 2009; 1:(1). OCTOBER 2009

82

www.slm-hematology.com


38. Tsai HM. Physiologic cleavage of von Willebrand factor by a plasma protease is dependent on its conformation and requires calcium ion. Blood. 1996;87(10):4235–4244. 39. Fischer BE, Thomas KB, Schlokat U, Dorner F. Triplet structure of human von Willebrand factor. Biochem J. 1998;331(Pt 2):483– 488. 40. O’Brien LA, Sutherland JJ, Weaver DF, Lillicrap D. Theoretical structural explanation for Group I and Group II, type 2A von Willebrand disease mutations. J Thromb Haemost. 2005;3(4):796– 797. 41. Levy GG, Nichols WC, Lian EC, et al. Mutations in a member of the ADAMTS gene family cause thrombotic thrombocytopenic purpura. Nature. 2001;413(6855):488–494. 42. Ai J, Smith P, Wang S, Zhang P, Zheng XL. The proximal carboxyl-terminal domains of ADAMTS13 determine substrate specificity and are all required for cleavage of von Willebrand factor. J Biol Chem. 2005;280(33):29428–29434. 43. Gao W, Anderson PJ, Majerus EM, Tuley EA, Sadler JE. Exosite interactions contribute to tension-induced cleavage of von Willebrand factor by the antithrombotic ADAMTS13 metalloprotease. Proc Natl Acad Sci USA. 2006;103(50):19099–19104. 44. de Groot R, Bardhan A, Ramroop N, Lane DA, Crawley JT. Essential role of the disintegrin-like domain in ADAMTS13 function. Blood. 2009;113(22):5609–5616. 45. Gao W, Anderson PJ, Sadler JE. Extensive contacts between ADAMTS13 exosites and von Willebrand factor domain A2 contribute to substrate specificity. Blood. 2008;112(5):1713–1719. 46. Majerus EM, Anderson PJ, Sadler JE. Binding of ADAMTS13 to von Willebrand factor. J Biol Chem. 2005;280(23):21773–21778. 47. Zhang P, Pan W, Rux AH, Sachais BS, Zheng XL. The cooperative activity between the carboxyl-terminal TSP1 repeats and the CUB domains of ADAMTS13 is crucial for recognition of von Willebrand factor under flow. Blood. 2007;110(6):1887–1894. 48. Bowen DJ, Collins PW. Insights into von Willebrand factor proteolysis: clinical implications. Br J Haematol. 2006;133(5):457– 467. 49. Bowen DJ. Molecular genetics of von Willebrand disease. In: Encyclopaedia of Life Sciences (ELS). Chichester: John Wiley & Sons Ltd; 2008. DOI: 10.1002/9780470015902.a0021447. 50. Collins PW, Cumming AM, Goodeve AC, Lillicrap D. Type 1 von Willebrand disease: application of emerging data to clinical practice. Haemophilia. 2008;14(4):685–696. 51. Levy GG, Motto DG, Ginsburg D. ADAMTS13 turns 3. Blood. 2005;106(1):11–17. 52. Feys HB, Deckmyn H, Vanhoorelbeke K. ADAMTS13 in health and disease. Acta Haematol. 2009;121(2–3):183–185. 53. Long DS, Smith ML, Pries AR, Ley K, Damiano ER. Microviscometry reveals reduced blood viscosity and altered shear rate and shear stress profiles in microvessels after hemodilution. Proc Natl Acad Sci USA. 2004;101(27):10060– 10065. Erratum in Proc Natl Acad Sci USA. 2004;101(39):14304. 54. Lipowsky HH, Kovalcheck S, Zweifach BW. The distribution of blood rheological parameters in the microvasculature of cat mesentery. Circ Res. 1978;43(5):738–749. 55. Lipowsky HH. Microvascular rheology and hemodynamics. Microcirculation. 2005;12(1):5–15. 56. Strony J, Beaudoin A, Brands D, Adelman B. Analysis of shear stress and hemodynamic factors in a model of coronary artery stenosis and thrombosis. Am J Physiol. 1993;265(5 Pt 2):H1787–1796. 57. Tsai HM, Sussman II, Nagel RL. Shear stress enhances the proteolysis of von Willebrand factor in normal plasma. Blood. 1994;83(8):2171–2179. 58. Federici AB, Elder JH, De Marco L, Ruggeri ZM, Zimmerman TS. Carbohydrate moiety of von Willebrand factor is not necessary for maintaining multimeric structure and ristocetin cofactor activity but protects from proteolytic degradation. J Clin Invest. 1984;74(6):2049–2055. 59. Sodetz JM, Paulson JC, McKee PA. Carbohydrate composition and identification of blood group A, B, and H oligosaccharide

www.slm-hematology.com

60.

61.

62.

63.

64.

65.

66.

67.

68.

69.

70.

71.

72.

73.

74.

75.

76.

77.

78.

83

structures on human Factor VIII/von Willebrand factor. J Biol Chem. 1979;254(21):10754–10760. Matsui T, Titani K, Mizuochi T. Structures of the asparaginelinked oligosaccharide chains of human von Willebrand factor. Occurrence of blood group A, B, and H(O) structures. J Biol Chem. 1992;267(13):8723–8731. Bowen DJ. An influence of ABO blood group on the rate of proteolysis of von Willebrand factor by ADAMTS13. J Thromb Haemost. 2003;1(1):33–40. O’Donnell JS, McKinnon TA, Crawley JT, Lane DA, Laffan MA. Bombay phenotype is associated with reduced plasma-VWF levels and an increased susceptibility to ADAMTS13 proteolysis. Blood. 2005;106(6):1988–1991. Gill JC, Endres-Brooks J, Bauer PJ, Marks WJ Jr, Montgomery RR. The effect of ABO blood group on the diagnosis of von Willebrand disease. Blood. 1987;69(6):1691–1695. Davies JA, Collins PW, Hathaway LS, Bowen DJ. Effect of von Willebrand factor Y/C1584 on in vivo protein level and function and interaction with ABO blood group. Blood. 2007;109(7):2840– 2846. McKinnon TA, Chion AC, Millington AJ, Lane DA, Laffan MA. N-linked glycosylation of VWF modulates its interaction with ADAMTS13. Blood. 2008;111(6):3042–3049. Kokame K, Matsumoto M, Soejima K, et al. Mutations and common polymorphisms in ADAMTS13 gene responsible for von Willebrand factor-cleaving protease activity. Proc Natl Acad Sci USA. 2002;99(18):11902–11907. Ruan CH. The frequency of P475S polymorphism in von Willebrand factor-cleaving protease in the Chinese population and its relevance to arterial thrombotic disorders. Thromb Haemost. 2004;91:1258–1259. Bongers TN, De Maat MPM, Dippel DWJ, Uitterlinden AG, Leebeek FWG. Absence of Pro475Ser polymorphism in ADAMTS13 in Caucasians. J Thromb Haemost. 2005;3:805. Plaimauer B, Fuhrmann J, Mohr G, et al. Modulation of ADAMTS13 secretion and specific activity by a combination of common amino acid polymorphisms and a missense mutation. Blood. 2006;107(1):118–125. Bowen DJ, Collins PW. An amino acid polymorphism in von Willebrand factor correlates with increased susceptibility to proteolysis by ADAMTS13. Blood. 2004;103(3):941–947. Davies JA, Bowen DJ. The association between the L1565 variant of von Willebrand factor and susceptibility to proteolysis by ADAMTS13. Haematologica. 2007;92(2):240–243. O’Brien LA, James PD, Othman M, et al. Association of Hemophilia Clinic Directors of Canada. Founder von Willebrand factor haplotype associated with type 1 von Willebrand disease. Blood. 2003;102(2):549–557. Sutherland JJ, O’Brien LA, Lillicrap D, Weaver DF. Molecular modeling of the von Willebrand factor A2 Domain and the effects of associated type 2A von Willebrand disease mutations. J Mol Model. 2004;10(4):259–270. Keeney S, Grundy P, Collins PW, Bowen DJ. C1584 in von Willebrand factor is necessary for enhanced proteolysis by ADAMTS13 in vitro. Haemophilia. 2007;13(4):405–408. Bowen DJ, Collins PW, Lester W, et al. UK Haemophilia Centre Doctors’ Organization. The prevalence of the cysteine1584 variant of von Willebrand factor is increased in type 1 von Willebrand disease: co-segregation with increased susceptibility to ADAMTS13 proteolysis but not clinical phenotype. Br J Haematol. 2005;128(6):830–836. Davies JA, Collins PW, Hathaway LS, Bowen DJ. von Willebrand factor: evidence for variable clearance in vivo according to Y/C1584 phenotype and ABO blood group. J Thromb Haemost. 2008;6(1):97–103. Davies JA, Collins PW, Hathaway LS, Bowen DJ. C1584: effect on von Willebrand factor proteolysis and von Willebrand factor antigen levels. Acta Haematol. 2009;121(2–3):98–101. Pruss CM, Notley CR, Hegadorn CA, O’Brien LA, Lillicrap D. ADAMTS13 cleavage efficiency is altered by mutagenic and, to a

JCD 2009; 1:(1). OCTOBER 2009


Journal of Coagulation Disorders

79.

80.

81.

82.

83.

84.

85. 86.

87.

88.

89.

lesser extent, polymorphic sequence changes in the A1 and A2 domains of von Willebrand factor. Br J Haematol. 2008;143(4): 552–558. Huizinga EG, Tsuji S, Romijn RA, et al. Structures of glycoprotein Ibalpha and its complex with von Willebrand factor A1 domain. Science. 2002;297(5584):1176–1179. Nishio K, Anderson PJ, Zheng XL, Sadler JE. Binding of platelet glycoprotein Ibalpha to von Willebrand factor domain A1 stimulates the cleavage of the adjacent domain A2 by ADAMTS13. Proc Natl Acad Sci USA. 2004;101(29):10578–10583. Tsai HM, Lian ECY. Antibodies to von Willebrand factorcleaving protease in acute thrombotic thrombocytopenic purpura. N Engl J Med. 1998;339:1585–1594. Furlan M, Robles R, Galbusera M, et al. von Willebrand factorcleaving protease in thrombotic thrombocytopenic purpura and the hemolytic-uremic syndrome. N Engl J Med. 1998;339(22): 1578–1584. Klaus C, Plaimauer B, Studt JD, et al. Epitope mapping of ADAMTS13 autoantibodies in acquired thrombotic thrombocytopenic purpura. Blood. 2004;103(12):4514–4519. Moake JL, Rudy CK, Troll JH, et al. Unusually large plasma factor VIII:von Willebrand factor multimers in chronic relapsing thrombotic thrombocytopenic purpura. N Engl J Med. 1982; 307(23):1432–1435. Tsai HM. Current concepts in thrombotic thrombocytopenic purpura. Annu Rev Med. 2006;57:419–436. Lyons SE, Bruck ME, Bowie EJ, Ginsburg D. Impaired intracellular transport produced by a subset of type IIA von Willebrand disease mutations. J Biol Chem. 1992;267(7):4424– 4430. Tsai HM, Sussman II, Ginsburg D, Lankhof H, Sixma JJ, Nagel RL. Proteolytic cleavage of recombinant type 2A von Willebrand factor mutants R834W and R834Q: inhibition by doxycycline and by monoclonal antibody VP-1. Blood. 1997;89(6):1954–1962. O’Brien LA, Sutherland JJ, Hegadorn C, et al. A novel type 2A (Group II) von Willebrand disease mutation (L1503Q) associated with loss of the highest molecular weight von Willebrand factor multimers. J Thromb Haemost. 2004;2(7):1135–1142. Lyons SE, Cooney KA, Bockenstedt P, Ginsburg D. Characterization of Leu777Pro and Ile865Thr type IIA von Willebrand disease mutations. Blood. 1994;83(6):1551–1557.

JCD 2009; 1:(1). OCTOBER 2009

84

www.slm-hematology.com


JOURNAL OF COAGULATION DISORDERS

REVIEW ARTICLE

Central Venous Access Devices (CVAD) for Pediatric Patients with Hemophilia: A Review Riten Kumar1, Rajiv K Pruthi2,3 and Vilmarie Rodriguez1,3 Affiliations: 1Division of Pediatric Hematology – Oncology, Mayo Clinic, Rochester, MN, USA; 2Division of Hematology – Oncology, Mayo Clinic, Rochester, MN, USA and 3Mayo Clinic Comprehensive Hemophilia Center, Rochester, MN, USA Submission date: 22nd July 2009, Revision date: 10th September 2009, Acceptance date: 17th September 2009

A B S T R A C T Hemophilia A and B are X-linked recessive bleeding disorders caused by deficiencies of blood coagulation factors VIII and IX, respectively. Management of hemophilia essentially entails the use of replacement factors which can be used on demand or on a prophylactic basis. Venous access is required irrespective of whether a patient receives prophylactic or on-demand therapy. While central venous access devices (CVAD) are frequently used in pediatric hemophilia patients to facilitate repeated factor infusions, especially in young patients on prophylaxis or immune tolerance therapy for inhibitors, their use is associated with an increased risk of infections and/or thrombosis. The need for CVAD should be individualized with each patient. Patient and parental education about potential complications, use and care of CVAD are extremely important to ensure proper care. Keywords: central venous access devices, hemophilia, complications, prophylaxis Correspondence: Vilmarie Rodriguez, MD, Mayo Clinic Comprehensive Hemophilia Center, Department of Pediatrics and Adolescent Medicine, 200 First Street SW, Rochester, MN 55905. Tel: 507-284-2695; fax: 507-284-0727; e-mail: Rodriguez.vilmarie@mayo.edu

the aim of prophylaxis is to maintain a factor level of above 1%, thereby converting the severe phenotype into a moderate one [6]. Observational and prospective studies demonstrating the benefits of prophylaxis were recently confirmed in a landmark prospective randomized trial [6–9, 10–12]. These studies have led to recommendations of initiation of prophylaxis beginning at age 1 to 2 years with the eventual goal of prophylactic dosing every other day for HA and every third day for HB [5].

INTRODUCTION Hemophilia A (HA) and B (HB) are X-linked recessive bleeding disorders caused by deficiencies of blood coagulation factors VIII (FVIII) and IX (FIX), respectively. Although Von Willebrand disease has a higher prevalence, the hemophilias are more commonly recognized; HA has an incidence of about 1 in 5000 men and HB has an incidence of approximately 1 in 30000 [1, 2]. Based on coagulation factor activity, hemophilia is classified into mild (FVIII/FIX 6 to 40%), moderate (FVIII/FIX 1 to 5%) and severe (FVIII/FIX ,1%), with approximate prevalences of 31%, 26% and 43%, respectively [2, 3]. Long-term management of hemophilia consists of prevention and treatment of bleeding. Hemarthrosis, which occurs in more that 90% of patients with severe hemophilia, may eventually result in irreversible joint damage [4]. Although treatment recommendations are to initiate prophylaxis early in childhood to prevent long-term sequelae associated with repetitive bleeds especially in the joints, some patients and families opt to utilize on-demand therapy as an approach to treat acute bleeds [5].

CENTRAL VENOUS ACCESS DEVICES (CVAD) IN HEMOPHILIA Irrespective of whether a patient is on prophylaxis or on-demand therapy, adequate venous access is required. While peripheral veins continue to remain the route of choice, CVAD are more frequently used in young patients with hemophilia [13–16]. Central venous access devices facilitate safe, effective and frequent administration of clotting factor concentrates. A recent report of 70 patients with severe hemophilia confirmed that CVAD allowed both earlier initiation of prophylactic therapy (P,0.001) and home treatment (P50.001) [17]. In addition, home infusions have been shown to reduce hospitalizations for bleeding complications [18]. CVAD are often needed in

The rationale for prophylactic infusions of factor concentrate (prophylaxis) is based on the observation that patients with moderate hemophilia rarely experience spontaneous hemorrhage or arthropathy. Thus, JCD 2009; 1:(1). OCTOBER 2009

85

www.slm-hematology.com


Journal of Coagulation Disorders

The risks and benefits of CVAD should be discussed with the families before placement. Caregivers should be allowed to practice accessing the port on CVAD models, and it should be ensured that they understand how to care for the site, maintain patency and recognize signs of infection. A multidisciplinary approach consisting of the surgeon, the hemophilia center and the anesthesia team is important for optimal outcomes. Each hemophilia center has guidelines regarding factor concentrate infusion (bolus vs continuous infusion) and duration of factor concentrate therapy following surgical placement of CVAD. Various institutional protocols have been reported addressing the need for pre- and postoperative factor concentrate replacement [13, 17, 20, 28, 30, 32–34]. All of these regimens consist of infusing factor concentrate to a level of 100% followed by a continuous or bolus administration in order to maintain a level of 50% to 100% for 48 to 72 hours and decreasing the dose thereafter to keep levels 30% to 75% for 1 to 5 days following port placement. For patients with inhibitors, activated prothrombin complex concentrates or recombinant activated factor seven are typically used [13]. The approach to timing of accessing the port after placement varies. Some centers avoid accessing the port for fear of introducing infections, whereas others immediately access the port [20].

hemophilia patients with inhibitors on immune tolerance therapy (ITT) given the requirement for daily infusions of factor concentrate. The use of CVAD is associated with potential complications such as infections and thrombosis which increase morbidity and might complicate long-term management [13–34]. Thus, a careful analysis of the risk–benefit ratio and an individualized case-by-case approach to the use of CVAD are recommended [13]. While medical factors such as young age, poor peripheral access and the need for multiple infusions must be considered, psychosocial issues such as the child’s poor tolerance to multiple venipunctures and the caregivers’ capability and commitment to aseptic technique also warrant special attention [13, 14]. Two types of CVAD are commonly used in hemophilia patients: implantable venous access systems (ports) which are subcutaneous devices such as Port-A-CathH (Deltec Inc., St. Paul, MN, USA) and external catheters such as HickmanH and BroviacH (Bard Access Systems, Salt Lake City, UT, USA). Ports are the most commonly used CVAD in patients with hemophilia. A port consists of a subcutaneous reservoir with a silicon septum that is surgically implanted in the anterior chest wall and usually appears as a bump under the skin [13, 15]. A radio-opaque catheter is attached to the reservoir, and its distal end extends to the junction of the superior vena cava and the right atrium. A lower risk for infection has been reported in recipients of fully implanted CVAD compared to those with external CVAD (incidence risk ratio (IRR) 0.31; confidence interval (CI) 0.12–0.86) [16]. Ports are discrete and do not require the use of special dressings during swimming and bathing. They are also preferred in smaller children who are at higher risk to dislodge an external CVAD [15]. One possible disadvantage of ports is the fact that the needle punctures the skin to gain access to the subcutaneous reservoir; this may lead to patient discomfort, anxiety and skin breakdown.

COMPLICATIONS OF CVAD Infection Infection is one of the most common complications of CVAD in hemophilia patients (Table 1 and Figure 1). Infectious complications can be categorized as local (exit site infection and tunnel infections) or systemic (bacteremia and septicemia). Exit site infections are superficial infections which extend less than 2 cm from the exit site of the port from the skin whereas tunnel or port-pocket infections extend along the entire subcutaneous course of the line (depth of involvement is greater than 2 cm) [15]. Serious complications such as endocarditis, septic arthritis and death secondary to sepsis while rare, have also been reported in the literature [22–25].

External catheters exit the skin through a subcutaneous tunnel. The proximal end is equipped with a Dacron cuff that anchors the catheter to the subcutaneous tunnel [15]. The distal end is again at the junction of the inferior vena cava and the right atrium. While patient discomfort from repeated injections is lower in external CVAD, there is an increased risk of accidental displacement and some reports suggest an increase risk for infections [15]. Recently, a new percutaneous port system, PercusealH (Percuseal Medical Systems AB, Husquvarna, Sweden) was reported in hemophilia patients. This device differs from a subcutaneous port in that the proximal portion protrudes above the skin, thereby allowing access without penetration of the skin. Initial data, however, show a high risk of infection in patients with hemophilia and von Willebrand’s disease and its use in its current form is not recommended [19]. JCD 2009; 1:(1). OCTOBER 2009

Catheter associated infections in hemophilia patients have varied from 0.14/1000 CVAD days to 3.4/1000 CVAD days in different studies [17, 20, 26–34] (Table 1) with a pooled estimate of 0.66/1000 catheter days and the pooled time to the first infection of 295 days (CI, 181–479 days) in a recent meta-analysis [16]. The true incidence of infections cannot be determined as denominators were not known and some patients were counted more than once or had more than one CVAD. Skin micro-organisms such as Staphylococcus epidermis, Staphylococcus aureus, and Streptococcus were found to be the most common culprits. Factors independently predictive of infections included age less than 6 years, presence of inhibitors and external 86

www.slm-hematology.com


Table 1. Frequency of Infections in Patients with Hemophilia Study

Year

Number of patients

Rate of infections per 1000 CVAD day

Rate of infections in patients with inhibitors

Titapiwatanakun et al [20]

2009

18

0.22

0.74

Van Djik et al [17]

2004

23

0.48

3.1

Domm et al [27]

2003

22

0.3

NA

Tarantino et al [26]

2003

59

0.45

0.66

Bollard et al [28]

2000

14

3.4

NA

McMahon et al [29]

2000

58

1.6

4.3

Ljung et al [30] Miller et al [31]

1998 1998

53 45

0.19 0.14

NA NA

Santagostino et al [32]

1998

15

0.42

NA

Perkins et al [33]

1997

35

1.2 (central); 0.7 (peripheral)

NA

Blanchette et al [34]

1996

19

0.7

NA

catheters [16]. Patients older than 6 years of age were only 46% as likely to develop infections compared to younger children. Similar results were observed by Tarantino et al, where older patients (greater than median age of 2.7 years) at CVAD placement had a significantly lower rate of infection [26]. The incidence of infections was found to be 67% higher in patients with inhibitors [13]. This increased risk might be related to the occurrence of bleeds around the port site while accessing the port [30]. Although the frequency with which inhibitor patients need to access ports has also been felt to be responsible for the high rate of infections, available data do not support such a hypothesis and in some reports, the number of punctures or frequency of CVAD use or access was not found to be an independent risk factor for infections in patients with CVAD [30, 33, 35].

hemophilia patients before accessing the port [33]. Simply wiping off the cream may leave behind a lipid layer, which can act as a nidus for infection. Scrubbing off the topical anesthetic with soap and water and subsequent cleaning with chlorhexidine is recommended. Further research is needed to confirm this observation. Management of established infections varies in each treating institution. Treatment of uncomplicated CVAD infection without pocket or tunnel infection consists of standard systemic antibiotic with or without antibiotic lock therapy for 10 to 14 days [36, 37]. If the infection is appropriately cleared, there is no need for CVAD removal. In the presence of persistent bacteremia, CVAD may need to be removed in presence of persistent bacteremia for more than 48–72 hours despite appropriate antibiotic therapy. Fungal infection and recurrent infection with the same organism after completing a previous antibiotic course, and tunnel, pocket and skin erosions are the main infectious complications where catheter removal is indicated [36–38].

Given that the most common organisms implicated in CVAD infections are bacteria commonly found on the skin surface, educating caregivers to adhere to aseptic precautions to reduce such infections is critical. There have been concerns regarding the use of topical lidocaine–prilocaine 2.5%–2.5% (eg, EMLA cream) in

Another potential risk factor for CVAD infection is thrombus formation. It has been described previously

Figure 1. Indications of removal of CVAD

www.slm-hematology.com

87

JCD 2009; 1:(1). OCTOBER 2009


Journal of Coagulation Disorders

that the formation of a fibrin sheath may increase the risk of infection in CVAD [39–41]. The use of prophylactic fibrinolytic agents such as tissue plasminogen activator (tPA; CathfloH, Genentech, Inc., San Francisco, CA, USA) in cancer patients with CVAD has been shown to reduce central line occlusion and infection [42, 43]. Tissue plasminogen activator is a recombinant serine protease glycoprotein that lyses plasmin-bound fibrin. The use of tPA in hemophilia patients with CVAD suggests that it might decrease the frequency of infections [44, 45]. In a pilot study of 5 patients with hemophilia and CVAD, the use of tPA reduced CVAD infections to 3.95 infections/1000 catheter days compared to 8.75 infections/1000 catheter days for patients without tPA [44]. In another study describing monthly tPA infusions for the prevention of infections in 18 hemophilia patients with CVAD, only one infection was reported, which was not related to CVAD (pneumonia) [tPA use 0.5 to 6.8 years, median 3 years] [45]. No bleeding complications were reported in these two studies. The use, efficacy and safety of tPA in the prevention of CVAD infection in hemophilia patients needs to be further investigated in a larger randomized study.

thrombosis. The limitations of these studies were small size, and lack of a uniform and consistent radiological approach for evaluation of thrombosis (Doppler ultrasound examination versus venogram). More recent studies, however, have reported a relatively higher incidence of thrombotic complications (Table 2). Journeycake et al prospectively evaluated 15 hemophilia patients, who had a CVAD in place for more than a year, with venograms [47]. Nine of these patients had been previously reported by Medeiros et al in 1998. Eight of the 15 patients (53%) were found to have abnormal venograms consistent with deep venous thrombosis (DVT). Three of these 8 patients had physical signs of thrombosis (swelling of affected arm, prominence of neck vessels) and 5 of the 8 reported catheter related problems. Interestingly, no patient with a catheter for less than 48 months had a DVT, whereas all of the patients with CVAD for more than 73 months had radiological evidence of thrombosis. Ettingshausen et al described 25 patients with tunneled subclavian CVAD [50]. Eight (32%) of these patients reported clinical problems, ranging from increased resistance during use of CVAD to the prominence of superficial veins over the ipsilateral shoulders. All patients were found to have a DVT of the neck veins on contrast venography and 5/7 patients tested for thrombophilia were found to have a prothrombotic risk factor (2-Factor V 1691 G.A mutation, 3-MTHFR 677 C.T genotype) – thereby suggesting multifactorial causation of thrombosis. An even higher prevalence of thrombosis was found in a prospective study of 16 children with hemophilia and CVAD. In this report, children underwent a Doppler ultrasound and bilateral upper arm venograms, 2 years apart to evaluate the patency of the upper venous systems [51]. At the time of the first study, 11/16 (69%) had evidence of DVT, while 13/16 (81%) had radiological evidence of an upper extremity DVT at the 2 year follow-up visit. A meta-analysis of CVAD in hemophilia patients reported 55 episodes of thrombosis. In 19 studies with a cumulative pooled estimate of 0.56/ 1000 CVAD days [CI 0.016–0.196 per 1000 CVAD days] [16]. Neither age, nor the presence of inhibitors were found to be independent risk factors for thrombosis.

Thrombosis While a well recognized complication of CVAD in oncology patients, thrombosis has historically not been of great concern in patients with hemophilia due to their congenital lack of clotting factors – in fact, it might seem almost paradoxical for patients with hemophilia to develop thrombosis [46, 47]. The reported incidence of thrombotic complications in hemophilia patients with CVAD has varied markedly in the literature [30–34, 48]. Ljung et al, for instance, did not see any clinical signs of thrombosis in 53 patients with severe and moderate hemophilia despite 1578 months of cumulative follow-up [30]. However, not all of these studies report on performing surveillance venograms of the upper extremities or Doppler sonograms of the neck. Medeiros et al reported 19 patients with moderate to severe hemophilia, who had CVAD in place for more than 6 months [49]. None of the patients had physical stigmata of thrombosis. Prospective venograms were performed on 13 of the 19 patients, and only one patient was found to have a nonocclusive thrombus at the same side of the catheter. It was concluded from this study that hemophiliacs do not have a significant risk of

There is a discrepancy in the incidence of thrombosis between different generations of studies. Many of the studies reported their own institutional experience and their sample size was small. Some of the more

Table 2. Incidence of Thrombosis in Hemophilia Patients Study

Year

Patients (n)

Median age of line placement (years)

Thrombosis (%)

Inhibitors

Mean duration (months)

Journeycake et al [47]

2001

15

4

53

3/15 5 20%

57.5

Ettingshausen et al [50]

2002

25

32

5/25 5 20%

43.3

Domm et al [27]

2003

22

5.9

18

9/22 5 41%

33.5

Price et al [51]

2004

16

1

69 (first evaluation) 89 (2-year follow-up)

5/16 5 31% 6/16 5 38%

– 76.8 [median]

JCD 2009; 1:(1). OCTOBER 2009

88

www.slm-hematology.com


(median age 2.7 years) [60]. Thirty-five AVF (81%) achieved maturation after a median of 58 days after placement and were used for a median of 5 years (range 0.4–8.5 years). Nine AVF had to be surgically dismantled due to transition to peripheral veins (4), aneurysmatic dilatation (3), limb hypertrophy (1) and AVF overflow (1). Age at AVF creation was younger in patients who lost AVF patency and aneurysms were more frequent in children who were on daily treatment. The use of AVF allowed long-term prophylaxis in 11 children and uninterrupted immune tolerance in 18 children [60]. No infection complications were reported in their AVF series [60, 61].

recent studies are performing routine venograms on asymptomatic patients; it is known from adult orthopedic data that the incidence of symptomatic deep venous thrombosis is much lower than that diagnosed by surveillance venograms [52, 53]. The duration of catheter placement can also be associated with thrombus formation – the mean duration of catheter placement has ranged from 33.5 months to 76.8 months in recent studies. The study by Journeycake et al suggested that the presence of a catheter for more than 48 months increases the risk of thrombogenesis [47]. Other factors such as younger children with a lower blood vessel to catheter diameter, frequent use of bypassing factor concentrates, higher doses of factor concentrates for immune tolerance and coinheritance of prothrombotic risk factors may predispose hemophilia patients with CVAD to develop thrombotic events [50, 51, 54].

McCarthy et al reported their experience with AVF in 9 patients with hemophilia. Patient median age was 5.5 years (range 1 to 27 years). None of the patients developed hematomas postoperatively. Of the mature AVF, 100% were patent at 1 year (7/7), 80% at 3 years (4/5) and 75% at 4 years (3/4) with a median follow-up of 22 months [62]. In summary, AVF appear to be an alternative for long-term venous access in patients with hemophilia although more data are needed before any recommendations can be made. Complications such as infections have not been reported so far in contrast to CVAD [60–62].

In summary, thrombosis in hemophilia patients is probably multifactorial. Routine screening for thrombophilia risk before CVAD in patients with hemophilia is not recommended unless patients have experienced a previous thrombosis unrelated to the use of CVAD. The duration of CVAD placement, age of placement, presence of inhibitors and coinheritance of prothrombotic conditions, may all predispose to thrombosis. It has been suggested that these clots develop slowly, thereby allowing collaterals to develop, masking the signs of acute DVT [51]. The significance of asymptomatic thromboses which are diagnosed on routine screening is not clearly understood and their long-term complications not reported. Bilateral venography and Doppler ultrasound examination may be needed to evaluate DVT in the upper venous system of a child [55]. Venography, although invasive, should be performed if DVT is suspected and other imaging modalities are negative. If a CVAD has been in place for more than 4 years, venograms and ultrasound are recommended to evaluate thrombosis even in the absence of symptoms [13]. A multicenter, prospective study to document the incidence of thrombosis in hemophilia patients with CVAD, with a follow-up reporting the clinical sequelae of asymptomatic thrombosis, is needed before screening recommendations guidelines can be made.

SUMMARY Even though peripheral veins are preferred, CVAD continue to remain a reliable means of venous access in children with hemophilia. These devices are particularly relevant in small children and allow the early initiation of both prophylactic therapy and immune tolerance therapy at home. They can also improve the quality of life for both patients and their caregivers. While infection and thrombosis continue to remain important complications of CVAD placement – a careful screening of patients and thorough education of caregivers with strict adherence of aseptic technique will help ensure a good outcome. The alternative use of AVF for venous access can be offered to a patient. However, surgical expertise, patient size and age and a comprehensive hemophilia center care are needed in order to provide safe care for these patients. Patients with CVAD should be followed up at a comprehensive hemophilia center, by a staff that is well versed with the common complications associated with these devices. Also patient and parent education need to be reinforced at each comprehensive visit.

ARTERIO-VENOUS FISTULAS FOR LONG-TERM VENOUS ACCESS IN HEMOPHILIA The arteriovenous fistula (AVF) is one of the preferred modalities for venous access in patients undergoing dialysis. It has several favorable qualities such as autologous angio-access leading to lower risk of infections (1% to 4% per year in patients undergoing dialysis), and longer duration (up to 52% are still in use for up to 6.5 years after placement) [56–59].

Disclosures: The authors have no financial interests to disclose related to the contents of this article.

REFERENCES 1. Werner EJ, Broxson EH, Tucker EL, Giroux DS, Shults J, Abshire TC. Prevalence of von Willebrand disease in children: a multiethnic study. J Pediatr. 1993;123(6):893–898. 2. Soucie JM, Evatt B, Jackson D. Occurrence of hemophilia in the United States. The Hemophilia Surveillance System Project Investigators. Am J Hematol. 1998;59(4):288–294.

An Italian group has reported the largest experience of AVF in patients with hemophilia [60, 61]. Fortythree AVF were placed in 38 children with hemophilia www.slm-hematology.com

89

JCD 2009; 1:(1). OCTOBER 2009


Journal of Coagulation Disorders

22. Hothi DK, Kelsall W, Baglin T, Williams DM. Bacterial endocarditis in a child with haemophilia B: risks of central venous catheters. Haemophilia. 2001;7(5):507–510. 23. Titapiwatanakun R, Pruthi RK, Schmidt K, Slaby JA, Rodriguez V. Bacterial endocarditis and septic arthritis in a patient with severe hemophilia A: a case report. J Pediatr Hematol Oncol. 2009;31(5):344–345. 24. Li CH, Ou Y, Lee AC, So KT. Septic arthritis in hemophilia with central venous catheter: a case report. Pediatr Hematol Oncol. 2000;17(2):187–189. 25. Crary SE, Buchanan GR, Journeycake JM. Fatal venouscatheter-related infection in haemophilia. Haemophilia. 2006;12: 183–186. 26. Tarantino MD, Lail A, Donfield SM, et al. Surveillance of infectious complications associated with central venous access devices in children with haemophilia. Haemophilia. 2003;9(5): 588–592. 27. Domm JA, Hudson MG, Janco RL. Complications of central venous access devices in paediatric haemophilia patients. Haemophilia. 2003;9(1):50–56. 28. Bollard CM, Teague LR, Berry EW, Ockelford PA. The use of central venous catheters (portacaths) in children with haemophilia. Haemophilia. 2000;6(2):66–70. 29. McMahon C, Smith J, Khair K, Liesner R, Hann IM, Smith OP. Central venous access devices in children with congenital coagulation disorders: complications and long-term outcome. Br J Haematol. 2000;110(2):461–468. 30. Ljung R, van den Berg M, Petrini P, et al. Port-A-Cath usage in children with haemophilia: experience of 53 cases. Acta Paediatr. 1998;87(10):1051–1054. 31. Miller K, Buchanan GR, Zappa S, et al. Implantable venous access devices in children with hemophilia: a report of low infection rates. J Pediatr. 1998;132(6):934–938. 32. Santagostino E, Gringeri A, Muca-Perja M, Mannucci PM. A prospective clinical trial of implantable central venous access in children with haemophilia. Br J Haematol. 1998;102(5):1224–1228. 33. Perkins JL, Johnson VA, Osip JM, et al. The use of implantable venous access devices (IVADs) in children with hemophilia. J Pediatr Hematol Oncol. 1997;19(4):339–344. 34. Blanchette VS, al-Musa A, Stain AM, Filler RM, Ingram J. Central venous access catheters in children with haemophilia. Blood Coagul Fibrinolysis. 1996;7(suppl 1):S39–S44. 35. Hacker MR, Page JH, Shapiro AD, Rich-Edwards JW, MancoJohnson MJ. Central venous access device infections in children with hemophilia: a comparison of prophylaxis and episodic therapy. J Pediatr Hematol Oncol. 2007;29(7):458–464. 36. Mermel LA, Farr BM, Sherertz RJ, et al. Guidelines for the management of intravascular catheter-related infections. J Intraven Nurs. 2001;24:180–205. 37. Berrington A, Gould FK. Use of antibiotic locks to treat colonize central venous catheters. J Antimicrob Chemother. 2001;48:597– 603. 38. Whitman ED. Complications associated with the use of central venous access devices. Curr Probl Surg. 1996;33:309–378. 39. Raad II, Luna M, Khalil S-A, et al. The relationship between the thrombotic and infectious complication of central venous catheters. JAMA. 1994;271:1014–1016. 40. Timsit JF, Farkas JC, Boer JM, et al. Central vein catheterrelated thrombosis in intensive care patients: Incidence, risk factors and relationship with catheter related sepsis. Chest. 1998; 114:207–213. 41. Mehall JR, Saltzman DA, Jackson RJ, et al. Fibrin sheath enhances central venous catheter infection. Crit Care Med. 2002; 30:908–912. 42. Dillon PW, Jones GR, Bagnall-Reeb HA, Buckley JD, Wiener ES, Haase GM; Children’s Oncology Group. Prophylactic urokinase in the management of long-term venous access devices in children: A Children’s Oncology Group study. J Clin Oncol. 2004;22:2718–2723.

3. White GC 2nd, Rosendaal F, Aledort LM, Lusher JM, Rothschild C, Ingerslev J; Factor VIII and Factor IX SubcommitteeDefinitions in hemophilia. Recommendation of the scientific subcommittee on factor VIII and factor IX of the scientific and standardization committee of the International Society on Thrombosis and Haemostasis. Thromb Haemost. 2001; 85(3):560. 4. Aledort LM, Haschmeyer RH, Pettersson H. A longitudinal study of orthopaedic outcomes for severe factor-VIII-deficient haemophiliacs. The Orthopaedic Outcome Study Group. J Intern Med. 1994;236(4):391–399. 5. Berntorp E, Astermark J, Bjorkman S, et al. Consensus perspectives on prophylactic therapy for haemophilia: summary statement. Haemophilia. 2003;9(suppl 1):1–4. 6. Nilsson IM, Berntorp E, Lofqvist T, Pettersson H. Twenty-five years’ experience of prophylactic treatment in severe haemophilia A and B. J Intern Med. 1992;232(1):25–32. 7. Lofqvist T, Nilsson IM, Berntorp E, Pettersson H. Haemophilia prophylaxis in young patients – a long-term follow-up. J Intern Med. 1997;241(5):395–400. 8. Petrini P, Lindvall N, Egberg N, Blomback M. Prophylaxis with factor concentrates in preventing hemophilic arthropathy. Am J Pediatr Hematol Oncol. 1991;13(3):280–287. 9. Panicker J, Warrier I, Thomas R, Lusher JM. The overall effectiveness of prophylaxis in severe haemophilia. Haemophilia. 2003;9(3):272–278. 10. Gringeri A. Prospective controlled studies on prophylaxis: an Italian approach. Haemophilia. 2003;9(suppl 1):38–42; discussion 43. 11. Feldman BM, Pai M, Rivard GE, et al. Tailored prophylaxis in severe hemophilia A: interim results from the first 5 years of the Canadian Hemophilia Primary Prophylaxis Study. J Thromb Haemost. 2006;4(6):1228–1236. 12. Manco-Johnson MJ, Abshire TC, Shapiro AD, et al. Prophylaxis versus episodic treatment to prevent joint disease in boys with severe hemophilia. N Engl J Med. 2007;357(6):535–544. 13. Ewenstein BM, Valentino LA, Journeycake JM, et al. Consensus recommendations for use of central venous access devices in haemophilia. Haemophilia. 2004;10(5):629–648. 14. Ljung R. The risk associated with indwelling catheters in children with haemophilia. Br J Haematol. 2007;138(5):580–586. 15. Komvilaisak P, Connolly B, Naqvi A, Blanchette V. Overview of the use of implantable venous access devices in the management of children with inherited bleeding disorders. Haemophilia. 2006; 12(suppl 6):87–93. 16. Valentino LA, Ewenstein B, Navickis RJ, Wilkes MM. Central venous access devices in haemophilia. Haemophilia. 2004;10(2): 134–146. 17. Van Dijk K, Van Der Bom JG, Bax KN, Van Der Zee DC, Van Den Berg MH. Use of implantable venous access devices in children with severe hemophilia: benefits and burden. Haematologica. 2004;89(2):189–194. 18. Soucie JM, Symons J 4 th, Evatt B, Brettler D, Huszti H, Linden J; Hemophilia Surveillance System Project Investigators. Homebased factor infusion therapy and hospitalization for bleeding complications among males with haemophilia. Haemophilia. 2001;7(2):198–206. 19. Berntorp E, Piitulainen E, Lethagen S, Mattson E, Tengborn L, Matzsch T. Experience with a new percutaneous port system, Percuseal, for intravenous injection in patients with haemophilia, von Willebrand disease and severe alpha1-antitrypsin deficiency. Haemophilia. 2003;9(2):173–178. 20. Titapiwatanakun R, Moir C, Pruthi RK, Stavlo PL, Schmidt KA, Rodriguez V. Central venous access devices for paediatric patients with haemophilia: a single-institution experience. Haemophilia. 2009;15(1):168–174. 21. Male C, Chait P, Andrew M, Hanna K, Julian J, Mitchell L. Central venous line-related thrombosis in children: association with central venous line location and insertion technique. Blood. 2003;101(11):4273–4278.

JCD 2009; 1:(1). OCTOBER 2009

90

www.slm-hematology.com


43. Ray CE, Shenoy S, McCarthy PL, et al. Weekly prophylactic urokinase installation in tunneled central venous access devices. J Vasc Interv Radiol. 1999;10:1330–1334. 44. Dunn AL, Abshire TC. Recombinant tissue plasminogen activator may reduce frequency of central venous access device infection in hemophilia patients undergoing immune tolerance therapy. Pediatr Blood Cancer. 2008;50:627–629. 45. Ragni MV, Journeycake JM, Brambilla DJ. Tissue plasminogen activator to prevent central venous access device infections: a systematic review of central venous access catheter thrombosis, infection and thromboprophylaxis. Haemophilia. 2008;14:30–38. 46. Glaser DW, Medeiros D, Rollins N, Buchanan GR. Catheterrelated thrombosis in children with cancer. J Pediatr. 2001; 138(2):255–259. 47. Journeycake JM, Quinn CT, Miller KL, Zajac JL, Buchanan GR. Catheter-related deep venous thrombosis in children with hemophilia. Blood. 2001;98(6):1727–1731. 48. Collins PW, Khair KS, Liesner R, Hann IM. Complications experienced with central venous catheters in children with congenital bleeding disorders. Br J Haematol. 1997;99(1):206–208. 49. Medeiros D, Miller KL, Rollins NK, Buchanan GR. Contrast venography in young haemophiliacs with implantable central venous access devices. Haemophilia. 1998;4(1):10–15. 50. Ettingshausen CE, Kurnik K, Schobess R, et al. Catheter-related thrombosis in children with hemophilia A: evidence of a multifactorial disease. Blood. 2002;99(4):1499–1500. 51. Price VE, Carcao M, Connolly B, et al. A prospective, longitudinal study of central venous catheter-related deep venous thrombosis in boys with hemophilia. J Thromb Haemost. 2004; 2(5):737–742. 52. Eriksson BI, Bauer KA, Lassen MR, Turpie AG. Fondaparinux compared with enoxaparin for the prevention of venous thromboembolism after hip-fracture surgery. N Engl J Med. 2001;345(18):1298–1304. 53. Bauer KA, Eriksson BI, Lassen MR, Turpie AG. Fondaparinux compared with enoxaparin for the prevention of venous thromboembolism after elective major knee surgery. N Engl J Med. 2001;345(18):1305–1310. 54. Girolami A, Scandellari R, Zanon E, Sartori R, Girolami B. Noncatheter associated venous thrombosis in hemophilia A and B. A critical review of all reported cases. J Thromb Thrombolysis. 2006;21(3):279–284. 55. Male C, Chait P, Ginsberg JS, et al. Comparison of venography and ultrasound for the diagnosis of asymptomatic deep vein thrombosis in the upper body in children: results of the PARKAA study. Prophylactic antithrombin replacement in kids with ALL treated with asparaginase. Thromb Haemost. 2002;87:593–598. 56. Kherlakian GM, Roedersheimer LR, Arbaugh JJ, et al. Comparison of autologous fistula versus expanded polytetrafluoroethylene graft fistula for angioaccess in hemodialysis. Am J Surg. 1986;152:238–243. 57. Brittenger WP, Walker G, Twittenhoff WD, et al. Vascular access for hemodialysis in children. Pediatr Nephrol. 1997;11:87–95. 58. Bonalumi U, Civalleri P, Rovida S, et al. Nine years experience with end-to-end arteriovenous fistula at the anatomical snuffbox for maintenance hemodialysis. Br J Surg. 1982;69:486–488. 59. Winsett OE, Wolma FJ. Complication of vascular access for hemodialysis. Southern Med J. 1985;78:513–517. 60. Mancuso ME, Berardinelli L, Boretta C, et al. Improved treatment feasibility in children with hemophilia using arteriovenous fistulae: the results after seven years of follow-up. Haematologica. 2009;94:687–692. 61. Santagostino E, Gringeri A, Berardinelli L, et al. Long-term safety and feasibility of arteriovenous fistulae as vascular accesses in children with haemophilia: a prospective study. Br J Haematol. 2003;123:502–506. 62. McCarthy WJ, Valentino LA, Bonilla AS, et al. Arteriovenous fistula for long-term venous access for boys with hemophilia. J Vasc Surg. 2007;45:986–991.

www.slm-hematology.com

91

JCD 2009; 1:(1). OCTOBER 2009


JOURNAL OF COAGULATION DISORDERS

REVIEW ARTICLE

Pharmacoeconomic studies of bypassing agents in mild-tomoderate bleeding episodes in patients with hemophilia and inhibitors: a critical appraisal Erich V De Paula and Margareth C Ozelo Affiliation: Hemocentro da UNICAMP, Campinas, Brazil

A B S T R A C T The development of neutralizing antibodies against factor VIII and IX complicates the clinical management of patients with hemophilia, and adds significant costs to their treatment. For this reason, several formal economic analysis have been performed comparing treatment of mild and moderate bleeding episodes in these patients in an effort to determine whether one of the two available bypassing agents would be cost-effective compared with the other. Here we review and discuss the available evidence about the economic aspects of the alternatives for on-demand treatment of mild and moderate bleeding episodes for hemophilia patients with inhibitors. Keywords: Bleeds, hemophilia, inhibitor, economic, recombinant activated factor VII (rFVIIa), activated prothrombin complex concentrate (aPCC) Correspondence: Margareth Castro Ozelo, Rua Carlos Chagas, 480, Distrito Bara˜o Geraldo, PO. Box 6198, 13084-878 Campinas, Sa˜o Paulo, Brazil. Tel: +55-19-3521-8756; fax: +55-19-3521-8600; e-mail: margaret@unicamp.br

inhibitors (>5 BU/mL) require the use of bypassing agents to achieve hemostasis [5]. Currently, there are two bypassing agents available, the activated prothrombin complex concentrate (aPCC (FEIBAH, Baxter, Deerfield, IL)), and recombinant activated factor VII (rFVIIa (NovoSevenH, Novo Nordisk A/S, Bagsvaerd, Denmark)). The clinical efficacy of both products is difficult to predict; furthermore, there is no appropriate laboratory assay available for monitoring their haemostatic response [5]. These issues contribute to the complexity for bleeding episodes management in the presence of high-titer inhibitors.

INTRODUCTION Hemophilia is an X-linked, recessive, bleeding disorder caused by deficiency of factor VIII (hemophilia A) or factor IX (hemophilia B) that occurs at a rate of approximately 1 in 5000 and 1 in 30 000 males, respectively, in all ethnic groups [1]. The disease phenotype correlates with the residual activity of the factor VIII or IX and can be classified as severe (plasma coagulation factor levels ,1% of normal), moderate (plasma coagulation factor levels 2–5% of normal), or mild (plasma coagulation factor levels 6– 30% of normal). In the severe and moderate forms, the disease is characterized by bleeding episodes into the joint (hemarthroses), soft tissues, and muscles after minor trauma or even spontaneously [1]. These bleeding episodes when there is no neurological compromise or any other major complication are usually classified as mild–moderate bleeds.

This article reviews and discusses the formal economic analysis performed in different countries about the on-demand treatment of mild–moderate bleeding episodes for hemophilia patients with inhibitors, with emphasis on the quality of the evidence that support these studies. For the purpose of this review, mild and moderate bleeding episodes are defined as hemorrhagic events without significant blood loss, or severe pain, or neurological symptoms. In general, these episodes involve joints, soft tissues, or muscles, and usually can be treated in the home setting or in an outpatient clinic, and do not require hospitalization.

The primary treatment for hemophilia is replacement of the missing factor using recombinant or plasmaderived factor VIII or IX concentrates. However, patients may develop inhibitors (neutralizing antibodies) to the infused factor VIII or IX, which results in partial or complete lack of efficacy of clotting factor concentrates [2–4]. Patients with low-titer inhibitors (,5 Bethesda units (BU)/mL) can be treated with increased dosages of clotting factors concentrates. However, bleeding episodes associated with high-titer JCD 2009; 1:(1). OCTOBER 2009

OVERVIEW OF INHIBITOR ECONOMICS The management of patients with inhibitors is complicated not only by the clinical problems raised 93

www.slm-hematology.com


Journal of Coagulation Disorders

by the development of these antibodies, but also by the significant added expense of products used to treat bleeding episodes. It is generally accepted that the development of an inhibitor increases the overall costs of hemophilia treatment about 2- to 3-fold, even when compared with patients under primary prophylaxis but without inhibitors. The development of an inhibitor can raise the proportion of costs with replacement of clotting factor concentrates (CFCs) to up to 99% of the overall healthcare expenses [6], an increase that is explained by the higher cost of bypassing agents than most factor VIII concentrates, and by the lower efficacy of these agents, which usually require more doses to achieve adequate hemostasis. Furthermore, the unavailability of laboratory monitoring can lead practitioners to err on the side of overtreatment, further increasing treatment costs [7].

Mild and moderate bleedings, which represent the scope of the present review, are usually treated at home or on an outpatient basis. Therefore, the commonly accepted figures of 2- to 3-fold higher costs of treating inhibitor patients is not applicable to inhibitor patients with mild and moderate bleedings, for whom treatment costs are probably much closer to general hemophilia care than to overall inhibitor care.

KEY VARIABLES INFLUENCING THE RESULTS OF PHARMACOECONOMIC STUDIES OF INHIBITOR TREATMENT Within the past few years, several studies have been designed in an effort to define which of the available bypassing agents are more cost-effective (Table 1). By using pharmacoeconomic models, authors from different countries have formally modeled and examined the costs of treating mild and moderate bleeding episodes in inhibitor patients, primarily comparing rFVIIa and aPCC, the two primary treatment options [11–19]. These data have been extensively reviewed in the last 10 years [4, 9, 20, 21]. Rather than performing an additional systematic review of these results, we will address some characteristics of these studies that could limit their generalization. More specifically, we will discuss in more detail two key variables of any of these formal economic analysis: (1) the relative efficacy of each agent; and (2) the price.

In the last two decades, important economic aspects of the treatment of inhibitor patients have been demonstrated. It is now well known that patients with inhibitors do not use more CFCs than patients without these antibodies [7]. Perhaps one of the most important observation of inhibitor economics, also demonstrated by several authors, is that the higher overall costs of treating inhibitor patients can be largely attributed to a small number (1–2%) of patients that use huge amounts of bypassing agents compared with the remaining 98–99% [2, 3, 6]. The disproportionate consumption of CFCs by these patients skews the means to extremely high figures and increases the costs of treatment of this population, which would otherwise be very similar to that of patients without inhibitors [7]. This phenomenon seems to be universal and is referred to in the literature as the ‘‘outlier effect’’. It has also been demonstrated that costs of treating these ‘‘outliers’’ increase exponentially when inhibitor patients undergo surgical procedures, major bleeding episodes, or require second-line therapies [8, 9], situations that tend to occur during hospitalizations. In contrast, outpatient costs among inhibitor patients were not significantly different from those for non-inhibitor patients in a retrospective analysis in a North American center [10].

Relative Efficacy of rFVIIa and aPCC A key aspect of all formal economic evaluations about the treatment of bleedings in inhibitor patients is the fact that these studies are extremely sensitive to the relative efficacy of each agent, a value that is input in the early steps of any economic analysis. When analyzing these economic evaluations, it is not difficult to note that by changing the relative efficacy of each bypassing agent, one can switch the direction of the final conclusion. As a consequence, any imperfection of this set of data can jeopardize the validity of the entire analysis. Anyone who has ever treated a mild or moderate bleeding in a patient with hemophilia is aware

Table 1. Published Pharmacoeconomic Studies Comparing rFVIIa and aPCC for the Treatment of Mild–Moderate Bleedings in Inhibitor Patients Study

Country

Source of efficacy assumptions

Conclusion summary

Ekert et al [11]

Australia

Retrospective (medical records)/prospective

rFVIIa cost higher, but with q efficacy and q QOL

Knight et al [12]

UK

Literature (single arm observational studies)

rFVIIa cost lower than aPCC-based regimens

Joshi et al [13] Odeyemi and Guest [14]

USA UK

Literature (single arm observational studies) Literature/expert opinion

rFVIIa cost lower than aPCC-based regimens rFVIIa cost lower than aPCC-based regimens

Steen Carlsson et al [15]

USA/Sweden

Head-to-head trial (FENOC)

aPCC cost lower than rFVIIa

Dundar et al [16]

Turkey

Literature (single arm observational studies)

rFVIIa cost lower than aPCC-based regimens

Ozelo et al [17]

Brazil

Retrospective (medical records)/prospective

rFVIIa cost lower than aPCC-based regimens

Putnam et al [18]

USA

Expert opinion

aPCC cost lower than rFVIIa

You et al [19]

South Korea

Retrospective (medical records)/prospective

rFVIIa cost lower than aPCC-based regimens

rFVIIa, recombinant activated factor VII; aPCC, activated prothrombin complex concentrate.

JCD 2009; 1:(1). OCTOBER 2009

94

www.slm-hematology.com


of the difficulties in establishing the exact moment of when a bleeding has been completely controlled. Hemophilia treatment center and even personal-oriented practices can strongly influence the decision when it comes to give or not an additional dose to treat such bleedings. The impact of such variability is more evident outside clinical trials, and can largely influence the results of economic evaluations in mild–moderate bleedings, since one single additional dose can represent up to a 50% increase in the total dose needed to achieve hemostasis. For these reasons, one can conclude that efficacy data from sources other than comparative controlled studies should be regarded as poor-quality evidence, and, as such, not be used in economic models. Ideally, efficacy data used in these models should derive from well-designed head-to-head trials that should reflect at least in part the so-called ‘‘real world practice’’, the most relevant endpoint of any economic analysis. Two open-label, head-to-head trials have compared the relative efficacy of aPCC and rFVIIa for the treatment of mild and moderate bleedings [22, 23]. However, given the relatively low number of patients, the non-blinded design, and differences in dosing, a precise figure for the efficacy of each agent cannot be obtained from either study. In this context, data from several sources have been used to in economic models: literature-based estimates, expert opinion, retrospective and prospective analysis of treatment of bleeding episodes, and head-tohead trials. None of these strategies is perfect, but their use is necessary since it is unlikely that a clinical trial that provides unequivocal answers to the question of the relative efficacy of each of these agents (such as an adequately powered double-blind study) will ever be performed. However, it is important to acknowledge the limitations of each study design in providing reliable data to be used in pharmacoeconomic models. Literature-based estimates and retrospective studies use heterogeneous criteria to define efficacy, and vary in key variables such as time before initiation of treatment, so that comparison across different studies is questionable. Expert opinions are strongly influenced by personal experience and preferences, and tend to reflect literature-based estimates. And finally, prospective studies, and even open-label, head-to-head trials, are not immune to the influence of patient and doctor preferences given the lack of objective parameters to assess response to treatment. In order to minimize the impact of these factors on the quality of the generated data, authors have systematically performed sensitivity analysis of the results of these economic models [12, 14– 19]. This tool allows one to check the results and conclusions of a model using a different set of relative efficacy data, thereby strengthening the conclusions of the model as a whole.

studies comparing treatment with bypassing agents discuss with detail the potential influence of price in their overall conclusions. Price is indeed a key variable in all economic analysis, and slight variations have the potential to switch the direction of the conclusion of a study, even when relative efficacy is unchanged. This has been formally demonstrated by at least two studies, in which conclusions were changed by the input of prices from different sources, with variations of as little as 15% [15, 24]. Drug prices tend to vary across different markets. Therefore, the translation of results of any economic analysis into realities other than the ones that were studied should be performed cautiously, given the relevance of differences in price assumptions used in the study and local market prices.

KEY STUDIES ON THE RELATIVE EFFICACY OF BYPASSING AGENTS IN MILD AND MODERATE BLEEDINGS As stated above, relative efficacy is the single most important variable of economic evaluations of inhibitor treatment cost. Therefore, we will discuss in more detail the most relevant studies that generated data on the relative efficacy of bypassing agents. There are two head-to-head trials comparing treatment of mild with moderate bleedings in inhibitor patients [22, 23]. Both studies compare aPCC and rFVIIa in an open-label fashion, with important differences in treatment protocols. In the FEIBAH NovoSevenH Comparative (FENOC) Study trial, 1 dose of aPCC (75–100 U/kg) was compared with 2 doses of rFVIIa (90–120 mg/kg) 2 h apart. Patients reported similar effective or partially effective rates for the two treatments (60.4%, 78.7%, and 84.4% for rFVIIa, and 75.0%, 80.9%, and 80.0% for aPCC at 2, 6, and 12 h respectively) [22]. In the other comparative study, 3 doses of rFVIIa 90 mg /kg 3 h apart were compared with 1 dose of aPCC (75 U/kg). After 9 h, the proportion of patients that did not need rescue hemostatic medication was 90.9% for rFVIIa and 63.6% for aPCC (P50.07). The study also compared a single dose of rFVIIa (270 mg/kg) with a statistically significant difference in favor of the latter strategy compared with the aPCC arm [23]. Several observational studies have also addressed the efficacy of aPCC or rFVIIa for the treatment of mild and moderate bleedings in inhibitor patients. Comparison of efficacy based on these individual studies are open to bias due to differences in study design, efficacy definition, among other variables. Nevertheless, efficacy assumptions used in several pharmacoeconomic analysis were obtained from these studies. Recently, a meta-analysis of these studies, including more than 2000 bleeds was published, suggesting that rFVIIa was more effective when compared with aPCC for the treatment of these bleedings [25]. However, over-representation of non-

Price of Bypassing Agents The importance of price should be obvious for any comparative economic analysis. Surprisingly, few www.slm-hematology.com

95

JCD 2009; 1:(1). OCTOBER 2009


Journal of Coagulation Disorders

5. Kempton CL, White GC, 2nd. How we treat a hemophilia A patient with a factor VIII inhibitor. Blood. 2009;113:11–17. 6. Gringeri A, Mantovani LG, Scalone L, Mannucci PM. Cost of care and quality of life for patients with hemophilia complicated by inhibitors: the COCIS Study Group. Blood. 2003;102:2358–2363. 7. Teitel J. Inhibitor economics. Semin Hematol 2006;43(Suppl 4): S14–17. 8. Gautier P, D’Alche-Gautier MJ, Coatmelec B, et al. Cost related to replacement therapy during hospitalization in haemophiliacs with or without inhibitors: experience of six French haemophilia centres. Haemophilia. 2002;8:674–679. 9. Stephens JM, Joshi AV, Sumner M, Botteman MF. Health economic review of recombinant activated factor VII for treatment of bleeding episodes in hemophilia patients with inhibitors. Expert Opin Pharmacother. 2007;8:1127–1136. 10. Ullman M, Hoots WK. Assessing the costs for clinical care of patients with high-responding factor VIII and IX inhibitors. Haemophilia. 2006;12(Suppl 6):74–79; discussion 79–80. 11. Ekert H, Brewin T, Boey W, Davey P, Tilden D. Cost-utility analysis of recombinant factor VIIa (NovoSeven) in six children with long-standing inhibitors to factor VIII or IX. Haemophilia. 2001;7:279–285. 12. Knight C, Paisley S, Wight J, Jones ML. Economic modelling of different treatment strategies for haemophilia A with highresponding inhibitors. Haemophilia. 2003;9:521–540. 13. Joshi AV, Stephens JM, Munro V, Mathew P, Botteman MF. Pharmacoeconomic analysis of recombinant factor VIIa versus APCC in the treatment of minor-to-moderate bleeds in hemophilia patients with inhibitors. Curr Med Res Opin. 2006;22:23–31. 14. Odeyemi IO, Guest JF. Modelling the economic impact of recombinant activated Factor VII and activated prothrombincomplex concentrate in the treatment of a mild to moderate bleed in adults with inhibitors to clotting factors VIII and IX at a comprehensive care centre in the UK. J Med Econ. 2002;5:51–64. 15. Steen Carlsson K, Astermark J, Donfield S, Berntorp E. Cost and outcome: comparisons of two alternative bypassing agents for persons with haemophilia A complicated by an inhibitor. Thromb Haemost. 2008;99:1060–1067. 16. Dundar S, Zu¨lfikar B, Kavakli K, et al. A cost evaluation of treatment alternatives in mild-to-moderate bleeding episodes in haemophilia patients with inhibitors in Turkey. J Med Econ. 2005;8:46–54. 17. Ozelo MC, Villaca PR, De Almeida JO, et al. A cost evaluation of treatment alternatives for mild-to-moderate bleeding episodes in patients with haemophilia and inhibitors in Brazil. Haemophilia. 2007;13:462–469. 18. Putnam KG, Bohn RL, Ewenstein BM, Winkelmayer WC, Avorn J. A cost minimization model for the treatment of minor bleeding episodes in patients with haemophilia A and high-titre inhibitors. Haemophilia. 2005;11:261–269. 19. You CW, Lee SY, Park SK. Cost and effectiveness of treatments for mild-to-moderate bleeding episodes in haemophilia patients with inhibitors in Korea. Haemophilia. 2009;15:217–226. 20. Knight C, Dano AM, Kennedy-Martin T. A systematic review of the cost-effectiveness of rFVIIa and APCC in the treatment of minor/moderate bleeding episodes for haemophilia patients with inhibitors. Haemophilia. 2009;15:405–419. 21. Lyseng-Williamson KA, Plosker GL. Recombinant factor VIIa (eptacog alfa): a pharmacoeconomic review of its use in haemophilia in patients with inhibitors to clotting factors VIII or IX. Pharmacoeconomics. 2007;25:1007–1029. 22. Astermark J, Donfield SM, DiMichele DM, et al. A randomized comparison of bypassing agents in hemophilia complicated by an inhibitor: the FEIBA NovoSeven Comparative (FENOC) Study. Blood 2007;109:546–551. 23. Young G, Shafer FE, Rojas P, Seremetis S. Single 270 microg kg(-1)-dose rFVIIa vs. standard 90 microg kg(-1)-dose rFVIIa and APCC for home treatment of joint bleeds in haemophilia patients with inhibitors: a randomized comparison. Haemophilia 2008;14: 287–294.

controlled studies is again a significant limitation of the analysis. Taken together, the results from these key studies do not allow a precise definition of the percentage of mild–moderate bleedings that are effectively controlled with each of the two bypassing agents currently available. However, it seems reasonable to assume that the very difficulty in defining the magnitude of the difference in efficacy between rFVIIa and aPCC may indicate that, if present at all, this difference must be small, and very sensitive to variations in dosing, time to initiate treatment and other variables that are not expected to be controlled in our ‘‘real-world’’ daily practice.

CONCLUSION Several formal economic analysis have been performed comparing treatment alternatives for mild and moderate bleedings in inhibitor patients [11–19], as well as systematic evaluations of these pharmacoeconomic studies [4, 9, 20, 21]. Two predefined variables, (1) relative efficacy and (2) price, are key parameters that can switch the conclusions of these models in either direction. Therefore, when analyzing these pharmacoeconomic studies one should consider the source of these data as key quality parameters of their conclusions. The bulk of studies evaluating relative efficacy of each bypassing agent suggests that, if present, differences between them for the treatment of mild and moderate bleedings should be small, and can vary according to dosing protocols, time to initiate treatment, and other variables that are hard to control in our ‘‘real-world’’ practice. In conclusion, pharmacoeconomic studies about the treatment of mild and moderate bleeding in inhibitor patients have provided us important information about one of the most complex aspects of hemophilia care. However, the generalization of their conclusions to realities other than the ones that were studied should be always preceded by a critical analysis of the suitability of assumptions used in the model compared with local realities. Disclosure: Erich V De Paula and Margareth C Ozelo have previously received a paid consultant from Novo Nordisk A/ S, and support for scientific symposia from Novo Nordisk A/ S and Baxter Healthcare.

REFERENCES 1. Mannucci PM, Tuddenham EG. The hemophilias–from royal genes to gene therapy. N Engl J Med. 2001;344:1773–1779. 2. Bohn RL, Aledort LM, Putnam KG, Ewenstein BM, Mogun H, Avorn J. The economic impact of factor VIII inhibitors in patients with haemophilia. Haemophilia. 2004;10:63–68. 3. Chang H, Sher GD, Blanchette VS, Teitel JM. The impact of inhibitors on the cost of clotting factor replacement therapy in Haemophilia A in Canada. Haemophilia. 1999;5:247–252. 4. Knight C. Health economics of treating haemophilia A with inhibitors. Haemophilia. 2005;11(Suppl 1):11–17.

JCD 2009; 1:(1). OCTOBER 2009

96

www.slm-hematology.com


24. Seremetis S, Joshi AV, Asmussen M. Cost minimization model for treatment of minor bleeding episodes in inhibitor patients – methodological issues. Haemophilia. 2006;12:108–109. 25. Treur MJ, McCracken F, Heeg B, et al. Efficacy of recombinant activated factor VII vs. activated prothrombin complex concentrate for patients suffering from haemophilia complicated with inhibitors: a Bayesian meta-regression. Haemophilia. 2009;15: 420–436.

www.slm-hematology.com

97

JCD 2009; 1:(1). OCTOBER 2009


Journal of Coagulation Disorders  

The Journal of Coagulation Disorders is distributed in print and electronic form across Europe to thousands of physicians, researchers, acad...

Read more
Read more
Similar to
Popular now
Just for you