Proefschrift barents by Nicole Nijhuis

Chronic heart failure diagnostics and application of neuropeptides in residential elderly

Maaike Barents

Chronic heart failure diagnostics and application of neuropeptides in residential elderly

Maaike Barents geboren 27 december 1950 te Utrecht

Chronic heart failure diagnostics and application of neuropeptides in residential elderly Proefschrift

ter verkrijging van de graad van doctor aan de Rijksuniversiteit Groningen op gezag van de rector magnificus prof. dr. E. Sterken en volgens besluit van het College voor Promoties.

De openbare verdediging zal plaatsvinden op woensdag 23 december 2015 om 16:15 uur door

Maaike Barents geboren op 27 december 1950 te Utrecht 3

Promotores Prof. F.A.J. Muskiet Prof. J.L. Hillege

Copromotor Dr. M.J.L. de Jongste

Beoordelingscommissie Prof. A.W. Hoes Prof. W.P.A. Achterberg Prof. A.A. Voors

Financial support by the Dutch Heart Foundation for the publication of this study was gratefully acknowledged.

We recognize thankfully the support for this thesis given by Corparis Welzijn met Zorg, www.corparis.nl.

CONTENTS List of abbreviations

Chapter 1

General introduction to heart failure 1.2 Rationale

1.3 Mortality

1.4 Morbidity

1.5 Defining heart failure and its causes

1.6 Prevalence of heart failure (HF)

1.7 HF disease burden (DALY)

1.8 Quality of life

1.9 Diagnostics and algorithm

1.10 Natriuretic peptides

1.11 Therapies for heart failure

1.12 Residential elderly

1.13 Heart failure in residential elderly

1.14 Aim and outline of the thesis

Chapter 2 Diagnostics and prevalence of heart failure in residential elderly. Chapter 2.1

Prevalence and misdiagnosis of chronic heart failure in nursing home residents: the role of B-type natriuretic peptides. Barents M, van der Horst IC, Voors AA, Hillege JL, Muskiet FA, de Jongste MJ.

Neth Heart J 2008;16(4):123-8. 7

Chapter 2.2

Underestimated prevalence of chronic heart failure among the nursing home residents in Aruba. Stoutjesdijk E, Brouns RM, Barents M, de Jongste MJ, Besselink HJ, Cheng JD, Wever R, Muskiet FA.

West Indian Med J 2014;63(6):610-5.

Chapter 2.3

Validation of neuropeptides in stable heart failure patients. Patients with stable chronic heart failure exhibit also high intraindividual Biological variation of amino-terminal pro-brain natriuretic peptide (NT-proBNP) in urine when compared to the variation in plasma. Schimmel AM, Barents M, de Jongste MJ, Rรถmer JW, Steward HN, Muskiet FA.

Accepted for publication in part in Clinical Chemistry.

Chapter 3

111

BNP and NT-proBNP, predictors of 1-year mortality in nursing home residents. Barents M, Hillege HH, van der Horst IC, de Boer RA, Koster J, Muskiet FA, de Jongste MJ.

J Am Med Dir Assoc 2008;9(8):580-5.

Chapter 4

129

Chronic heart failure, related to help with activities of daily living in nursing home residents. Barents M, Hillege HH, Muskiet FA, de Jongste MJ.

Act Adapt Aging 2011;35(2):98-110.

Chapter 5

151

Physical exercise performed by residential elderly of high age is safe and does not affect chronic heart failure symptoms, NT-proBNP and cardiovascular-metabolic indices: a pilot, single-blinded RCT. Barents M, Weening E, Muskiet FA, Hillege HH, de Jongste MJ. Submitted.

Chapter 6

173

Chapter 6.1 Summary

174

Chapter 6.2 Discussion

185

Chapter 6.3 Recommendations and future perspectives

191

Chapter 7

199

Chapter 7.1 Samenvatting

200

Chapter 7.2 Discussie

211

Chapter 7.3 Aanbevelingen en toekomst perspectieven

217

Dankwoord en CV

221

ABBREVIATIONS AND DEFINITIONS ACE-i

angiotensin-converting enzyme inhibitor

ADL

activities of daily living

Advance care planning aims to help patients, establish decisions about future care that take effect when they lose capacity AF

atrial fibrillation

ARB

angiotensin-2 receptor blocker

AWBZ

Algemene Wet Bijzondere Ziektekosten, funding source of elderly care medicine except geriatric rehabilitation (Ziektekosten Wet)

BNP

B-type natriuretic peptide or brain natriuretic peptide

CHF

chronic heart failure. Actually, “non-acute” heart failure is a better term because a physician does not wait to diagnose heart failure till it has become chronic” (three months). But, we have used “chronic” heart failure in all publications and decided therefore not to replace the term

comorbidity

diseases of which a person suffers, near the disease which is focussed at. Comorbidity is standardised by the Charlson comorbidity index. This combined age and comorbidity is developed to estimate the relative risk of death from prognostic clinical variables. In many studies, the index is applied to correct for existing comorbidity (1)

COPD

chronic obstructive pulmonary diseases

CRT-P/CRT-D

chronic resynchronization therapy-pacemaker/defibrillator

CVD

cardiovascular diseases

CVA

cerebrovascular accident

CVi

intraindividual biological variation

CVRM guideline

cardiovascular risk management guideline (2)

DALY

disability adjusted life years is a measure of overall disease burden, expressed as the number of years lost due to ill-health, disability or early death

ECG

electrocardiogram

echo

echocardiogram

eGFR

estimated glomerular filtration rate

elderly

persons of 65 years and older; also older persons elderly care physician the elderly care physician is a physician who specializes in longterm care for frail elderly people and chronic patients with complex health problems (3)

ESC

European Society of Cardiology

general practitioner

HbA1c

HbA1c ≤8,5% = GTA (Gron. Transmurale Afspraak ) T2DM

HDL-C

high-density lipoprotein cholesterol; HDL-C ≥ 1 mmol/L (39 mg/dL) males, HDL-C ≥1,2 mmol/L (46 mg/dL) females

heart failure

HFpEF

heart failure with preserved ejection fraction

HFrEF

heart failure with reduced ejection fraction

mercury

ICD

implantable cardioverter defibrillator

insulin resistance

Lab Noord

Laboratorium Noord (Laboratory North) in The Netherlands

LDL-C

low-density lipoprotein cholesterol; LDL-C < 2,5mmol/L (100 mg/dL)

LTCFs

long-term care facilities comprising nursing homes and care homes for the elderly

LVEF

left ventricular ejection fraction

MDS

minimum data set, items of the Resident Assistant Instrument (RAI). MDS-items consist of systematic observations of dependency and de livered aid through which dependence and aid can be quantified. These observations focus on self-reliance of the individual residential elderly during predefined activities of daily living.

MET

Metabolic Equivalent Task (MET), an unit for metabolism in

rest

METS

metabolic syndrome

MeSH

medical subject headings; controlled key words of MEDLINE database

myocardial infarction

MMSE

minimum mental state estimation

multi medication use a daily intake of at least five different drugs (4) multimorbidity

multiple co-occurring, chronic or long-term diseases or conditions, none considered as index disease (5)

MSD

musculoskeletal disorders

NICE guideline

National Institute for Health and Care Excellence

nursing home

NLM database

National Library of Medicine database (Maryland, USA)

NND

neurodegenerative disorders

norm physical activity a minimum of physical activity at â&#x2030;Ľ4 METs for individuals of 20-55 years during 30 min at 5 days a week and â&#x2030;Ľ3 METs for 55+ years (6) NP

natriuretic peptide

NT-proBNP

amino-terminal proB-type natriuretic peptide

NYHA class

New York Heart Association classification of heart failure

oldest old persons

persons aged > 85 years (medicaldictionary.thefreedictionary.com)

PMC database

Pub Med Central database

polypharmacy

polypharmacy is defined as the use of five or more drugs, including prescribed, over-the-counter, and complementary medicines (4)

primary care physicians physicians, general practitioners and elderly care physicians Quol

quality of life

RAAS

renin-angiotensin-aldosterone system

RAI

Residentâ&#x20AC;&#x2122;s Assessment Instruments, software program for providing problem lists as base for the individual care-treatment plan made by the direct care-giver and for the State Governors (7)

RCT

randomized controlled trial

RIVM

Rijks Instituut voor Volksgezondheid en Milieu

renal dysfunction

residential elderly

elderly residing in nursing homes or care homes; also institutionalised frail elderly

TIA

transient ischemic attack

TMWD

ten meter walk distance

TUG

timed up and go test

triglycerides; fasting TG â&#x2030;¤8mmol/L

T2DM

type 2 diabetes mellitus

Verenso

vereniging van specialisten ouderengeneeskunde en sociaal geriaters (Society of elderly care physicians and social geriatricians)

VZH

verzorgingshuis, care home for the elderly. One of the institutes of LTCFs

Chapter 1 GENERAL INTRODUCTION. This chapter provides a comprehensive review of heart failure (HF), which is the central theme of this thesis. Since HF is a syndrome predominantly affecting the elderly, it imposes an important disease burden in elderly medicine. Elderly medicine is a relatively young specialism. On a daily basis, elderly care physicians deal with the question when and how to apply general medicinal guidelines and in which cases the residentâ&#x20AC;&#x2122;s high age has to be taken into account. Issues as diagnostics and exercise are relevant for the frailest elderly with HF. Due to the complexity of frail elderly, optimal diagnosis and treatment, remain a major challenge for the practicing physician in order to improve outcome. At the end of this chapter we focus on HF in the frailest elderly, i.e. those who have to rely on long-term care facilities (LTCFs), since heart failure in the frailest elderly is the subject of this thesis.

1.2 RATIONALE. In The Netherlands, the average life expectancy has increased from 70 years for men and 73 years for women in 1950 to 79 and 83 years in 2013, respectively (www.cbs.nl and www.nationaalkompas.nl). Main determinants of the increased life expectancy are the implementation of better hygienic and preventive measures, easy availability of sufficient and healthy food in conjunction with a shrewd and balanced lifestyle and the development of an armamentarium of medical treatments, such as pharmacotherapy and surgical techniques. One of the most important achievements in reducing overall mortality originates from the enormous progress made in the management of cardiovascular disease (CVD) during the last 60 years. As a consequence the many survivors of CVD grow older and acquire other non-CV diseases, single or multiple, often accompanied by disabilities. When multimorbidity is accompanied by the loss of psychosocial resources, we speak of frailty (8). Frail elderly are at risk to deteriorate even

faster and geriatric interventions are required. The frailest elderly live in LTCFs, like nursing homes or care homes, and are referred to as â&#x20AC;&#x2122;residential elderlyâ&#x20AC;&#x2122;. Although this group consumes a growing percentage of the Dutch health care budget they are excluded from research studies because of high age, multimorbidity and cognitive disorders. Until recently, these elderly were diagnosed and treated in the same way as adults with a single disease. At the start of this study there was no HF guideline available for the group of residential elderly. This was because the HF guideline of Verenso from 1999 was out of date (Verenso, Vereniging van specialisten ouderengeneeskunde en sociaal geriaters). In 2003 and in 2008, the Dutch multidisciplinary HF guideline, based on the European Society of Cardiology (ESC) HF guideline, recognised the lack of research data for females and the 75+ age-group (both genders) (9;10). Meanwhile, natriuretic peptides (NP) proved their utility for HF diagnostics, prognosis and therapyoptimization. In addition the importance of non-pharmacological treatment such as physical exercise was well recognized. These niches of knowledge on CHF diagnostics and nonpharmacological therapy formed the rationale for studying this excluded group of the residential elderly.

1.3 MORTALITY FROM CARDIOVASCULAR DISEASES. Since its peak in the late 1950s (for women) and the early 1970s (for men) cardiovascular (CV) mortality has gradually declined. This tremendous decrease in CV mortality is remarkable when we take into account the increasing number of elderly people and the more advanced age of contemporary Dutch citizens. Also, after correction for age (standardisation), the drop in CV mortality remains considerable, with a reduction of about 70% (Figure 1).

Figure 1. Standardised Cardiovascular Mortality per 100,000 inhabitants in The Netherlands (8).

1.4 MORBIDITY. As spin-off of the impressive 68% decrease in CVD mortality, a substantial increase in CVD morbidity is nowadays reported, with an estimated prevalence of 700,000 subjects in The Netherlands currently suffering from CVD. Out of these 700,000, around 125,000 have HF (11). As for CV mortality to date more patients die from HF than from myocardial infarction (MI), since an increasing number of people survive their MI, but subsequently develop HF (Figure 2).

25000

20000

15000

10000

5000

0 1980

1985

1990

Myocardial infarction

1995

2000

2005

2010

Heart failure

Figure 2. Heart failure mortality has surpassed myocardial infarction mortality in The Netherlands since 2011. Number of deaths attributed to myocardial infarction (MI) or heart failure (HF) as the primary cause of death in The Netherlands from 1980 through 2012 (12).

1.5 DEFINING HEART FAILURE AND ITS CAUSES. Defining HF is not easy. The most straightforward approach is to define HF as: a condition during which the heart is unable to meet the metabolic needs of the body. Unfortunately in daily practice HF is often more difficult to define. Also, this is caused by the multi-interpretability of vague key symptoms of HF, notably, fatigue and dyspnea. The significance of symptoms and signs is even more difficult to interpret in elderly due to the presence of co-morbidities. At present three definitions are used to label HF. These definitions describe the diagnosis of HF on a clinical (symptomatic), a pathophysiological or a therapeutic basis (9;13;14). Given the difficulties to define HF, discovery of (the onset of) HF is often late. 9

Clinical (symptomatic) HF definition: HF is a syndrome in which the patients should have the following symptoms typical for HF: shortness of breath at rest or during exertion,

and/or fatigue; signs of fluid retention such as pulmonary congestion or ankle swelling; and objective evidence of an abnormality of the structure or function of the heart at rest (9). 9

Pathophysiological HF definition: HF is defined as a pathophysiological state in which an abnormality of cardiac function is responsible for failure of the heart to pump blood at a rate commensurate with metabolic requirements or to do so only from an elevated filling pressure (13).

Therapeutic definition: cardiomyopathies are a heterogeneous group of diseases of the myocardium associated with mechanical and/or electrical dysfunction that usually (but not invariably) exhibit inappropriate ventricular hypertrophy or dilatation and are due to a variety of causes. Cardiomyopathies either are confined to the heart or are part of generalized systemic disorders, often leading to cardiovascular death or progressive HF– related disability (14). In addition to the three definitions, HF is classified by course of disease and functionality.

With respect to course: after 3 months HF is considered as chronic. Chronic HF (CHF) can be stable or gradually progressive, or can exacerbate into acute HF. Regarding the functionality, HF is distinct in diastolic HF, also named HF with preserved left-ventricular ejection fraction >40-50% (LVEF; HFpEF), and systolic HF, also named HF with reduced LVEF (HFrEF) (9). For diagnostic purposes, for both HFrEF and HFpEF, typical signs and symptoms of HF are obligatory. For HFpEF two more requirements are needed: a non dilated left ventricle and documentation of diastolic dysfunction. The diagnosis of HFpEF is more difficult than that of HFrEF because it is largely one of exclusion, e.g. of potential non-cardiac causes of the patient’s symptoms. As for the individual patient’s restrictions, the New York Heart Association (NYHA) classification is used (15). NYHA places patients in one of four categories based on how much they are limited during physical activity.

All possible causes for HF may be divided into derailments of the heart muscle as a result of 19

cardiomyopathy, coronary artery disease (CAD), atrial fibrillation or other rhythm or conducting disturbances, infection or intoxication. In addition, diseases of the heart valves, afflictions of the pericardium or (cardiac) shunts may cause HF. Finally, circulatory derangements may induce HF, such as coarctation, thrombo-embolic process, intoxication, high output failure, infection, hyperthyroidism and fluid overload. In general, HFrEF is most often caused by CAD which is predominantly generated through hypertension (HT).

1.6 PREVALENCE OF HEART FAILURE. Due to an increasingly ageing population, already in 1990 a growing number of individuals with HF was predicted in The Netherlands (12). After 22 years, the estimation of HF became reality, with a prevalence of about 120,000 – 130,000 subjects. Moreover, the number of people with HF is expected to increase to reach about 200,000 individuals in 2018 (16). Between 20 to 30% of the general population will ultimately develop HF, when they are over 70 years of age (17).

1.7 HEART FAILURE DISEASE BURDEN. Apart from the increasing HF mortality, the morbidity or disease burden due to HF will rise as well. The “Rijksinstituut voor Volksgezondheid en Milieu” (RIVM) has quantified “disease burden” in a model of disability-adjusted life-years (DALY) (18), expressed as the number of years lost due to ill-health, disability or early death through HF. The authors have estimated the DALY for each of four risk factors for HF separately. These risk factors comprise hypertension (HT; moderate HT is SBP of 140-160 and serious HT ≥160 mm Hg), moderately-high total cholesterol (200-239 mg/dL; 5.17-6.18 mmol/L) and seriously-high total cholesterol (≥240 mg/dL; ≥6.20 mmol/L), overweight, obesity and inactivity. Inactivity is defined as <4 metabolic equivalents of tasks (METs) for age 20-55 years and <3 METs for age 55+. A DALY was found of 20

1.6 years for seriously-high total cholesterol and up to 2.9 years for serious HT (Figure 3). The consequences of DALY for the AWBZ will be that an increasing part of the Dutch health care budget have to be devoted to the HF disease burden. healthy

HF free

overweight

with HF

increased cholesterol

loss of LE

strongly increased cholesterol hypertension obesity serious hypertension inactivity 77

Figure 3. Disability-adjusted life-years (DALY). A DALY-model estimated for a Dutch population classified by different risk factors for heart failure. Disability-Adjusted Life-Years (DALY), expressed as number of years lost due to ill-health, disability or early death. The bars are equally long and the upper bar represents the lifeexpectancy (LE) of a 20 year old with a lifelong optimal HF risk profile. The middle parts of the bars reflect the years lived with HF. The right parts of the bars show the years lost due to HF (18).

1.7 QUALITY OF LIFE (QUOL). The DALY quantifies the time lived with reduced Quol due to HF. Not only the direct loss of Quol by HF symptoms but also the indirect loss by restrictions, as a result of HF, are important. These restrictions appear in several Quol domains like mobility and independency. For instance,

dyspnea limits the ability to execute activities of daily living (ADL) and walking distance (19). From the perspective of the individual, the severity of HF affects a personâ&#x20AC;&#x2122;s self-reported functioning (20). It is obvious that HF is related to assistance at ADL, but this relation has not been investigated by others. Such a relation, if causal, is important for the care-giver who needs to estimate the amount of required ADL assistance and for the individual who needs that help. The physician may estimate the prognosis of patients using independent predictors of HF such as high age, NYHA 3 and 4, loss of appetite (21), N terminal pro B-type natriuretic peptide (NT-pro)BNP (22) and chronic kidney diseases (CKD) (23). Among others these prognostic predictors enable a patient with CVD to make personal choices for the last phase of his life. The choices may permit the patient to retain control over his or her life and thereby prevent decrease of autonomy, at least on the domain of decision-making.

1.9 HF DIAGNOSTICS AND ALGORITHM. In primary care elderly-medicine, HF is diagnosed using CV medical history, for HF predisposing diseases and risk factors (chapter 1.5), clinical examination, electrocardiography (ECG), natriuretic peptides (chapter 1.10) and echocardiography. These are referred to as diagnostic tools. The reference standard for the, at least functionally undifferentiated HF diagnosis is based on all of these tools, occasionally extended with chest X-ray (9;10;24). For further examination of HF causes or treatment, referral to a cardiologist may be required. However, predictive values of the separate diagnostic tools are moderate which hamper initial HF diagnostics in primary care and in geriatric outpatients as well (25;26). This hindrance combined with no easy access to echocardiography or unwanted referral (residentâ&#x20AC;&#x2122;s preference), needed a solution to diagnose HF without outpatient referral. For that purpose HF diagnostic algorithms were developed based on combinations of diagnostic tools including NPs (21;25-28). However a few is known whether those HF diagnostic algorithms (9;10;21;24-28) are also suitable for the residential elderly suspected of HF. 22

1.10 NATRIURETIC PEPTIDES. At the end of the last century, NPs have emerged as reliable biomarkers for the diagnosis and prognosis of both systolic and diastolic HF. NP testing is regarded to be the single most useful test to add to the diagnostic pathway for HF in primary care. Tests of NPs have shown to be cost effective. However, in HFpEF levels of NPs are increased, but somewhat lower compared to those in patients with HFrEF. Therefore, the relatively higher NP levels are predictive for HFrEF, but the moderately increased NP levels do not predict HFpEF very well (29). Especially the NPs specificity and diagnostic accuracy in acute HF are prominent. Currently, NPs, notably B-type natriuretic peptide (BNP) and N-terminal-proBNP (NT-proBNP), are commonly applied diagnostic (30) and prognostic (31) biomarkers. Confounders of NP levels are non-cardiac factors such as high age, sex and renal dysfunction, since these also increase NP levels. Obesity, diuretics, ACE inhibitors, beta-blockers, ARBs and aldosterone antagonists reduce NP levels (32). All the above mentioned confounders occur widely in residential elderly. Many of them use CV medications and renal dysfunction is widespread. Therefore, NPs may not predict HF exclusively (32) and NP cut-off values are not fully validated in this particular group. With regard to treatment optimization, a meta-analysis shows that NP-guided therapy is superior to symptom-guided therapy in groups of patients with CHF (33). However a study on individual CHF patients reveals high intra-individual biological variations of NPs as major limitation of NPs in the optimization of HF treatments (34).

The secretion of NPs by the ventricle wall increases when HF is accompanied by pressure or volume overload (35). Only BNP regulates actively. BNP stimulates natriuresis, diuresis, vasodilatation and inhibits the renin-angiotensin-aldosterone system (RAAS) and sympathetic nervous system. BNP is metabolized actively in blood and cells by neuropeptidases (36). The halflife time of BNP is 20 minutes. These characteristics render BNP appropriate as marker of HF (30). NT-proBNP has an about six times longer half-life, viz. 1-2 hours (37). As opposed to BNP, NT-proBNP is passively filtered in the kidney (38), actively reabsorbed by the proximal 23

tubulus brush border cells and catabolised to amino acids. This process of NT-proBNP degradation is usually nearly complete. Thus, only minor amounts of NT-proBNP are recovered in urine (39). Both NPs are released into the circulation in an 1:1 molar ratio. The secretion of BNP exhibits a pulsatile pattern in the circulation (40) and it may be assumed that NT-proBNP is split-off pulse-shaped as well. Consequently both NPs exhibit high intra-individual biological variations which hamper their use for treatment optimization of an individual (34). Mainly experimentally, NPs are determined in urine using immunoassays. Urine NPs may exhibit less intra-individual biological variation compared to plasma NPs, assuming that the pulsatile secretion becomes attenuated by NT-proBNP accumulation in the bladder (41-43). Thus, the experimental use of urine NP is prompted by the high intra-individual biological variation of plasma NPs (34).

1.11 THERAPIES FOR HEART FAILURE. Following the diagnosis of HF, treatment is based on CV risk management (CVRM) (2) and correction of underlying causes (10). CVRM for HF differs in the time of intervention (2) from other HF guidelines. The current trend is not to wait until HF symptoms have developed, but to start intensive risk management for patients with increased risk of HF (www.RIVM.nl). It is assumed that if HF could be recognized before symptoms emerge, lifestyle adaptations or appropriate medication could reduce HF symptoms or the physical decline caused by HF. An example of lifestyle effects on CV risk factors is presented by a five-year lasting Dutch study. This study shows that 1% of adults has changed from physically inactive into moderately-active and 1-2% from overweight/obese or underweight to normal weight (6). Whether such endpoints can also be reached in HF patients belonging to the elderly and whether such interventions could prevent HF progression or even accomplish a reverse, is unknown.

In addition, it is advised to treat other diseases precipitating to HF. It is assumed that treatment of anemia, lung diseases, renal dysfunction, thyroid-gland dysfunction, T2DM and intoxications reduces the risks of HF (9). Treatment of underlying causes of HF consists, among others, of pharmacological therapy. Most pharmacological therapies are specific for HFrEF (Table 1).

Table 1. Treatment of Heart Failure with reduced left ventricular ejection fraction (HFrEF) (26). HF with reduced ejection fraction

treatment

start

diuretics to relieve signs/symptoms and ACE-i (or ARB)

When clinically stable, add

beta-blocker

Titrate to the target doses of

ACE-i (or ARB) and beta-blocker

When complaints are not persisting

no complementary therapy indicated

When complaints are persisting, add

MR antagonist

When instable (SR and HR ≥ 70/min), add

Ivabradine

When complaints does not persist, EF≤35%

No further specific treatment

When complaints are persisting and QRS≥120ms

Consider CRT-P/CRT-D

When complaints are persisting and QRS ≤120ms consider ICD When complaints are persisting

Consider digoxin and/or hydralazine or isosorbide nitrate

Angiotensin Converting Enzyme inhibitor, ACE-i; Angiotensin Receptor II Blocker, ARB; ejection fraction, EF; sinus rhythm, ARB; heart rate, HR; mineralocorticosteroid receptor, MR; cardiac resynchronization therapy pacemaker, CTR-P/CRT-D, defibrillator.

Since patients with advanced CHF (EF ≤35 - 40%) are at risk of sudden death, an implantable cardioverter defibrillator (ICD) is placed, for preventive reasons according to guidelines. Finally, non-pharmacological treatments to improve systolic function of the left ventricle function comprise, among others, revascularization procedures, valvular or ventricular surgery, rhythm optimization (electrocardioversion, ablation), cardiac resynchronization therapy (CRT) and heart transplantation. For HFpEF no specific therapy is available, yet. For these patients only the un26

derlying cause such as hypertension is treated (10). For the elderly, the same treatment is available provided that dosages are adjusted to decreased kidney function and the life expectancy is at least 0.5 years with good quality of life (10).

1.12 THE RESIDENTIAL ELDERLY. The residential elderly belong to the frailest elderly (chapter 1.2). They reside in care or nursing homes (LTCFs). Care and nursing homes differ in intermittently or continuously provided support and ADL-help, respectively. During the final stage of writing this thesis in 2015, financing of Dutch care homes had just ended. As a consequence most care home elderly had to move to private houses with care at home or to a nursing home. In primary care elderly medicine, preferences of the residential elderly are inventoried and taken as the starting point for further treatment decisions in case they become incompetent in future. These preferences concern diagnostics, treatment and hospital referral in the event of emergency. These choices or preferences are set out in agreements that are evaluated twice a year, or more often when indicated. These agreements may include renouncing invasive and intensive care treatments and hospital referral. As an exception, hospital referral for (surgical) treatment of bone-fractures are seldom renounced.

1.13 HEART FAILURE IN RESIDENTIAL ELDERLY. The increasing HF prevalence on the one hand, and lack of specific HF research data on the other hand challenge healthcare professionals in the provision of optimal patient care in the residential elderly. We expected the HF diagnostics to be less accurate due to high age, female sex, multimorbidity (44;45), â&#x20AC;&#x2122;treatment agreementsâ&#x20AC;&#x2122; and absence of a specific HF guideline. At the start of this study, it was unknown whether general HF guidelines could be applied to the residential elderly (9;10;24). Therefore, we set out to study HF prevalence and diagnostic ac27

curacy in this particular group of elderly people. We planned also to investigate whether HF diagnostics could be improved by adding NPs to the usual HF diagnostics. Furthermore, HF guidelines, the CVRM guideline and family members recommend residential elderly to be physically active (2;9;10;24) aiming to maintain or improve physical health. The residential elderly, however, seems to have little motivation to train physically on a regular basis. He or she faces a limited life expectancy wondering whether his exertion will provide benefits. Not much is known as to whether non-elective exercise might be beneficial in residential elderly (46). Staffs of nursing and care homes doubt whether they should invest in regular exercise programmes and whether such interventions will be safe and effective. Therefore, we experimented with an exercise program based on a review specific to the residential elderly (47). In addition, when facing a limited life expectancy, accurate information on prognosis will become more important. Prognostic information allows residential elderly to take their own decisions so that loss of Quol at least in the area of autonomy is reduced (chapter 3). Natriuretic peptides have been shown a powerful prognostic marker in adults with HF and may have similar prognostic value in residential elderly. However, the prognostic properties of NPs have not been not examined in this particular group. Moreover, many confounders influence NP levels in such a way that they hamper to inform on prognosis in this residential target group. Therefore we aimed to study whether NPs could be used as a prospective and independent predictor in those frail elderly with HF. Independency of care takers determines Quol to a high degree. It was surprising to learn, at the start of this study, that the HF diagnosis was not on the list of geriatric chronic diseases in the â&#x20AC;&#x2DC;Resident Assessment Instrumentâ&#x20AC;&#x2122; (RAI). The RAI is in use in American and Dutch nursing homes for care-planning at an individual, organizational and state level (48). Nevertheless, we presumed a relationship between HF state and help-with-care in residential elderly. In that line of reasoning, NP concentrations might be related to help-with-care as well. However a relation

between HF and or NPs with ADL-assistance was not observed until now. Therefore, we set out to test these research questions.

1.14 AIM AND OUTLINE OF THE THESIS. In a broad perspective, the studies in this thesis aim to determine whether application of parts of the HF guidelines can be accurately applied to the specific group of the residential elderly. The doubts on the applicability of HF guidelines in residential elderly is based on the systematic exclusion of this group from HF research. There is also dubiety based on the observed higher cutoff values of NPs at a higher age, in renal dysfunction and comorbidity. Evidence as to what extent HF guidelines can be applied is important to elderly care physicians. These questions have undermined the accuracy of HF diagnostics resulting in no treatment or inappropriate treatment, and less well-being. Therefore, we aimed to obtain insight into certain HF aspects, such as prevalence, diagnostic accuracy, diagnostic and prognostic values of NPs and the relation of HF with help-at-care. To this end, we screened the residential elderly of a single nursing home in Groningen on the presence of CHF. To examine its external validity we repeated this screening program in Aruban nursing homes. Furthermore, we sought to investigate whether physical activity was to the benefit of care home elderly on the basis of cardio-metabolic endpoints. These comprise waist circumference, blood pressure, glucose-homeostasis and lipids. We were also interested to see whether complaints and NPs ameliorate in the care home elderly with HF. With both aims, we conducted a randomized controlled trial employing a 16-weeks lasting supervised exercise intervention. For this, we collaborated with investigators examining partly the same care home elderly for other, more functional, endpoints (46). At last, aiming to gain insight into the feasibility of NPs for treatment optimization, we estimated the intra- and inter-individual variations of NPs in urine and plasma of stable CHF patients living in Curaรงao. 29

Specific aims of this study are described in the following paragraphs: In Chapter 1 we provide a general introduction of HF with special emphasis on HF diagnostics in residential elderly. In Chapter 2.1 we study both prevalence and accuracy of HF diagnostics in the Groningen residential elderly. In the meantime, we aim to find out whether improvements can be achieved by adding NPs to the diagnostic process in use in a Dutch nursing home. The aim of Chapter 2.2 is similar to that of Chapter 2.1. but now applied to all residential elderly in Aruba. In Chapter 2.3 we study the intra-individual variations of NPs with the purpose of finding out whether urine NPs may exhibit less variations compared to those in plasma in the Curacao population In Chapter 3 we study the prognostic value of NPs on one year mortality of the Groningen residential elderly, aiming to better inform those facing a limited life expectancy. The aim of Chapter 4 is to determine the relationship between CHF and the required need for ADL-help in the residential elderly in Groningen. The aim of Chapter 5 is to examine the effects of a physical exercise program for the Groningen care home elderly. We investigated whether such a program may be beneficial with respect to cardiovascular-metabolic (soft) endpoints. Here, we could collaborate with an investigator who would perform the same intervention, but aiming at other, more functional endpoints. In Chapter 6.1 we summarized and in Chapter 6.2 we discussed the results of the above described studies and in Chapter 6.3 we provided recommendations for future studies. In Chapter 7 a Dutch summary (â&#x20AC;&#x153;Samenvattingâ&#x20AC;?) is given.

Heart failure diagnostics

All nursing homes on Aruba N=51

Heart failure and 1-year mortality Heart failure and

1 nursing home Groningen N=103; ADL-assist. N=93

ADL-assistance

Natriuretic peptides urine versus plasma

Heart failure and exercise

Heart failure patients on Curaรงao

N=25

4 care homes in Groningen N=52

Figure 4. Flowchart of studies presented in this thesis with the corresponding populations in which the studies have been conducted. ADL, Activities of daily living.

References 1

Charlson ME, Charlson RE, Peterson JC, Marinopoulos SS, Briggs WM, Hollenberg JP. The Charlson comorbidity index is adapted to predict costs of chronic disease in primary care patients. J Clin Epidemiol 2008;61(12):1234-40.

Wiersma T, Smulders YM, Stehouwer CD, Konings KT, Lanphen J. [Summary of the multidisciplinary guideline on cardiovascular risk management (revision 2011)]. Ned Tijdschr Geneeskd

2012;156(36):A5104.

Koopmans RT, Lavrijsen JC, Hoek JF, Went PB, Schols JM. Dutch elderly care physician: a new generation of nursing home physician specialists. J Am Geriatr Soc 2010;58(9):1807-9.

Hilmer SN, Mager DE, Simonsick EM, Cao Y, Ling SM, Windham BG, et al. A drug bur den in dex to define the functional burden of medications in older people. Arch Intern Med 2007;167(8):781-7.

Akner G. Analysis of multimorbidity in individual elderly nursing home residents. Development of a multimorbidity matrix. Arch Gerontol Geriatr 2009;49(3):413-9.

Wendel-Vos W, Droomers M, Kremers S, Brug J, van Lenthe F. Potential environmental determinants of physical activity in adults: a systematic review. Obes Rev 2007;8(5):425-40.

Hirdes JP, Ljunggren G, Morris JN, Frijters DH, Finne, Soveri H, et al. Reliability of the interRAI suite of assessment instruments: a 12-country study of an integrated health information system. BMC Health Serv Res 2008;8:277-88.

Smets IH, Kempen GI, Janssen-Heijnen ML, Deckx L, Buntinx FJ, van den Akker M. Four screening instruments for frailty in older patients with and without cancer: a diagnostic study. BMC Geriatr 2014;14:26.

McMurray JJ, Adamopoulos S, Anker SD, Auricchio A, Bohm M, Dickstein K, et al. ESC Guide lines for the diagnosis and treatment of acute and chronic heart failure 2012: The Task Force for the Diagnosis and Treatment of Acute and Chronic Heart Failure 2012 of the European Society of Cardiology. Developed in collaboration with the Heart Failure Association (HFA) of the ESC. Eur Heart J 2012 33(14):1787-847.

Voors AA, Walma EP, Twickler TB, Rutten FH, Hoes AW. [Multidisciplinary guideline 'Heart failure 2010']. Ned Tijdschr Geneeskd 2011;155:A2957.

11 National Institute for Public Health and the Environment (RIVM). National Public Health

Compass.

2013

[November

19,

2013];

Available

from:

http://www.nationaalkompas.nl/ gezondheid-en-ziekte/ziekten-en- aandoeningen/hartvaatstelsel).

Vaartjes I, Koopman C, van Dis I, et al. Hart- en vaatziekten in Nederland 2013, cijfers over leefstijl, risicofactoren, ziekte en sterfte. The Hague: Dutch Heart Foundation; 2013.

Braunwald E. Pathophysiological of HF. In: Braunwaldâ&#x20AC;&#x2122;s Heart Disease. A textbook of Cardio vascular Medicine. 10th ed. 2014.14.

Maron BJ, Towbin JA, Thiene G, Antzelevitch C, Corrado D, Arnett D, et al. Contemporary definitions and classification of the cardiomyopathies: an American Heart Association Scientific Statement from the Council on Clinical Cardiology, Heart Failure and Trans 33

plantation Committee; Quality of Care and Outcomes Research and Functional Genomics and Translational Biology

Interdisciplinary Working Groups; and Council on

Epidemiology and Prevention. Circulation 2006;113(14):1807-16.

Raphael, Claire; Briscoe, C; Davies, J; Ian Whinnett, Z; Manisty, C, et al (2007). "Limitations of the New York Heart Association functional classification system and self-reported walking dis-

tances in chronic heart failure." Heart (British Cardiac So-

ciety) 93(4):476â&#x20AC;&#x201C;82.

Engelfriet PM, Hoogenveen RT, Poos MJJC, Blokstra A, van Baal PHM, Verschuren WMM. Hartfalen: epidemiologie, risicofactoren en toekomst. Heart failure: epidemiology, risk factors and future. In: Rijksinstituut voor Volksgezondheid en Milieu, RIVM. Rapport 260401006. 2012.

Bleumink GS, Knetsch AM, Sturkenboom MCJM, Straus SM, Hofman A, Deckers JW, et al. Quantifying the heart failure epidemic: prevalence, incidence rate, lifetime risk and prognosis of heart failure the Rotterdam study. Eur Heart J 2004;25:1614â&#x20AC;&#x201C;1619.

Murray, C. J. and Lopez, A. D. Evidence-based health policy-lessons from the Global Burden of Disease Study. Science 274(5288), 740-743. 1-11-1996.

Steptoe A, Deaton A, Stone AA. Subjective wellbeing, health, and ageing. Lancet 2015;385(9968):640-8.

20 Munyombwe T, Hofer S, Fitzsimons D, Thompson DR, Lane D, Smith K, et al. An evaluation of the Minnesota Living with Heart Failure Questionnaire using Rasch analy

sis. Qual Life Res 2014;23(6):1753-65.

21 Oudejans I, Mosterd A, Zuithoff NP, Hoes AW. Applicability of current diagnostic algorithms in geriatric patients suspected of new, slow onset heart failure. Age Ageing 2012;41(3):309-16.

22 Don-Wauchope AC, Santaguida PL, Oremus M, McKelvie R, Ali U, Brown JA, et al. Incremental

predictive value of natriuretic peptides for prognosis in the chronic stable

heart failure population: a systematic review. Heart Fail Rev 2014;19(4):521-40.

23 Damman K, Voors AA, Navis G, van Veldhuisen DJ, Hillege HL. Current and novel renal biomarkers in heart failure. Heart Fail Rev 2012;17(2):241-50.

24 NICE Guideline No 5: Chronic Heart Failure; www.pathways.nice.org.uk/pathways/chronic-heart-failure.

25 Kelder JC, Cramer MJ, van Wijngaarden J, van Tooren R, Mosterd A, Moons KG, et al. The diagnostic value of physical examination and additional testing in primary care pa tients with suspected heart failure. Circulation 2011;124:2865â&#x20AC;&#x201C;2873.

26 Oudejans I, Mosterd A, Bloemen JA, Valk MJ, van Velzen E, Wielders JP, et al. Clinical evalua tion of geriatric outpatients with suspected heart failure: value of symptoms, signs, and additional tests. Eur J Heart Fail 2011;13:518â&#x20AC;&#x201C;527.

27 Roalfe AK, Mant J, Doust JA, Barton P, Cowie MR, Glasziou P, et al. Development and initial validation of a simple clinical decision tool to predict the presence of heart failure in 35

primary care: the MICE (Male, Infarction, Crepitations, Edema) rule. Eur J Heart Fail 2012;14(9):1000-8.

28 Mant J, Doust J, Roalfe A, Barton P, Cowie MR, Glasziou P, et al. Systematic review and individual patient data meta-analysis of diagnosis of heart failure, with modelling of implications of different diagnostic strategies in primary care. Health Technol Assess. 2009;13(32):1-207, iii.

29 Paulus WJ, Tschope C, Sanderson JE, Rusconi C, Flachskampf FA, Rademakers FE, et al. How to diagnose diastolic heart failure: a consensus statement on the diagnosis of heart failure with normal left ventricular ejection fraction by the Heart Failure and Echo cardiography Associations of the European Society of Cardiology. Eur Heart J 2007;28(20):2539-50.

30 Januzzi JL, Jr. Natriuretic peptide testing: a window into the diagnosis and prognosis of heart failure. Cleve Clin J Med 2006;73(2):149-7.

31 Sato T, Yamauchi H, Suzuki S, Yoshihisa A, Yamaki T, Sugimoto K, et al. Distinct prognostic factors in patients with chronic heart failure and chronic kidney disease. Int Heart J 2013;54(5):311-7.

32 Fent G, Hall I, Watt V, O'Toole L, Al-Mohammad A. 43 Raised Natriuretic Peptides are not Ex clusively Indicative of Heart Failure: A Service Review of a Diagnostic Heart Failure Clinic. Heart 2014;100 Suppl 3:A23-A24.

33 De Vecchis R, Esposito C, Di BG, Ariano C, Giasi A, Cioppa C. B-type natriuretic pep tide-guided versus symptom-guided therapy in outpatients with chronic heart failure: a 36

systematic review with meta-analysis. J Cardiovasc Med (Hagerstown) 2014;15(2):12234.

34 Bruins S, Fokkema MR, Romer JW, Dejongste MJ, van der Dijs FP, van den Ouweland JM, et al. High intraindividual variation of B-type natriuretic peptide (BNP) and aminoterminal

proBNP

patients

with

stable

chronic

heart

failure.

Clin

Chem

2004;50(11):2052-8.

35 Iwanaga Y, Nishi I, Furuichi S, Noguchi T, Sase K, Kihara Y, et al. B-type natriuretic peptide strongly reflects diastolic wall stress in patients with chronic heart failure: comparison between systolic and diastolic heart failure. J Am Coll Cardiol 2006;47(4):7428.

36 Levin ER, Gardner DG, Samson WK. Natriuretic peptides. N Engl J Med 1998;30;339(5):321-8.

37 Balion CM, Santaguida P, McKelvie R, Hill SA, McQueen MJ, Worster A, et al. Physiological, pathological, pharmacological, biochemical and hematological factors affecting BNP and NT- proBNP. Clin Biochem 2008;41(4-5):231-9.

38 Gekle M. Renal tubule albumin transport. Annu Rev Physiol 2005;67:573-94.

39 Palmer SC, Endre ZH, Richards AM, Yandle TG. Characterization of NT-proBNP in human urine. Clin Chem 2009;55(6):1126-34.

40 Pedersen EB, Bacevicius E, Bech JN, Solling K, Pedersen HB. Abnormal rhythmic oscillations of atrial natriuretic peptide and brain natriuretic peptide in chronic renal failure. 37

Clin Sci (Lond) 2006;110(4):491-501.

41 Cortes R, Portoles M, Salvador A, Bertomeu V, Garcia de BF, Martinez-Dolz L, et al. Diagnostic and prognostic value of urine NT-proBNP levels in heart failure patients. Eur J Heart Fail 2006;8(6):621-7.

42 Song J, Li P, Li H, Wang Q. The clinical significance of a urinary B-type natriuretic pep tide as say for the diagnosis of heart failure. Clin Chim Acta 2011;412(17-18):1632-6.

43 Michielsen EC, Bakker JA, Kimmenade RR, Pinto YM, Dieijen-Visser MP. The diagnostic value of serum and urinary NT-proBNP for heart failure. Ann Clin Biochem 2008;45(Pt 4):389-94.

44 Daamen MA, Schols JM, Jaarsma T, Hamers JP. Prevalence of heart failure in nursing homes: a systematic literature review. Scand J Caring Sci 2010;24(1):202-8.

45 Rutten FH, Cramer MJ, Grobbee DE, Sachs AP, Kirkels JH, Lammers JW, et al. Unrecognized heart failure in elderly patients with stable chronic obstructive pulmonary dis ease. Eur Heart J 2005;26(18):1887-94.

46 Chin APM, van Poppel MN, van MW. Effects of resistance and functional-skills training on habitual activity and constipation among older adults living in long-term care facilities: a randomized controlled trial. BMC Geriatr 2006;6:9.

47 Weening-Dijksterhuis E, de Greef MH, Scherder EJ, Slaets JP, van der Schans CP. Frail institutionalized older persons: A comprehensive review on physical exercise, physi

cal fitness, activities of daily living, and quality-of-life. Am J Phys Med Rehabil 2011;90(2):156-68.

48 Grebe C, Brandenburg H. [Resident assessment instrument. Application options and relevance for Germany]. Z Gerontol Geriatr 2015;48(2):105-13.

Chapter 2.1

PREVALENCE AND MISDIAGNOSIS OF CHRONIC HEART FAILURE IN NURSING HOME RESIDENTS: THE ROLE OF B-TYPE NATRIURETIC PEPTIDES. Barents M, van der Horst IC, Voors AA, Hillege JL, Muskiet FA, de Jongste MJ.

Neth Heart J. 2008;16(4):123-8.

ABSTRACT BACKGROUND/OBJECTIVES. Without knowing the exact CHF prevalence, chronic heart failure (CHF) occurs frequently in elderly people both inside and outside nursing homes (NH). For a diagnosis we have to rely on physical examination and additional tests. We therefore run risks of missing CHF diagnoses or of diagnosing CHF when we should not. Natriuretic peptides assays have emerged as a diagnostic test but their use in NH residents is limited. We examined the number of misdiagnoses, the CHF prevalence and the role of natriuretic peptide. METHOD. Residents in one centre without aphasia, cognitive impairments or metastatic cancer were screened on CHF. The natriuretic peptide levels were measured separately. RESULTS. Of the 150 residents, 103 (64%) were included (79 Âą 11 years). The diagnosis of CHF was established in 24 of these 103 residents with NT-proBNP 1871 (interquartile range (IQR) 539-4262) and of BNP 194 (IQR 92-460) pg/mL. A striking result was that of the 24 residents found to have CHF after the screening, 15 (66%) had previously been undetected: NT-proBNP 1146 (IQR 228-3341) and BNP 200 (IQR 107-433) pg/mL). Moreover, in 13 out of 22 (62%) residents who had previously been thought to have CHF, the diagnosis was rejected: NT-proBNP 388 (IQR 174-719) and BNP 90 (IQR 35-128) pg/mL respectively. Regarding

the diagnostic accuracy of NT-proBNP and BNP, the optimal cut-off level of NT-

proBNP was 450 pg/mL with a sensitivity 0.71 and a specificity 0.67, and of BNP was 100 pg/mL with a sensitivity 0.71 and a specificity 0.70. CONCLUSION. Both undetected and incorrect CHF diagnoses were largely observed. NTproBNP and BNP were moderately accurate at CHF diagnosis. The CHF prevalence was 23%.

INTRODUCTION Nursing home (NH) residents are characterized by high age and co-morbidity. Chronic heart failure (CHF) occurs frequently, but is a relatively neglected diagnosis (1). The main reasons why CHF is less frequently detected are non-specific complaints, co-morbidity and limited access to echocardiography (2;3;4). On the other hand, some residents may be diagnosed with CHF, while their symptoms may be caused by other disorders. Both undetected diagnosis and incorrect diagnosis will lead to inadequate treatment of the underlying disease, and will impair the quality of life. Natriuretic peptides are increasingly used to aid the diagnosis of CHF. Its diagnostic value is particularly strong in dyspnoeic patients admitted to the emergency department but less strong in general practices (5). NH residents are mostly of high age and have multiple chronic diseases, such as renal dysfunction and diabetes. The role of natriuretic peptides for the diagnosis of CHF in this specific population may be limited, but so far it has not extensively been studied. In the present study we therefore aim to look at the prevalence of CHF, both undetected and incorrect diagnoses of CHF, and the diagnostic accuracy of NT-proBNP and BNP in NH residents.

METHODS During the course of this study there were 150 persons in the departments for residents with diseases of somatic origin in NH â&#x20AC;&#x153;het Zonnehuisâ&#x20AC;?. Their impairments are based on cardiovascular, pulmonary, neurodegenerative, skeletal muscle and other disorders. Most of them were receiving long-term care (longstay). Some residents with musculoskeletal or cerebrovascular disorders were rehabilitated , one-third of whom could be discharged (short-stay) but should remain care dependent in primary care. Both longstay and short-stay groups were invited to take part between 25 September 2004 and 24 May 2005. Both groups of residents were included if they understood the impact of the study on themselves (competent) and if they agreed to participate by written informed consent. Residents with aphasia or a cognitive impairments, measured by the mini-mental state estimation (MMSE) test, were excluded. The MMSE test contains 30 questions and a MMSE â&#x2030;¤ 20 is suspect for cognitive impairments (6). Persons with metastatic cancer who stayed in a department specialized in palliative terminal care were excluded. Persons were also excluded if the echocardiographic frames could not be sufficiently visualized or if they

refused to have their blood sample taken. CHF was defined according to the definition of the Chronic Heart Failure Guidelines of the European Society of Cardiology (7). Two experienced heart failure cardiologists independently decided on the diagnosis of CHF, based on medical history, physical examination, ECG, routine blood tests and echocardiography. In case of disagreement a third cardiologist reviewed the case and made the final diagnosis. At this stage none of the cardiologist knew the levels of BNP or NT-proBNP. The study was approved of by the Medical Ethical Committee in Groningen, University Medical Center, The Netherlands.

MATERIALS In this cross-sectional study, all data were collected anonymously and within one week (questionnaire, neurohormone sampling, ECG, echocardiography). One physician collected the data of the patients’ medical history (CHF with NYHA classification, coronary artery diseases (CAD), hypertension, peripheral vascular disorders, diabetes mellitus and other comorbidity), of their symptoms (fatigue and dyspnea at great or light effort and in rest, when laying down and at night, breast pain, palpitation), and of their medication. He also examined blood pressure, central jugular pressure (CJP), the heart (third heart sound, gallop rhythm, cardiomegaly by percussion), the lungs (rales), the liver (liver-percussion), length and weight and looked at periphery oedema (8). A 12 leads electrocardiogram (ECG) was made with

the

electrocardiograph

Cardioline

delta

three

plus

(Cardioline,

Milan,

Italy,

www.cardioline.it) with the patient in a horizontal position. One blood sample per resident (12 mL) was taken if one was in fasting condition and at rest. Assessments of creatinine, haemoglobin and mean corpuscular volume (MCV) were performed in the NH laboratory. Since the coefficient of variation of neurohormones is reported to be about 100% we chose not to repeat the neurohormone sampling (9). For determination of levels of (NT-pro)BNP, a 5 μl aprotinine solution was adjusted to the 250 μl plasma samples in EDTA and to 250 μl serum samples. At the UMCG Clinical Chemical Laboratory (CCL) both were frozen at -20˚C and stored in batches

for a maximum of 10 months. The assays were run in one go for both NT-proBNP and BNP. Both NT-proBNP and BNP were measured by immunoassays (Elecsys®1010/2010/modular analytics 2004 Roche diagnostics Indianapolis IN US, and AXSYM system® BNP 2003 Axis– shield diagnostics LTD ABBOT Wiesbaden Germany). NT-proBNP and BNP had coefficients of variation of 3.3% and 7.8% respectively, and measuring ranges of 5-35000 pg/mL and 03465 pg/mL. The renal function has been defined as glomerular filtration rate (GFR) measured by the Cockcroft and Goult formula in mL/min. On the UMCG department of Cardiology, echocardiography was performed by four experienced echocardiographists using a hand held cardiograph ‘Opti Go’ (Philips, Eindhoven, The Netherlands, www.philips.com). The left-ventricular ejection fraction was assessed semi-quantitatively by the two-dimensional visual estimate method (10). A LVEF ≤ 45% was considered to be a left ventricular systolic dysfunction (LVSD). If the echocardiogram was of insufficient quality, the subject was transported to the UMCG hospital where an echocardiogram was produced with the General Electric Vingmed Ultrasound five (www.gemedicalsystems.com, Zoetermeer, The Netherlands).

Statistics Statistical analyses were performed using SPSS 12.0.1 software (SPSS Inc., 233 S. Wacker, 11th Floor, Chicago IL US). Differences of base characteristics of residents with and without CHF were submitted at Student t tests, of gender at Chi-square test and the Mann Whitney test for non-parametric continuous data, as appropriate. ROC curves were made of NT-proBNP and BNP levels. The Area Under the Curve (AUC) presents the test accuracy and the significance of the difference of both AUCs was estimated by a Chi-square test. Influences of age, gender, renin-angiotensin system blocker (ACE-inhibitor or angiotensin receptor blocker) and

renal

dysfunction on (NT-pro)BNP cut-off points were determined by Mann Whitney tests. All statistical comparisons were two-tailed, and a P-value <0.05 was considered to be statistically significant. 45

RESULTS A total of 150 individuals was screened. Fourteen residents refused to participate, 30 were excluded because of cognitive impairment and aphasia. Three persons were excluded because of incomplete data (two poor quality echocardiograms and one refusing the blood sample). The remaining 103 residents were included. The mean age of the population was 78 Âą 11 years. Twenty two persons had CHF before the study. Twenty four persons had CHF after the study and we found 3 groups: subjects diagnosed with CHF before the study whose diagnosis was confirmed afterwards (CHF confirmed), subjects diagnosed with CHF before the study whose diagnosis was rejected afterwards (CHF rejected) and subjects not diagnosed with CHF before the study but diagnosed with CHF after the study (CHF de novo). Of the 24 CHF, nine had CHF confirmed and 15 had CHF de novo. Thirteen subjects had CHF rejected (Table 1). The CHF prevalence was 23% (24/103). Residents with CHF more often had CAD, used a renin-angiotensin system blocker, complained of fatigue, had a raised CJP, had a cardiac thoracic ratio of more than a half, had a third heart tone and or gallop rhythm, had pulmonary rales, had left ventricle hypertrophy, suffered more from renal dysfunction and from left ventricular dysfunction (see Table 2). Residents with CHF after compared to those without CHF had NT-proBNP medians of 1871 and 239 pg/mL (P<0.001) respectively and BNP medians of 194 and 68 pg/mL (P<0.001). Residents with CHF before compared to those without CHF had NT-proBNP medians of 552 and 241 pg/mL.(P=0.013) and BNP medians of 92 and 76 pg/mL respectively (P=0.270) (see Table 3). The predictive values of the two neurohormones presented by the AUC of NT-proBNP (0.815) and of BNP (0.758) did not differ significantly (P= 0.234) as is shown in Figure 1. Table 4 presents the sensitivity, specificity, positive and negative predictive values at several cut-offs points of NT-proBNP and BNP. In search of threshold values with the best combination of excluding and including CHF, threshold values of NT-proBNP at 450 pg/mL (sensitivity 0.71, specificity 0.67, positive and negative predictive values of 0.42 and 0.91) and of BNP at 100 46

pg/mL (sensitivity 0.71, specificity 0.70, positive and negative predictive values of 0.63 and 0.89 respectively) were found. Both BNP and NT-proBNP were influenced by renal function and estimated in subjects with GFR > 75 mL/min and <75 mL/min: NT-proBNP means and SDs were 619 ± 1240 compared to 2010 ± 4191 pg/mL (P =0.005). BNP means and SDs were 110 ± 171 compared to 251 ± 457 pg/mL (P =0.004). BNP and NT-proBNP were influenced significantly by age and renal dysfunction (age and BNP Z=-2,220, P=0.026; renal dysfunction and BNP Z=-2,001, P=0.045; age and NT-proBNP Z=-4,196, P=0.000; renal dysfunction and NT-proBNP Z=-3,663, P=0.000). NT-proBNP and BNP cut-off points were not significantly influenced by gender nor by use of renin angiotensin blockers.

DISCUSSION In this NH study, CHF was present in almost one quarter of the residents. We found that the majority of these residents had previously not been diagnosed with CHF. In addition, there were several residents, who had been diagnosed with CHF before the study, whose diagnosis had to be rejected after careful examination. Blinded values of both BNP and NT-proBNP differed significantly between residents who were diagnosed with CHF and those whose previous CHF diagnosis was rejected. In initial screening, NT-proBNP and to a lesser extent BNP, were moderately accurate in predicting the presence of CHF. A prevalence of 23% in this NH cohort is roughly twice as much compared with the prevalence of CHF in the general population but concomitant with another NH population and with a population of 70 years and over (11-13). Both undetected diagnoses and incorrect diagnoses of CHF were expected, but they exceeded our expectations to a large extent. Reasons for an incorrect diagnosis of CHF were a history of atrial fibrillation (5 cases) and the coexistence of COPD (3 cases). COPD as a reason for misdiagnosis has earlier been described by Rutten (14). In addition, the majority of individuals with rejected CHF had NYHA class I and II (12/13) and one had class III. Eleven out of 15 CHF de 47

novo may have been overlooked because of mild complaints (7 NYHA class I and 4 class II). All 4 other subjects with CHF de novo with NYHA class III and IV were missed because multiple co-morbidity confused their symptoms of cardio-vascular origin. Thus, both incorrect CHF diagnoses and undetected CHF diagnoses seemed to have been caused by the existence of comorbidity. The unspecific clinical presentation of elderly people and the aspecificity of the diagnostic tools add to the number of misdiagnoses. Moreover, this study shows that thorough physical examination in combination with echocardiography can improve the accuracy of the CHF diagnostics. In addition, natriuretic peptides can improve the CHF diagnostics if added to the usual diagnostic procedure even without cardiac ultrasounds. Failures of the expert panel are possible but unlikely, since standard diagnostic tests were performed, and the panel existed of specialized heart failure cardiologists. In addition, NT-proBNP and BNP levels supported both undetected diagnoses and incorrect CHF diagnoses. Both NT-proBNP and BNP were moderately accurate in predicting CHF. However, depending on the cut-offs, the CHF diagnosis will still be missed in a substantial number of residents. Moreover, one should realize that (NT-pro) BNP levels will increase not only with CHF but also in conditions like high age and renal dysfunction. So the use of natriuretic peptides alone will not be sufficient for an adequate diagnosis, although the results have been significantly improved when compared to the baseline situation, especially for physicians not specialized in cardiology (15). If natriuretic peptides are used as an initial test, we suggest NT-proBNP and BNP cut-offs of 450 and 50 pg/mL respectively for excluding CHF, and 900 and 200 pg/mL for establishing CHF. When values are within the cut-off values, a regular assessment should be performed. The NT-proBNP cut-off of 450 pg/mL is suggested as an age specific cut-point. The BNP cut-off of 50 pg/mL is lower than in other studies where 100 pg/mL is proposed. NT-proBNP and BNP were influenced by age and renal dysfunction in accordance with other studies (16-19). In contrast with other studies, they were not influenced by gender nor by the use of renin angiotensin system blockers (20-22). 48

A limitation of the study is the small population of 103 NH-residents in one centre, which could lead to a selection bias in arriving at the prevalence of CHF being 23%. On the other hand, our prevalence of chronic heart failure is concomitant with the prevalence in other NH residents and in the elderly (12;23;24). Moreover, the population studied is representative of other NH populations, in terms of distribution of gender, age, diabetes and the use of ACE-inhibitor therapy, although not representative in terms of the presence of hypertension and renal dysfunction (24;25). Furthermore, despite good criteria, setting the diagnosis of CHF remains difficult because of the lack of diagnostic accuracy of physical examination and the lack of easy access to echocardiography.

CONCLUSION At present the overall accuracy of the diagnosis CHF is limited. In this study both undetected and incorrect CHF diagnoses were observed in a large number of residents. Moreover, the study shows that a more thorough investigation of residents on CHF will lead to more accurate diagnostics, which in turn will improve residents’ quality of life. The use of NT-proBNP and BNP as additional diagnostic instruments is promising also in nursing home residents but needs further evaluation. The CHF prevalence in care-dependent elderly people was almost 25%, which is twice as much as in old people living independently. The use of NT-proBNP and BNP as additional diagnostic instruments is promising also in NH residents but needs further evaluation.

ACKNOWLEDGMENTS This study was subsidised by the “Vereniging het Zonnehuis”, Soesterberg, The Netherlands. Financial disclosure: the authors have nothing to disclose. Author contribution: Dr. Voors was responsible for the echocardiography, Dr. Muskiet was responsible for the biochemistry, Dr. Hillege was responsible for the statistics, Dr. de Jongste was responsible for the design and Drs. 49

Barents was principal investigator and responsible for the manuscript. (Dr. Voors, Dr. van der Horst and Dr. de Jongste were also independent panel members and responsible for the proofreading).

References 1

Cleland JG, Swedberg K, Follath F, Komajda M, Cohen-Solal A, Aguilar JC, Dietz R, Gavazzi A, Hobbs R, Korewicki J, Madeira HC, Moiseyev VS, Preda I, van Gilst WH, Widimsky J, Freemantle N, Eastaugh J, Mason J. The Euro Heart Failure survey pro gramme - a survey on the quality of care among patients with heart failure in Europe. Part 1: patient characteristics and diagnosis. Eur Heart J 2003 March;24(5):442-463.

Luchi RJ, Taffet GE, Teasdale TA. Congestive heart failure in the elderly. J Am Geriatr Soc 1991 August;39(8):810-825.

Januzzi JL, Jr., Camargo CA, Anwaruddin S, Baggish AL, Chen AA, Krauser DG, Tung R, Cameron R, Nagurney JT, Chae CU, Lloyd-Jones DM, Brown DF, Foran-Melanson S, Sluss PM, Lee-Lewandrowski E, Lewandrowski KB. The N-terminal Pro-BNP investigation of dyspnea in the emergency department (PRIDE) study. Am J Cardiol 2005 April 15;95(8):948-954.

Morrison LK, Harrison A, Krishnaswamy P, Kazanegra R, Clopton P, Maisel A. Utility of a rapid B-natriuretic peptide assay in differentiating congestive heart failure from lung disease in patients presenting with dyspnea. J Am Coll Cardiol 2002 January 16;39(2):202-209.

Smith H, Pickering RM, Struthers A, Simpson I, Mant D. Biochemical diagnosis of ventricular dysfunction in elderly patients in general practice: observational study. BMJ 2000 April 1;320(7239):906-908.

Tombaugh TN, McIntyre NJ. The mini-mental state examination: a comprehensive re view. J Am Geriatr Soc 1992 September;40(9):922-935.

Krum H. The Task Force for the diagnosis and treatment of chronic heart failure of the European Society of Cardiology. Guidelines for the diagnosis and treatment of chronic heart failure: full text (update 2005). Eur Heart J 2005 November;26(22):24722474.

Shamsham F, Mitchell J. Essentials of the diagnosis of heart failure. Am Fam Physician 2000 March 1;61(5):1319-1328.

Bruins S, Fokkema MR, Romer JW, Dejongste MJ, van der Dijs FP, van den Ouweland JM, Muskiet FA. High intraindividual variation of B-type natriuretic peptide (BNP) and amino-terminal proBNP in patients with stable chronic heart failure. Clin Chem 2004 November;50(11):2052-2058.

10 Quinones MA, Waggoner AD, Reduto LA, Nelson JG, Young JB, Winters WL, Jr., Ribeiro LG, Miller RR. A new, simplified and accurate method for determining ejection fraction with two-dimensional echocardiography. Circulation 1981 October;64(4):744753.

11 Mosterd A, Hoes AW, de Bruyne MC, Deckers JW, Linker DT, Hofman A, Grobbee DE. Prevalence of heart failure and left ventricular dysfunction in the general population; The Rotterdam Study. Eur Heart J 1999 March;20(6):447-455.

12 Heckman GA, Misiaszek B, Merali F, Turpie ID, Patterson CJ, Flett N, McKelvie RS. Management of heart failure in Canadian long-term care facilities. Can J Cardiol 2004 August;20(10):963-969.

13 Mair J. Monitoring of patients with heart failure. Scand J Clin Lab Invest Suppl 2005;240:99-106.

14 Rutten FH, Cramer MJ, Grobbee DE, Sachs AP, Kirkels JH, Lammers JW, Hoes AW. Unrecognized heart failure in elderly patients with stable chronic obstructive pulmonary disease. Eur Heart J 2005 September;26(18):1887-1894.

15 Cowie MR, Jourdain P, Maisel A, Dahlstrom U, Follath F, Isnard R, Luchner A, McDonagh T, Mair J, Nieminen M, Francis G. Clinical applications of B-type natriuretic peptide (BNP) testing. Eur Heart J 2003 October;24(19):1710-1718.

16 Bando M, Ishii Y, Sugiyama Y, Kitamura S. Elevated plasma brain natriuretic peptide levels in chronic respiratory failure with cor pulmonale. Respir Med 1999 July;93(7):507-514.

17 Buckley MG, Sethi D, Markandu ND, Sagnella GA, Singer DR, MacGregor GA. Plasma concentrations and comparisons of brain natriuretic peptide and atrial natriuretic peptide in normal sub jects, cardiac transplant recipients and patients with dialysis-independent or dialysis-dependent chronic renal failure. Clin Sci (Lond) 1992 October;83(4):437444.

18 Sayama H, Nakamura Y, Saito N, Kinoshita M. Why is the concentration of plasma brain natriuretic peptide in elderly inpatients greater than normal? Coron Artery Dis 1999 Octo ber;10(7): 537-540.

19 Wallen T, Landahl S, Hedner T, Saito Y, Masuda I, Nakao K. Brain natriuretic peptide in an elderly population. J Intern Med 1997 October;242(4):307-311. 53

20 Kohno M, Yokokawa K, Yasunari K, Kano H, Minami M, Hanehira T, Yoshikawa J. Changes in plasma cardiac natriuretic peptides concentrations during 1 year treatment with angiotensin-con verting enzyme inhibitor in elderly hypertensive patients with left ventricular hypertrophy. Int J Clin Pharmacol Ther 1997 January;35(1):38-42.

21 Missouris CG, Grouzmann E, Buckley MG, Barron J, MacGregor GA, Singer DR. How does treatment influence endocrine mechanisms in acute severe heart failure? Effects on cardiac natriuretic peptides, the renin system, neuropeptide Y and catecholamines. Clin Sci (Lond) 1998 June;94(6):591-599.

22 Redfield MM, Rodeheffer RJ, Jacobsen SJ, Mahoney DW, Bailey KR, Burnett JC, Jr. Plasma brain natriuretic peptide concentration: impact of age and gender. J Am Coll Cardiol 2002 September 4;40(5):976-982.

23 Mair J, Friedl W, Thomas S, Puschendorf B. Natriuretic peptides in assessment of leftventricular dysfunction. Scand J Clin Lab Invest Suppl 1999;230:132-142.

24 Gambassi G, Lapane K, Sgadari A, Landi F, Carbonin P, Hume A, Lipsitz L, Mor V, Bernabei R. Prevalence, clinical correlates, and treatment of hypertension in elderly nursing home residents. SAGE (Systematic Assessment of Geriatric Drug Use via Epidemiology) Study Group. Arch Intern Med 1998 November 23;158(21):2377-2385.

25 Kohno M, Yokokawa K, Yasunari K, Kano H, Minami M, Hanehira T, Yoshikawa J. Changes in plasma cardiac natriuretic peptides concentrations during 1 year treatment with angiotensin-con verting enzyme inhibitor in elderly hypertensive patients with left ventricular hypertrophy. Int J Clin Pharmacol Ther 1997 January;35(1):38-42. 54

Table 1. Three chronic heart failure groups found after the study. n

nyha

NT-proBNP

III

2632 (1044-5737)

CHF rejected

388 (174-719)

CHF de novo

1146 (228-3341)

CHF confirmed

BNP 118 (59-706) 90 (35-128) 200 (107-433)

Chronic heart failure, CHF; CHF diagnosed before the study and confirmed afterwards, CHF confirmed; CHF diagnosed before the study and rejected afterwards, CHF rejected; CHF undetected before the study and diagnosed afterwards, CHF de novo; number, n; New York Heart Association, nyha; N-terminal-pro B-type natriuretic peptide, NT-proBNP; B-type natriuretic peptide, BNP; median and interquartile range of pg/mL.

Table 2. Baseline characteristics of nursing home residents. Total age

CHF+ (%)

CHF- (%)

78±11

80±9

78±11

0.351

39 (38)

11 (46)

28 (35)

0.358

40 (39)

13 (54)

27 (34)

0.078

Hypertension (%)

57 (55)

13 (54)

44 (49)

0.680

CAD (%)

28 (27)

14 (58)

14 (18)

<0.001

rhythm disorder (%)

24 (23)

9 (38)

15 (19)

0.060

Diabetes type 2 (%)

26 (25)

9 (38)

17 (22)

0.114

PAD (%)

16 (16)

66 (25)

10 (13)

0.144

ACE and ARB (%)

42 (41)

14 (58)

28 (35)

0.046

Beta blockers (%)

24 (23)

7 (29)

17 (22)

0.438

Ca-antagonists (%)

10 (10)

4 (17)

6 (8)

0.189

Anti-coagulants (%)

63 (61)

19 (78)

44 (53)

0.055

Diuretics (%)

50 (49)

15 (62)

35 (44)

0.118

Anti-diabetics (%)

21 (20)

8 (33)

13 (16)

0.072

Fatigue (%)

40 (39)

15 (52)

25 (32)

0.007

Dyspnea

31 (30)

11 (45)

20 (25)

0.055

vCJP

11 (11)

7 (29)

4 (5)

<.001

Cardiomegaly

22 (21)

10 (42)

12 (15)

0.006

Murmur

24 (23)

9 (38)

25 (32)

0.587

3th tone, gallop

20 (19)

10 (20)

10 (13)

0.587

Rales

17 (17)

9 (40)

8 (8)

<0.001

4 (4)

1 (4)

3 (3)

0.935

40 (39)

15 (60)

35 (44)

0.116

male (%) immobility wheelchair/bed (%)

medical history

Medication

Major symptoms

Physical examination

Hepatomegaly edema

systolic blood pressure

141±22

143 (27)

140 (20)

0.534

diastolic blood pressure

81±13

79 (13)

81 (13)

0.590

body mass index

25± 5

25 (5)

0.897

10 (10)

3 (13)

7 (8)

0.598

5 (5)

2 (8)

3 (28)

0.365

32(31)

16(67)

16(20)

<0.001

Hemoglobin (SD)

7.7±1

0.834

eGFR (SD)

76±32

63±26

80±32

0.022

324 (163-1146)

1871(539-4262)

239 (118-674)

<0.001

87 (28-187)

194 (92-460)

68 (20-123)

<0.001

51±9

43±9

54±6

0.016

Electrocardiogram atrial fibrillation ischemia LVH Laboratory

NT-proBNP md IQR BNP md IQR echocardiogram mn SD

Coronary artery diseases, CAD; peripheral artery disease, PAD; angiotensin converting enzyme inhibitor, ACE-inhibitor and alfa 2-receptor blocker, ARB; anti-coagulants and anti-platelets, anti-coagulants; venous central jugular pressure, vCJP; left ventricular hypertrophy, LVH; estimated glomerular filtration rate, eGFR; N-terminal-pro B-type natriuretic peptide, NT-proBNP; B–type natriuretic peptide, BNP; md, median IQR, interquartile range; SD, standard deviation.

Table 3. Median NT-proBNP and BNP concentrations in residents with and without chronic heart failure (CHF + and CHF -), before and after the study. NT-proBNP

CHF + median (IQR) (n)

CHF -

P-value

median (IQR) (n)

before study

552 (309-2477) (22)

241 (128-892) (81)

0.013

after study

1871(539-4262) (24)

239 (118-674) (79)

<0.001

BNP

median (IQR) (n)

before study

92 (33-187) (22)

76 (23-187) (81)

0.270

after study

194 (92-460) (24)

68 (20-123) (79)

<0.001

median (IQR) (n)

Interquartile range, IQR; number of subjects, n; natriuretic peptides in pg/mL; Before the study, medians of NT-proBNP in residents with and without CHF (CHF+ and CHF -) differed (P=0.013) and medians of BNP in CHF+ and CHF- did not differ (P=0.270). After the study, medians of both NT-proBNP and BNP differed convincingly (NT-proBNP in CHF+ 1871 and in CHF – 239, P<0.001, BNP in CHF+ 194 and in CHF – 68 pg/mL, P<0.001).

Table 4. N-terminal-pro B-type natriuretic peptide (NT-proBNP) and B-type natriuretic peptide (BNP) cut-off values with positive and negative predictive values. Cut-off pg/mL

Sensitivity

Specificity

PPV

NPV

NT-proBNP 450

0.71

0.67

0.42

0.91

NT-proBNP 900

0.67

0.85

0.56

0.89

NT-proBNP 1350

0.58

0.86

0.56

0.86

NT-proBNP 1800

0.46

0.94

0.69

0.85

Cut-off pg/mL

Sensitivity

Specificity

PPV

NPV

BNP 50

0.88

0.63

0.42

0.89

BNP 100

0.71

0.70

0.63

0.89

BNP 150

0.54

0.79

0.69

0.82

BNP 200

0.50

0.87

1.00

0.80

PPV , positive predictive value; NPV, negative predictive value.

Chapter 2.2

UNDERESTIMATED PREVALENCE OF CHRONIC HEART FAILURE AMONG RESIDENTS OF CARE-HOMES IN ARUBA. Stoutjesdijk E, Brouns RM, Barents M, DeJongste MJ, Besselink HJ, Cheng JD, Wever R, Muskiet FA.

West Indian Med J. 2014 Nov 4;63(6). doi: 10.7727/wimj.2013.277.

ABSTRACT OBJECTIVES. We aimed to study the frequency and accuracy of chronic heart failure diagnoses in a group of stable Aruban residential elderly with low physical activity. METHODS. A total of 235 residents were assessed in a cross-sectional designed pilot study. Residents with dementia, aphasia, a history of psychiatry or terminal illness were excluded. Physical examination and electrocardiograms were performed in conjunction with blood samples to determine a./o. B-type natriuretic peptides (BNP) and renal function. A general practitioner and a cardiologist established or withdraw the diagnosis CHF. RESULTS. Out of, 235 residents, 184 were excluded. Fifty-one (22%) subjects, 78Âą8 years of age, were included in the study. CHF was established in 16 [BNP 156 (72-1029) ng/L] out of the 51 residents. Eleven residents were not previously diagnosed with CHF [BNP 152 (611029) ng/L]. Of two out of the seven residents diagnosed with CHF before the screening, the diagnosis CHF was rejected [BNP 71 (59-83) ng/L]. CONCLUSION. In Aruban residential elderly, identification of chronic heart failure is severely underestimated. The detection of CHF was improved by implementing an appropriate guideline and by including the determination of BNP.

INTRODUCTION During the last one-and-half-century, in the industrialized society the average age of dying has doubled. This observation is mainly resulting from introduction of hygienic measures (sewers, water supply, food, life style) and improved treatments such as for ischemic heart diseases. The drawback of increased survival from cardiac diseases is a subsequent growth in sequelae, such as heart failure. In addition, the onset of cardiovascular diseases is adversely affected by the development of an unhealthy lifestyle, as is the case in more prosperous societies. As a result of an insalubrious life style, the prevalence is raising of acquired afflictions like hypertension, diabetes, obesity and hypercholesterolemia more often. In this regard it is worth notifying that Mexican Americans, to which Arubans belong develop atherosclerosis much more frequently most likely because of an imbalance between intake and leisure-time physical activities (1;2). Hence, in Arubans cardiovascular morbidity has become a substantial part of the multimorbidity in the elderly. Multimorbidity is related to care dependency and more help in activities of daily living (3). Accordingly, the most care dependent elderly live in a nursing home, here referred to as residential elderly. It is unknown however likely that a substantial part of Aruban residential elderly suffer from CHF. In Canada and The Netherlands CHF prevails in one to two% of all adults, ten% in a community dwelling elderly and 20% in residential elderly (4-6). The life-time risk to develop CHF is estimated to be 20%. The prognosis of CHF greatly depends on the severity of symptoms (7). For instance, persons with CHF of NYHA class III-IV have an one-year mortality of 60%, those with NYHA class I-II have 20% and half of the latter is deceased within five years (8-10). Heart failure is a syndrome defined as the presence of at least one symptom and a heart disorder. Key symptoms are dyspnea and fatigue, both at rest and during exertion. Heart failure may be caused by hypertension, atherosclerotic heart diseases, valve and rhythm disorders and a variety of cardiomyopathies. When CHF is accompanied by pressure or volume overload, natriuretic peptides, as B-type Natriuretic Peptide (BNP) secretion by the ventricle wall is one of the regulating mechanisms. BNP stimulates natriuresis, diuresis, vasodilatation and inhibition of 63

the renin-angiotensin-aldosterone system - and sympathetic nervous system. BNP is cleared by binding to natriuretic peptide receptors and the half-life time of this binding is 20 minutes. These characteristics make BNP appropriate as marker of CHF. However, many factors increase the secretion of BNP as well (11;12). In residential elderly, it is even more difficult to identify subjects with CHF since there are no CHF guidelines for residential elderly (4;13;14). Moreover, BNP is not validated for elderly ≥ 75 years or for elderly with multimorbidity (15). Finally, access to echocardiography is limited. All these confounders have attributed to regular misjudgment of CHF in Dutch community-dwelling and residential elderly (4;13). In Aruba, it can be predicted that in a population of residential elderly, with a presumably higher incidence of not recognized CHF, comparable difficulties are met. Therefore, we aim to investigate the diagnostic accuracy and role of BNP in diagnosing CHF in Aruban residential elderly with low physical activity.

METHOD Design: this study was designed as multi-centre cross-sectional pilot study. Persons: from February till August 2009, 235 residents live in the “Stichting Algemeen Bejaardenzorg Aruba” located in San Nicolas, Savaneta and Oranjestad. Half the residents were immobile for the most part because they had an amputation. Some were admitted because of lacking family care. All residents were invited to participate in the study. They were included if they signed the informed consent. They were excluded in case of cognitive impairments as dementia (Korsakov’s syndrome), incompetency (one does not understand the impact of the study on himself), aphasia, serious psychiatric and neurological disorders and terminal illness. From the included residents, one of the investigators collected data on medical history and medication. One general practitioner examined the residents on CHF symptoms. He also measured blood pressure and heart rate. He examined the central jugular pressure, cardiac size (percussion) and heart rhythm. He also looked for the presence of murmurs, deviate breath 64

sounds and peripheral edema. A 12-leads electrocardiogram was made at rest (Burdick Eclipse LE (www.Burdick.com)). The general practitioner and cardiologist decided on the diagnosis CHF, independently. The general practitioner suggested the diagnosis CHF based on medical history, physical examination, electrocardiogram and blood tests except BNP. The cardiologist came to the diagnosis CHF based on the same data as the general practitioner but now with BNP included. They discussed all cases and in case of disagreement, the cardiologist prevailed. All data were retrieved anonymously from the resident files. According to CHF guidelines (16;17), BNP is determined when CHF is suspected. When BNP is lower than 100ng/L and the electrocardiogram is not deviating, CHF can be excluded in most cases leaving a small risk of still having CHF of less than 10% (5;18). Laboratory assistants took non-fasting blood samples in tubes prepared with lithium heparin and EDTA. They stored all samples in a cooler and analyzed them within four hours only in the Laboratory. Sodium, potassium, urea, kreatinin, thyroid-stimulating hormone, BNP, hemoglobin, haematocrit and mean corpuscular volume were determined. BNP was determined with the Triage BNP assay (www.biosite.com) at a Beckman Coulter Unicel DxC600i Immunoassay System (www.beckmancoulter.com). BNP had coefficients of variation of <7% with a range of 5 to 4970ng/L. Kreatinin was measured by a standardized isotope dilution mass spectrometry. Renal function was defined by the estimated glomerularfiltration rate (eGFR) measured by the Modification of Diet in Renal Disease formula (mL/min/1.73 m2) = 175 * (Sc/88.4)-1.154 * (age)-0.203 * 0,742 if female *1.212 if negroĂŻd. Serious renal dysfunction was defined as eGFR less than 30 mL/min/1.73m2 (19;20). We requested the laboratory to deliver us the percentage of BNP tests ordered by general practitioners during the study period. Statistical analyses were performed using PASW Statistics 17 software. We compared the characteristics of residents with and without CHF with the Studentâ&#x20AC;&#x2122;s t-test and Mann-Whitney test for nonparametric continuous data, as appropriate (Table 2). The x2- test was used in parametric and the Mann-Whitney test in nonparametric binominal variables. The predictive values of the 65

BNP test was calculated at different cut-off points from 50 â&#x20AC;&#x201C; 200 ng/L as described in Table 3. Influences of age, gender, angiotensin-converting enzyme inhibitor (ACE-i) and eGFR on BNP were examined with linear regression analyses with a log transformation. All statistical comparisons were two-tailed, and a p value less than 0.05 was considered to be statistically significant.

RESULTS Out of all 235 residents, 184 (78%) were excluded for a variety of reasons such as dementia in half of them. Out of 184 residents, 165 were not eligible according the exclusion criteria. And, 19/184 dropped out because of refusal to participate (eight), cognitive impairments (five), admission to a hospital beforehand (four), incomplete data (one), and for not living in a nursing home (one). The remaining 51 (22%) were included and screened. The included residents were aged 78 (range 56-93) years and 29 were females with 22 males. The majority (55%) of the individuals was immobilized and bound to wheelchair or bed (55%), as a consequence of various disorders, such as cerebrovascular accidents (31%), CHF (31%) and serious sequelae of T2DM (59%), like amputation of a leg. The estimated GFR was 46Âą22 mL/min. Chronic heart failure was diagnosed in sixteen residents (16/51, 31%). We distinguished four groups: seven residents were judged to have CHF before the screening (CHF before study). In two of these seven, the CHF diagnoses were not confirmed (CHF rejected). In five of these seven, the CHF diagnoses were confirmed (CHF confirmed). In addition, eleven CHF diagnoses were first established after the screening while they were undetected before (CHF de novo) (Figure 1 and Table 2) The cardiologist overruled the general practitioner in 4 out of 16 residents in which there was disagreement with regard to CHF (BNP 134, 392, 440, 459ng/mL).

Comparing the 16 residents with CHF to those without, the first group used more anticoagulants and suffered more from dyspnea, fatigue and edema. In addition, the CHF group had higher BNP levels and a lower renal function, on average. Regarding the ECG observations, residents with CHF had more LVH and atrial fibrillation compared to those without CHF. And, within the other ECG disorders, residents with and without CHF contained two and four atrial-ventricular conduction disorders, zero and two rhythm disorder and seven and six implanted artificial cardiac pacemaker, respectively and none significantly. One resident with CHF had LVH and an atrial-ventricular conduction disorders. The predictive values of BNP, calculated at different cut-off points from 50 to 200ng/L with each 25 ng/L difference in between, were presented in Table 3. Of the included confounders, we found renal function inversely related to BNP (Î˛= -0.49, p=0.002) while age, gender and ACE-i did not influence BNP levels, significantly (p=0.452, p=0.314, p=0.787, respectively). During the study period, general practitioners ordered 6-15% of all ordered BNP tests.

DISCUSSION In Aruban residential elderly, the CHF frequency was unknown and undetected CHF was presumed because of several reasons. We aimed to investigate the frequency and diagnostic accuracy in diagnosing CHF in that population. Therefore, all eligible residential elderly were screened on CHF in three nursing homes. We found that the CHF frequency was about onethird of the residents. And, that in the majority of them the diagnosis of CHF had not been made previously. Undetected diagnoses were expected but the number exceeded our expectations. Reasons for undetected CHF diagnoses were the lack of consensus on the sequence of the diagnostic procedures, incomplete data documentation in the medical files and limited use of BNP testing by general practitioners.

The main etiology of CHF existed of atherosclerosis. Arguments for this assumption were high coexistence of T2DM in residents with CHF and physical inactive lifestyle. Quite a percentage of residential elderly had T2DM which was well documented. This occurrence corresponded to Palloni et al. who found an high or higher frequency of T2DM in residential elderly of 60+ years living in the Caribbean compared to those from the U.S.A.. And, the high T2DM frequency corresponded to a study in the Netherland Antilles. In that study, T2DM was also related to gender (more females), social status and a Western lifestyle. Near the high T2DM frequency, the studied population was characterized, surprisingly, by a residentsâ&#x20AC;&#x2122; mean age comparable to the Aruban life expectancies of 75 and 76 years in 2009 and 2011, respectively (www.indexmundi.com). So, it is likely that our population represented the Aruban elderly as regards T2DM occurrence and age. However, our population represented only the Aruban residential elderly as regards the distribution over the Island. Although the population was too small to speak of predictive values of BNP, the calculated sensitivity and specificity were in line with a comparable study on CHF diagnosis in NH residents of similar age (4) (Table 3). And, from the estimated confounders, BNP was only inversely influenced by renal function which was to be expected from the moderate to severe renal dysfunction (23;25). Thus, these findings do not limit the additional value of BNP to the CHF diagnostics in residential elderly. The sense of improving CHF diagnoses for residential elderly was a better quality of life. Since, more correct CHF diagnoses will lead to adequate treatment, ADL help and planning of care givers (3;26). And, from another view, less unjust CHF diagnoses save unnecessary treatments and costs. Altogether, we observed that the carefully medical work-up according to the applied guideline improved the CHF diagnostics in residential elderly (17). This study had some shortcomings. The studied residents did not represent all since widely half of them were demented and the characteristics of all excluded could not have been uncovered because of ethical reasons. However, the studied residents represented Aruban elderly as re68

gards T2DM occurrence and age. Therefore, we could not rule out a selection bias. So, we cited a frequency of CHF rather than prevalence. The lack of a panel as standard of diagnosing CHF was another shortcoming. This could have influenced the CHF frequency but to a limited extent because the cardiologist was specialized in heart failure and the diagnosis CHF was made up according to the guideline including BNP (17). Main limitation of this pilot study was the small population. In the future, a study is recommended on the detection of CHF in a representative cohort of residential elderly.

CONCLUSION In Aruban residential elderly, identification of chronic heart failure is severely underestimated. The detection of CHF was improved by implementing an appropriate guideline including determination of BNP.

ACKNOWLEDGMENTS We thank Stichting Algemeen Bejaardenzorg Aruba, dr. Horacio E. Oduber Hospitaal and Centro Medico: dr. Rudy Engelbrecht for their participation in this study. None of the authors reported conflicts of interest.

References 1

Bielak LF, Yu P, Ryan KA, Rumberger JA, Sheedy PF, Turner ST, et al. Differences in preva lence and severity of coronary artery calcification between two non-Hispanic white populations with diverse lifestyles. Atherosclerosis 2008;196(2):888-95.

Kurian AK, Cardarelli KM. Racial and ethnic differences in cardiovascular disease risk factors: a systematic review. Ethn Dis 2007;17(1):143-52.

Barents M, Boersma F, Van Der Horst ICC, Hillege HH, de Jongste MJL. Increase of Help at Activities of Daily Life Given to Frail Elderly With Chronic Heart Failure in a Nursing Home. Activities, Adaptation & Aging 2011;35(2):98-110.

Barents M, van der Horst IC, Voors AA, Hillege JL, Muskiet FA, de Jongste MJ. Prevalence and misdiagnosis of chronic heart failure in nursing home residents: the role of B-type natriuretic peptides. Neth Heart J 2008;16(4):123-8.

Rutten FH, Cramer MJ, Grobbee DE, Sachs AP, Kirkels JH, Lammers JW, et al. Un recognized heart failure in elderly patients with stable chronic obstructive pulmonary

dis

ease. Eur Heart J 2005;26(18):1887-94.

Heckman GA, Misiaszek B, Merali F, Turpie ID, Patterson CJ, Flett N, et al. Manage ment of heart failure in Canadian long-term care facilities. Can J Cardiol 2004;20(10):963-9.

Barents M, Hillege HH, van der Horst IC, de Boer RA, Koster J, Muskiet FA, et al. BNP and NT-proBNP, predictors of 1-year mortality in nursing home residents. J Am Med Dir Assoc 2008;9(8):580-5.

Cowie MR, Fox KF, Wood DA, Metcalfe C, Thompson SG, Coats AJ, et al. Hospitalization of patients with heart failure: a population-based study. Eur Heart J 2002;23(11):877-85.

Bleumink GS, Knetsch AM, Sturkenboom MC, Straus SM, Hofman A, Deckers JW, et al. Quantifying the heart failure epidemic: prevalence, incidence rate, lifetime risk and prognosis of heart failure The Rotterdam Study. Eur Heart J 2004;25(18):1614-9.

Ho KK, Anderson KM, Kannel WB, Grossman W, Levy D. Survival after the onset of congestive

heart

failure

Framingham

Heart

Study

subjects.

Circulation

1993;88(1):107-15.

Januzzi JL, Jr. Natriuretic peptide testing: a window into the diagnosis and prognosis of heart failure. Cleve Clin J Med 2006;73(2):149-7.

Beleigoli A, Diniz M, Nunes M, Barbosa M, Fernandes S, Abreu M, et al. Reduced brain natriuretic peptide levels in class III obesity: the role of metabolic and cardiovascular factors. Obes Facts 2011;4(6):427-32.

Rutten FH, Grobbee DE, Hoes AW. Differences between general practitioners and cardiologists in diagnosis and management of heart failure: a survey in every-day practice. Eur J Heart Fail

2003;5(3):337-44.

Voors AA, Walma EP, Twickler TB, Rutten FH, Hoes AW. [Multidisciplinary guideline 'Heart failure 2010']. Ned Tijdschr Geneeskd 2011;155:A2957.

Dickstein K, Cohen-Solal A, Filippatos G, McMurray JJ, Ponikowski P, Poole-Wilson PA, et al. ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure 2008: the Task Force for the Diagnosis and Treatment of Acute and Chronic Heart Failure 2008 of the

European Society of Cardiology. Developed in collaboration

with the Heart Failure Association of the ESC (HFA) and endorsed by the European Society of Intensive Care Medicine (ESICM). Eur Heart J 2008;29(19):2388-442.

van LJ, Rutten FH, Walma EP, Wiersma T, Goudswaard AN. [Summary of the practice guide line 'Heart failure' (first revision) from the Dutch College of General Practitioners]. Ned Tijdschr Geneeskd 2005;149(48):2668-72.

Cleland JG, Swedberg K, Follath F, Komajda M, Cohen-Solal A, Aguilar JC, et al. The EuroHeart Failure survey programme- a survey on the quality of care among patients with heart

failure in Europe. Part 1: patient characteristics and diagnosis. Eur Heart

J 2003;24(5):442-63.

Kenchaiah S, Evans JC, Levy D, Wilson PW, Benjamin EJ, Larson MG, et al. Obesity and the risk of heart failure. N Engl J Med 2002;347(5):305-13.

20 Levey AS, Stevens LA, Schmid CH, Zhang YL, Castro AF, III, Feldman HI, et al. A new equation to estimate glomerular filtration rate. Ann Intern Med 2009;150(9):604-

12.

Palloni A, McEniry M. Aging and health status of elderly in Latin America and the Caribbean: preliminary findings. J Cross Cult Gerontol 2007;22(3):263-85.

22 Grievink L, Alberts JF, O'Niel J, Gerstenbluth I. Waist circumference as a measure ment of obesity in The Netherlands Antilles; associations with hypertension and diabetes mellitus. Eur J Clin Nutr 2004;58(8):1159-65.

23 DeFilippi C, van Kimmenade RR, Pinto YM. Amino-terminal pro-B-type natriuretic peptide test

ing in renal disease. Am J Cardiol 2008;101(3A):82-8.

24 deFilippi CR, Christenson RH. B-type natriuretic peptide (BNP)/NT-proBNP and renal function: is the controversy over? Clin Chem 2009;55(7):1271-3.

25 van Kimmenade RR, Januzzi JL, Jr., Bakker JA, Houben AJ, Rennenberg R, Kroon AA, et al. Renal clearance of B-type natriuretic peptide and amino terminal pro-B-type natriuretic peptide a mechanistic study in hypertensive subjects. J Am Coll Cardiol 2009;53(10):884-90.

235 residents

184 excluded

51 included

16 (31%) CHF

2 CHF rejected

5 CHF confirmed

11 CHF de novo

35 (69%) no CHF

After the study

7 CHF before the study Figure 1. Flow chart of included and excluded residents, and of CHF diagnoses before and after the study in Aruba. Chronic heart failure, CHF.

Table 1. Characteristics of 51 residents. CHF+ Age (years) mn (SD)

(%)

CHF-

(%)

n=16

n=35

value

79 (9)

78 (8)

0.460

Female

0.248

Wheelchair/bed

0.288

Cerebrovascular accident

0.517

Coronary artery disease

0.298

Diabetes Mellitus type 2

0.151

0.018

Diuretics

0.211

Calcium-ion antagonists

0.680

Beta-blockers

0.287

ACE-i or AT2Ri

0.533

Heart glycosides

0.665

0.049

Dyspnea

0.020

Fatigue

0.009

Edema

0.020

Systolic tension mm Hg

135

133

0.782

Diastolic tension mmHg

0.378

Heart rate beats/min.

0.133

Increased CVP

0.002

Cardiomegaly

0.164

Pacemaker

0.243

0.011

Medical history

Medication Antidiabetics

Anticoagulants

Symptoms, signs

Electrocardiogram Normal electrocardiogram

Left ventricular hypertrophy

0.041

Atrial fibrillation

0.413

Ischemia

0.917

Other

0.343

BNP ng/L md (range)

156

72-1029

9-191

0.008

Hb mmol/L mn (SD)

7(1)

8(1)

0.290

MCV femtoL mn (SD)

91 (6)

90 (6)

0.763

33(20)

58(24)

0.001

1(1)

2(1)

0.346

Laboratory data

eGFR mL/min/1.73mÂ˛ n(SD) TSH mE/L mn (SD)

Numbers, n and percentage unless other indicated; mn, mean and SD, standard deviation ; CVP, central jugular pressure; ACE-I, angiotensin-converting enzyme inhibitor; AT2Ri, angiotensin II receptor inhibitor; other, rhythm or conduction disorder, pacemaker or no-specified disorder.

Table 2. CHF diagnoses after the screening with matching BNP values. N residents B-type natriuretic peptide median (ng/L) chronic heart failure rejected

71 (59 – 83)

chronic heart failure confirmed

105 (84 – 392)

chronic heart failure de novo

152 (61 – 1029)

Table 3. Predictive values of BNP at different cut-off points. BNP cut-off point ng/L

sensitivity

1.00

specificity

PPV

NPV

0.37

0.42

1.00

0.64

0.54

1.00

100

0.75

0.69

0.52

0.86

125

0.63

0.74

0.53

0.81

150

0.56

0.89

0.69

0.82

200

0.44

1.00

0.80

PPV, positive predictive value; NPV, negative predictive value; cut-off points are rounded numbers.

Chapter 2.3

HIGH INTRA-INDIVIDUAL VARIATION OF AMINO-TERMINAL PROBRAIN NATRIURETIC PEPTIDE (NT-PROBNP) IN URINE OF PATIENTS WITH STABLE CHRONIC HEART FAILURE. COMPARISON WITH PLASMA. Schimmel AM, Barents M, de Jongste MJL, Rรถmer JWP, Steward HN, Muskiet FAJ.

Accepted for publication in part in Clinical Chemistry.

ABSTRACT BACKGROUND. The large intra-individual variation (CVi) of plasma N-terminal prohormonebrain natriuretic peptide (NT-proBNPp) limits its applicability for guided therapy optimization in individual patients with chronic heart failure (CHF). We investigated whether urine NT-proBNP (NT-proBNPu) has lower CVi compared with NT-proBNPp. METHODS. CVis were measured in 25 patients with stable CHF (NYHA classes I-III). Blood and urine samples were collected on a single day (6 2-h blood samples and spontaneously voided urines during 24-h), 5 consecutive days (5 blood samples, 5 enforced urine samples, 5 full 24-h urines) and 6 consecutive weeks (6 blood samples, 6 enforced urine samples, 6 full 24-h urines). NT-proBNP was measured by immunoassay. Within-day, day-to-day and weekto-week total CVs (CVts), CVis and reference change values (RCVs) were calculated. RESULTS. Median age was 61 y (range 36-80), 60% was male and meanÂąSD left ventricular ejection fraction was 36Âą15%. Median CVis for NT-proBNPp were 9 (within-day), 18 (day-today) and 30% (week-to-week). For NT-proBNPu they were 34, 21 and 28%, respectively. RCVs of NT-proBNPu, NT-proBNPu/creatinineu and NT-proBNPu/h were higher than, or equal to, those of NT-proBNPp. NT-proBNPu did not correlate with NT-proBNPp up to 310 pg/mL NTproBNPp, but related almost linearly from thereon. NT-proBNPp increased during the day, while NT-proBNPu tended to decrease. CONCLUSION. Measurement of NT-proBNP in urine provides no advantages over plasma NTproBNP for therapy optimization in CHF patients.

INTRODUCTION Improvements in diagnostics and treatment have contributed considerably to the increasing survival of patients with cardiac diseases (1). As a consequence, also the prevalence and hence the impact on society of chronic heart disease, notably chronic heart failure (CHF), has grown tremendously (2). CHF has become a major health care issue in therapy, hospital admissions and costs. Measurements of biomarkers may favourably affect the cost-benefit ratio. An ideal biomarker has high diagnostic sensitivity and specificity, should be a true reflection of (patho)physiological changes, has low analytical variation (CVa), and is applicable in (sub)groups irrespective of their demographic background (3). Natriuretic peptides, notably brain natriuretic peptide (BNP) and amino-terminal pro-brain NP (NT-proBNP) are well established CHF biomarkers. They are widely used for CHF diagnosis, follow-up and prognosis (4), while they also provide opportunities for biomarker-guided treatment optimization (5-8). Metaanalyses have shown that natriuretic peptide (NP)-guided therapy is superior to symptomguided therapy in groups (5-7). However, because of high intra-individual variations compared to expected changes, it is not possible to establish whether longitudinal changes of BNP or NTproBNP in individual patients differ with 95% certainty following therapy-adjustment (8).

BNP has natriuretic, diuretic and vasodilator properties (9;10). NT-proBNP becomes cosecreted in equimolar amounts, but is physiologically inactive (11). Both are released by cardiac atria and ventricles upon myocyte stretch, mostly due to volume overload (9). Like many, if not all, hormones, natriuretic peptides exhibit pulsatile secretion patterns (12). Plasma BNP has a half-life of about 20 min, while that of NT-proBNP is about 120 min. BNP is rapidly deactivated in the circulation and tissues by neuropeptidases (10). In contrast, NT-proBNP becomes passively filtered in the kidneys to become excreted in urine, notably in the form of smaller peptides that are at least in part immunoreactive (13). In the brush border of the renal tubules lowmolecular-weight proteins may become hydrolyzed to smaller peptides and amino acids that are subsequently reabsorbed. Usually, this process is nearly complete, suggesting that only minor 81

amounts reach the finally voided urine (14;15). Another mechanism is by endocytosis followed by lysosomal degradation, secretion via either apical or basolateral membranes, or by transcytosis of the intact protein (16). With increasing plasma, and thereby filtered, concentrations all enzymatic and active processes may ultimately be expected to reach saturation. From this threshold, the renal clearance may be expected to increase to finally reach the glomerular filtration rate.

We have previously established that plasma NT-proBNP (NT-proBNPp) in patients with CHF exhibits high within-day (mean CVi 9%; range 0-20%), day-to-day (CVi 20%; 3-80%) and week-to week (CVi 35%; 8-103%) intra-individual variations (CVis) (17). This high CVi, calculated by correcting total variation (CVt) for analytical variation (CVa, 3%), hampers the usefulness of NT-proBNPp for treatment optimization of individual HF patients. Establishment of a genuine difference between pre and post-treatment NT-proBNP concentrations requires information of the minimal change value (also named reference change value; RCV). The RCV for 95% certainty is calculated by 2.8*CVt in which CVt is composed mostly of CVi in case of NTproBNPp (17).

Measurement of immunoreactive NT-proBNP in urine voidings (abbreviated NT-proBNPu) may provide us with a tool that is less affected by the pulsatile secretion of NT-proBNP into plasma. As a long term plasma ultrafiltrate, NTproBNPu may exhibit less intra-individual variation, at least theoretically (18-20), but there may be many confounding factors. Measurements in urine may especially be useful in CHF patients with NTproBNPp concentrations beyond the renal reabsorption threshold. In this study we calculated CVis for NT-proBNP in both plasma and urine of patients with stable CHF. To do so we collected corresponding plasma and urine samples for the establishment of the within-day, day-to-day and week-to-week CVt, CVa, CVi and RCV. The ultimate goal was to predict whether measurements in urine would allow treatment optimization for the individual patient at the level of 95% certainty. 82

PATIENTS AND METHODS The study was conducted from March–July 2008 in the island of Curaçao (12° 12' N, 68° 5' W). The protocol was approved by the Medical Ethics Committee of the St. Elisabeth Hospital and was in agreement with local ethical standards and the Helsinki Declaration of 1975, as revised in 2008. All patients were assured that their data were anonymized. Written informed consent was obtained from all of them.

Study design Within a study period of six weeks we collected blood and urine samples during home visits. Participants were asked whether they preferred to participate in one, two or all sampling protocols. These protocols were: 1) a within-day protocol; 6 blood samples (taken every 2-h starting at 08:00 and ending at 18:00) and collection of spontaneously voided urine portions during 24h (starting 08:00 after discarding an enforced voiding and ending with a collected enforced voiding at 08:00 the next day), 2) a day-to-day protocol; one blood sample (collected from 08:00 to 10:00), an enforced urine portion (immediately after blood sampling) and a full 24-h urine taken during 5 consecutive days, and 3) a week-to-week protocol; one blood sample (collected from 08:00 to 10:00 each week, an enforced urine portion (immediately after blood sampling) and a full 24-h urine, on the same day of the week during 6 consecutive weeks. Clock times of urine voidings were documented. The sampling protocols were integrated where possible, to limit the number of sampling moments. For example, the 08:00 samples of the within-day protocol were also part of the day-to-day protocol and samples of one sampling day of the day-to-day protocol was also part of the week-to-week protocol (Figure 1).

Patients CHF patients of 18 years and older were recruited from two cardiological practices. The inclusion criteria were: stable CHF (defined as unchanged CHF treatment during the last two months) living in the vicinity of the hospital and speaking Dutch or English. The exclusion criteria were obstructive heart valve diseases, obstructive or restrictive heart diseases, potentially transient causes of HF (e.g. acute myocarditis), myocardial infarction, unstable angina, uncontrolled arrhythmias and pregnancy or lactation. Information on date of birth, body mass index and use of medication was collected at baseline. The most recent New York Heart Association (NYHA) classes and the left ventricular ejection fractions (LVEF) were retrieved from the medical files. In addition to the NYHA classification, we established whether there was systolic or diastolic CHF, defined as having a LVEF ≥45%, or >45%, respectively (21).

Samples and analyses Ten mL EDTA-anticoagulated blood samples were collected by venapuncture. They were stored at 4°C until further handling in the laboratory. Urine voidings and full 24-h urines were collected in containers and immediately stored in cool boxes. Blood samples were centrifuged within 2 h after sampling (1,500 g-force, 15 min, 4 °C). Volumes of the spontaneous voidings, enforced voidings and full 24-h urines were measured. Ten mL of each urine portion was taken and centrifuged (1,500 g-force, 15 min, 4 °C). Plasma and urine samples were aliquoted and stored in Eppendorf tubes at -80 °C until transportation to The Netherlands. NT-proBNP was analyzed within nine months after collection in the Clinical Chemical Laboratory of the University Medical Center Groningen, The Netherlands. Plasma and urine samples of a single patient were analyzed in a single series to minimize the influence of the CVa. For measurements of NT-proBNP we applied the sandwich technique [Elecsys 2010 immunoassay system, Roche Diagnostics, Mannheim, Germany (22)]. Creatinine was measured in all urine samples (creatinineu). Creatinine was also measured in the firstly collected plasma sample of each patient to calculate the estimated glomerular filtration rate (eGFR). In these calculations we used the factor of 1.112 84

for patients with African-American backgrounds (23). NT-proBNPu concentration (pg/mL) was also expressed as NT-proBNPu/creatinineu (pg/nmol) and NT-proBNPu/h (pg/h). NTproBNPu/NT-proBNPp represents the apparent filtration fraction of NT-proBNP.

The detection limit of the NT-proBNP immunoassay was 5 pg/mL. An NT-proBNP <5 pg/mL was documented as 0 pg/mL. The within-series CVa of NT-proBNPp was 3% (17). Since NTproBNPu ranges from about 5-50 pg/mL, we used 4 quality-control samples with NT-proBNPu concentrations of 8, 10, 42 and 48 pg/mL. These samples were selected from the total patient sample collection. They were analyzed six times to calculate the CVa of NT-proBNPu. The CVa of creatinineu was calculated from all quality control measurements of creatinineu in the laboratory as performed in the month preceding the NT-proBNP analyses.

Statistical analysis Data of participants sampled three times or less within a sampling protocol were excluded from evaluation in that specific protocol. The remaining data were processed with Microsoft Office Excel 2007 and statistically analyzed using SPSS (SPSS 20.0; SPSS, Inc). Results were expressed as mean±SD or medians (range). The relations of NT-proBNPp and NT-proBNPu with the patients’ characteristics were analyzed by Spearman’s correlation tests. Spearman’s correlation tests were also employed for correlating NT-proBNPp with NT-proBNPu, using all data from the day-to-day and week-to-week samples.

For comparison of within-day NT-proBNPu with NT-proBNPp, we employed the spontaneously voided urine portion within 2 h following venapuncture. For example, the urine portion of 10:00 should be voided from 10:00 to 12:00, the portion of 12:00 from 12:00 to 14:00 etc. In case of two spontaneously voided urine portions obtained within 2 h, the mean NT-proBNPu concentration of those samples was taken. If no urine was voided within 2 h, the NT-proBNPu of 10:00 was labelled as missing. The Shapiro-Wilk test was used to examine the distribution of NT85

proBNP at different clock times for normality. NT-proBNPp, NT-proBNPu and NT-proBNPu/NTproBNPp ratio at 8:00 and 18:00 were compared with Wilcoxon rank-sum test. NT-proBNP levels at six different clock times within one day were analyzed for differences between those six measurements using Friedman’s’ test. To identify differences between NT-proBNPp, NTproBNPu and NT-proBNPu/NT-proBNPp measurements from the same patient at different clocktimes, we used the Wilcoxon signed-rank test. The outcomes were corrected for type 1 errors (p<0.01).

The plasma and urine total variations (CVt) of NT-proBNP for each patient and each of the 3 sampling protocols were calculated. CVi was calculated using (CVi)2 = (CVt)2 – (CVa)2 (17). Reference change values (RCVs) at 95% probability were calculated from median CVas and CVis, according to: RCV = Z × 21/2 × (CVa2/na + CVi2/ns)1/2 (22;24;25), where Z is 1.96 (i.e. Z-score for 95% CI); na is the number of replicate assays; and ns the number of patient samples needed to estimate each of the two homeostatic set points, ns=1 in the present study. RCVs reflect the minimum percentage change in serial results that is different from the combined analytical and biological variation, with 95% confidence. The CVis of the within-day, dayto-day and week-to-week data were compared with the Wilcoxon signed-rank test. Only for these comparisons the significance was set at p<0.01. All other significances were set at p<0.05 (see above).

RESULTS Patients and samples Twenty-five patients (mean age 61±10 years, 60% males) were included into the study (Table 1). Fifteen patients (60%) participated in all three sampling protocols. The main reasons for not participating in all protocols were interference with work or holidays. All together, we were able to collect samples from 20 patients participating in the within-day protocol, 18 patients following 86

the day-to-day protocol and 22 patients completing the week-to-week protocol (Figure 2). This translated into 329 out of the maximum of 342 blood samples (96%), 187 spontaneously voided urine portions, 212 of the maximum of 222 enforced urine voidings (95%), and 207 of the maximum of 222 full 24-h urines (93%). Four participants (16%) exhibited NT-proBNPu levels below the detection limit of 5 pg/mL. These levels were set at 0 pg/mL. In two of these four patients, all NT-proBNPu values were set at 0 pg/mL. The NYHA classes of these two patients were II and unknown. In the other two patients 89% and 50% of the NT-proBNPu values were below the detection limit and their NYHA classes were II and III, respectively.

Within-day, day-to-day and week-to-week intra-individual coefficients of variation The calculated CVas of NT-proBNPu quality-control samples were 12% at 8 pg/mL, 15% at 10 pg/mL, 3% at 42 pg/mL and 3% at 48 pg/mL. The CVa of NT-proBNPu was inversely related to NT-proBNPu (r= -0.983, p<0.05). The CVa for NT-proBNPu levels between 5 and 48 pg/mL was derived from CVa (%) = 119.67*(NT-proBNPu)-1.0016. For NT-proBNPu levels above 48 pg/mL, a CVa of 3% was used. The CVa of creatinineu, calculated from all measurements in a single month, was 2%. The median CVis for NT-proBNPp in the within-day, day-to-day and week-to-week protocols were 9, 18 and 30%, respectively (Table 2). The corresponding median CVis for NT-proBNPu were 34, 21 and 28%, respectively. The within-day CVi of NT-proBNPu was higher than the corresponding CVi of NT-proBNPp (p=0.003). This was also the case for the CVis of NTproBNPu/creatinineu (p=0.002) and NT-proBNPu/h (p<0.001). The within-day CVi of NTproBNPu/h (pg/h) was higher than that of NT-proBNPu (pg/mL; p=0.031). For the day-to-day and week-to-week protocols, the CVis of NT-proBNPu did not differ from the CVis of NTproBNPp. For the day-to-day protocol, the CVi of NT-proBNPu/h was higher compared to the CVi of NT-proBNPu (p=0.044). The same statistical differences as for the CVis are applicable for the comparisons between RCVs, because of the proportionality between these two parameters (Table 2). 87

Correlations Baseline NT-proBNPp (i.e. firstly collected blood sample of each patient) was inversely related to body mass index (r= -0.548, p=0.005). Inversely related to LVEF were: baseline NTproBNPp (r= -0.599, p=0.009), baseline NT-proBNPu (i.e. the firstly voided urine) (r= -0.596, p=0.009), baseline NT-proBNPu/creatinineu (r= -0.481, p=0.043), and baseline NTproBNPu/h (r= -0.538, p=0.021). Baseline NT-proBNPp and baseline NT-proBNPu were not related to other patient characteristics, such as age, eGFR and NYHA class. Related to baseline NT-proBNPp

were:

proBNPu/creatinineu

baseline

NT-proBNPu

(r=0.804,

p=<0.001),

(r=0.841, and

p=<0.001),

baseline

NT-proBNPu/h

NT-

(r=0.835,

p=<0.001).

When the data of the day-to-day and week-to-week protocols were pooled, NTproBNPp correlated with NT-proBNPu in the corresponding enforced urine voidings (r=0.739, p<0.001) and also with NT-proBNPu in the full 24-h urines (r=0.735, p<0.001). NT-proBNPu correlated also with NT-proBNPu/creatinineu (r=0.932, p<0.001) and NT-proBNPu/h (r=0.950, p<0.001). Figure 3 shows the relation between NT-proBNPp and NT-proBNPu in the corresponding enforced urine voidings collected in the day-to-day and week-to-week protocols. Initially, NTproBNPp did not correlate with NT-proBNPu. However, there was a clear relation from an NTproBNPp of about 500 pg/mL, obeying the equation NT-proBNPu = 0.0226 Ă&#x2014; (NTproBNPp)1.1563. The median NT-proBNPu in the lower NT-proBNPp region was 17 pg/mL. Substituting this value into the equation gave an estimated intersection at an NT-proBNPp of 310 pg/mL.

Analysis of a subgroup with NT-proBNPp of 310 pg/mL We investigated whether the RCVs for NT-proBNPp were better for concentrations beyond 310 pg/mL, i.e. the concentration from which NT-proBNPu correlated with NT-proBNPp (Figure 3). 88

This appeared not to be the case. Except for a decrease in RCVs for week-to-week NTproBNPp (from 83 to 59%) and NT-proBNPp/creatinineu (from 113 to 107%), and day-to-day NT-proBNPu/h (87 to 72%), all other RCVs increased while that of the day-to-day NTproBNPp remained constant (Supplemental Table 1).

Changes of NT-proBNP in plasma and urine during the day The within-day courses, relative to baseline, for NT-proBNPp, NT-proBNPu and NTproBNPu/NT-proBNPp, all calculated from the concentrations in pg/mL, are shown in Figure 4A-4C. Concomitant p-values are in Supplemental Table 2. Data were expressed relative to levels at 08:00 h (set at 100%). Within-day NT-proBNPp increased from 08:00 to 18:00 (Wilcoxon rank-sum p=0.006; Friedman p<0.001). No other changes were detected (Figure 4A). No concomitant changes were observed for the within-day NT-proBNPu. The apparent decrease of NT-proBNPu from 08:00 to 18:00 was insignificant (Figure 4B). There were also no changes in NT-proBNPu/creatinineu and NT-proBNPu/h from 08:00 to 18:00. NT-proBNPu/h fluctuated from 12:00 to 16:00: there was a decrease from 12:00 to 14:00 and an increase from 14:00 to 16:00 (Wilcoxon signed-rank test p=0.006 and p=0.007, respectively). There was no change of the NT-proBNPu/ NT-proBNPp ratio from 8:00 to 18:00, but it decreased from 8:00 to 10:00 (Wilcoxon signed-rank test p=0.009; Figure 4C).

DISCUSSION We investigated whether the within-day, day-to-day (5-days) and week-to-week (6-weeks; collection at the same day of week) CVis of NT-proBNPu were more favorable compared to the corresponding NT-proBNPp in patients with stable CHF. In contrast to expectation, we found that the within-day CVi of NT-proBNPu was higher than that of NT-proBNPp. The day-to-day and week-to-week CVis of NT-proBNPu did not differ from the CVis of NT-proBNPp. This conclusion also applies for NT-proBNPu levels adjusted for dilution (by creatinine) or expressed as 89

an excretion rate (per h). The CVis did not improve by restricting the analyses to NT-proBNPp concentrations â&#x2030;Ľ310 pg/mL, i.e. the levels from which there was an almost linear relation between NT-proBNPu and NT-proBNPp.

In the present study the within-day, day-to-day and week-to-week CVis of NT-proBNPp amounted to 9, 18 and 30%. These outcomes were similar to those of our previous study, where Bruins et al. (17) reported 9, 20 and 35%, respectively. They are also in line with the week-to-week CVis of 30% for natriuretic peptides reported by others (26;27). High NTproBNPp CVis are accompanied by proportionally higher RCVs that in the present study were 25 (within-day), 51 (day-to-day) and 83% (week-to-week). The latter RCV was comparable to the week-to-week (8-weeks) RCV of 90% for NTproBNPp as reported by Wu et al. (26). The nowadays well established high RCVs for NT-proBNP in both plasma and urine preclude the use of these parameters for guided therapy of individuals, if longitudinal differences are to be detected with 95% certainty (17).

It is as yet unclear why the CVis of (immunoreactive) NT-proBNPu are higher than those of (immunoreactive) NT-proBNPp. NT-proBNPu reflects NT-proBNPp over an extended period and thereby dampens fluctuation of NT-proBNPp, caused by its pulsatile secretion together with BNP in a stoichiometric manner (12). Fluctuations of filtration fraction, eGFR, glomerular permeability of proteins and proximal tubule function are candidates. Permeability may not be limiting for low molecular weight proteins that are usually filtered without difficulty. Larsson et al. (28) observed no consistent changes of the eGFR during the day in healthy subjects. Other determinants might relate to proteolytical cleavage in tubular microvillar membranes (14;15) and receptor-mediated endocytosis by proximal tubule cells. In the latter, the megalin/cubilin receptor complex may play a central role by its ability to bind structurally very different proteins (29), but other mechanisms of protein internalization, like the caveolin-pathway and fluid-phase endocytosis have also been described (16). Megalin is involved in proximal tubular uptake of 90

glomerular-filtered albumin, other low-molecular weight proteins and their cargos (such as vitamins, minerals), including albumin, hemoglobin, transcobalamin-vitamin B12, and vitamin D-, retinol- and folic acid-binding proteins, but also many peptides including parathormone, insulin, epidermal and insulin-like growth factor and leptin (16;29;30). Following endocytosis, the captured proteins may become cycled to the plasma or tubular fluid by exocytosis, either as the intact protein (transcytosis) or after (partial) degradation in lysosomes (16).

Our data suggest that (immunoreactive) NT-proBNPu is not related to (immunoreactive) NTproBNPp (Figure 3) up to an NT-proBNPp threshold of about 310 pg/mL. From this threshold filtration seems to be the dominant determinant, suggesting that at lower levels either filtration, reabsorption, or both, might be limiting. Given the passiveness of filtration, NT-proBNPp above 310 pg/mL might be consistent with saturation of tubular reabsorption/metabolism, either by co-filtered proteins like albumin, or the reach of maximum reabsorption/metabolism capacity of NT-proBNP itself. Whatever the cause, we found that CVis for NTproBNPu at NTproBNPp levels above 310 pg/mL were not more favorable than those applicable for the entire range. More detailed studies on the renal handling of plasma proteins and notably NT-proBNPp are needed to elucidate the factors causing the presently encountered high CVis of NT-proBNPu.

Consistent with our previous study (17), we found that NTproBNPp steadily increased during the day (Figure 4A). This seems to coincide with a decrease of NTproBNPu, notably between 08:00 and 10:00 (Figure 4B), although it did not reach significance. Increased volume load (e.g. salt and fluids) related to breakfast and other meals, but also the prescribed medication (Table 1) might contribute to the observed pattern, which is at least in part responsible for the observed high within-day CVis.

The strength of this study was its prospective design with simultaneous collections of blood samples and urines. This resulted in a reliable comparison of NT-proBNP levels in both plasma 91

and urine, collected in a single day (10-h), from day-to-day (5-days) and from week-to-week (6-weeks), all taken during usual daily life activities. A limitation was some shortcomings in the sampling within a single day. The number of urine portions differed per patient. Our solution was to categorize the samples in periods of 2 h to provide the most reliable figure. The patients noted the times of voiding themselves and these were verified by us during every visit to minimize potential bias. Second, participants may not have collected the entire 24-h volumes, despite our frequent visits. Correction for creatinine and expression as an excretion rate at least partially overcame this problem. Finally, most participants used diuretics which may have caused higher water and salt excretion from 08:00 to 14:00 h (Figure 4A and 4B).

CONCLUSION We conclude that measurement of urine NT-proBNP provides no advantages over plasma NTproBNP for therapy optimization in patients with chronic heart failure.

ACKNOWLEDGEMENTS We thank all patients for their kind cooperation and Roche Diagnostics Netherlands b.v. (Mr. Ben Aalderink) for partial financial support.

References 1

Leening MJ, Siregar S, Vaartjes I, Bots ML, Versteegh MI, van Geuns RJ, et al. Heart disease in The Netherlands: a quantitative update. Neth Heart J 2014;22(1):3-10.

Heidenreich PA, Albert NM, Allen LA, Bluemke DA, Butler J, Fonarow GC, et al. Fore casting the impact of heart failure in the United States: a policy statement from the American Heart Association. Circ Heart Fail 2013;6(3):606-19.

Morrow DA, de Lemos JA. Benchmarks for the assessment of novel cardiovascular biomarkers. Circulation 2007;115(8):949-52.

McMurray JJ, Adamopoulos S, Anker SD, Auricchio A, Bohm M, Dickstein K, et al. ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure 2012: The Task Force for the Diagnosis and Treatment of Acute and Chronic Heart Failure 2012 of the European Society of Cardiology. Developed in collaboration with the Heart Failure Association (HFA) of the ESC. Eur Heart J 2012;33(14):1787-847.

Pfisterer M, Buser P, Rickli H, Gutmann M, Erne P, Rickenbacher P, et al. BNP-guided vs symptom-guided heart failure therapy: the Trial of Intensified vs Standard Medical Therapy in Elderly Patients With Congestive Heart Failure (TIME-CHF) randomized trial. JAMA 2009;301(4):383-92.

De Vecchis R, Esposito C, Di BG, Ariano C, Giasi A, Cioppa C. B-type natriuretic peptide-guided versus symptom-guided therapy in outpatients with chronic heart failure: a systematic review with meta-analysis. J Cardiovasc Med (Hagerstown ) 2014;15(2):122-34.

Savarese G, Trimarco B, Dellegrottaglie S, Prastaro M, Gambardella F, Rengo G, et al. Natriuretic peptide-guided therapy in chronic heart failure: a meta-analysis of 2,686 pa tients in 12 randomized trials. PLoS One 2013;8(3):e58287.

Fent G, Hall I, Watt V, O'Toole L, Al-Mohammad A. 43 Raised Natriuretic Peptides are not Exclusively Indicative of Heart Failure: A Service Review of a Diagnostic Heart Failure Clinic. Heart 2014;100 Suppl 3:A23-A24.

Rodeheffer RJ. Measuring plasma B-type natriuretic peptide in heart failure: good to go in 2004? J Am Coll Cardiol 2004;44(4):740-9.

Levin ER, Gardner DG, Samson WK. Natriuretic peptides. N Engl J Med 1998;339(5):321-8.

Balion CM, Santaguida P, McKelvie R, Hill SA, McQueen MJ, Worster A, et al. Physiological, pathological, pharmacological, biochemical and hematological factors affecting BNP and NT- proBNP. Clin Biochem 2008 Mar;41(4-5):231-9.

Bentzen H, Pedersen RS, Pedersen HB, Nyvad O, Pedersen EB. Abnormal rhythmic oscillations of atrial natriuretic peptide and brain natriuretic peptide in heart failure. Clin Sci (Lond) 2003;104(3):303-12.

Palmer SC, Endre ZH, Richards AM, Yandle TG. Characterization of NT-proBNP in hu man urine. Clin Chem 2009;55(6):1126-34.

Kenny AJ, Maroux S. Topology of microvillar membrance hydrolases of kidney and intestine. Physiol Rev 1982;62(1):91-128.

Maack T, Johnson V, Kau ST, Figueiredo J, Sigulem D. Renal filtration, transport, and metabolism of low-molecular-weight proteins: a review. Kidney Int 1979;16(3):251-70.

Dickson LE, Wagner MC, Sandoval RM, Molitoris BA. The proximal tubule and albuminuria: really! J Am Soc Nephrol 2014;25(3):443-53.

Bruins S, Fokkema MR, Romer JW, Dejongste MJ, van der Dijs FP, van den Ouweland JM, et al. High intraindividual variation of B-type natriuretic peptide (BNP) and amino-terminal proBNP in patients with stable chronic heart failure. Clin Chem 2004;50(11):2052-8.

Ng LL, Geeranavar S, Jennings SC, Loke I, O'Brien RJ. Diagnosis of heart failure us ing urinary natriuretic peptides. Clin Sci (Lond) 2004;106(2):129-33.

Cortes R, Portoles M, Salvador A, Bertomeu V, Garcia de BF, Martinez-Dolz L, et al. Diagnostic and prognostic value of urine NT-proBNP levels in heart failure patients. Eur J Heart Fail 2006;8(6):621-7.

Michielsen EC, Bakker JA, Kimmenade RR, Pinto YM, Dieijen-Visser MP. The diag nostic value of serum and urinary NT-proBNP for heart failure. Ann Clin Biochem 2008;45(Pt 4):389-94.

Solomon SD, Anavekar N, Skali H, McMurray JJ, Swedberg K, Yusuf S, et al. Influence of ejection fraction on cardiovascular outcomes in a broad spectrum of heart failure patients. Circulation 2005;112(24):3738-44.

Barnes SC, Collinson PO, Galasko G, Lahiri A, Senior R. Evaluation of N-terminal pro-B type natriuretic peptide analysis on the Elecsys 1010 and 2010 analysers. Ann Clin Biochem 2004;41(Pt 6):459-63.

Stevens LA, Coresh J, Feldman HI, Greene T, Lash JP, Nelson RG, et al. Evaluation of the modification of diet in renal disease study equation in a large diverse population. J Am Soc Nephrol 2007;18(10):2749-57.

Steele BW, Wang E, Palomaki GE, Klee GG, Elin RJ, Soldin SJ, et al. An evaluation of analytic goals for assays of drugs: a College of American Pathologists Therapeutic Drug Monitoring Sur vey Study. Arch Pathol Lab Med 2001;125(6):729-35.

Petersen PH, Sandberg S, Fraser CG, Goldschmidt H. Influence of index of individuality on false positives in repeated sampling from healthy individuals 162. Clin Chem Lab Med 2001;39(2):160-5.

Wu AH, Smith A, Wieczorek S, Mather JF, Duncan B, White CM, et al. Biological variation for N-terminal pro- and B-type natriuretic peptides and implications for therapeutic monitoring of patients with congestive heart failure. Am J Cardiol 2003;92(5):628-31.

Melzi dG, Tagnochetti T, Nauti A, Klersy C, Papalia A, Vadacca G, et al. Biological variation of N-terminal pro-brain natriuretic peptide in healthy individuals. Clin Chem 2003;49(9):1554-5.

Larsson A, Akerstedt T, Hansson LO, Axelsson J. Circadian variability of cystatin C, creatinine, and glomerular filtration rate (GFR) in healthy men during normal sleep and after an acute shift of sleep. Chronobiol Int 2008;25(6):1047-61.

Shankhajit De 1 SK2aAS2. The Endocytic Receptor Megalin and its Associated Proteins in Proximal Tubule Epithelial Cells. 4[membranes 4030333], 333-355. 2014. ww.mdpi.com/journal/membranesw, Membranes. 11-7-2014.

Christensen EI, Birn H, Storm T, Weyer K, Nielsen R. Endocytic receptors in the renal proximal tubule. Physiology (Bethesda ) 2012;27(4):223-36.

Week-to-week protocol week 1

week2

week 3

week 4

week 5

week 6

Day-to-day protocol day 1 day 2 day 3 day 4 day 5 08:00 10:00 12:00 14:00 16:00 18:00

Within-day protocol

Figure 1. Protocol for blood and urine sampling: an example. The total study duration was six weeks. We employed three different protocols for establishment of within-day, day-to-day and week-to-week variation, respectively. The samples of the week-to-week protocol were collected on the same day of the week.

patient recruitment from 2 cardiologists' practices

25 participants

Within-day 20 patients

Day-to-day 6 times

maximum: 120 samples

116 plasma samples

18 patients

Week-to-week 5 days

maximum: 90 samples

4 missing

22 patients

6 weeks

maximum: 132 samples

89 plasma samples

1 missing

124 plasma samples

8 missing

187 spont.voided urine portions

89 forced urine portions

1 missing

123 forced urine portions

9 missing

mean 9 (6-14) urine portions p.p.

84 full 24-h urine portions

6 missing

123 full 24-h urine portions 9 missing

Figure 2. Overview of collected and missing blood samples and urine portions of withinday, day-to-day and week-to-week sampling protocols. Spont., spontaneously; p.p., per person.

Figure 3. Relation between NT-proBNP in plasma and NT-proBNP in the enforced urine voidings collected in the day-to-day and week-to-week protocols. Data derive from 22 patients and in total 182 comparisons of NT-proBNP in plasma and urine. Samples were from the day-to-day and week-to-week protocols. For urine we used the enforced voidings. NT-proBNPp was unrelated to NT-proBNPp up to a concentration of 310 pg/mL (corresponding with NT-proBNPp of 17 pg/mL). The relation from NT-proBNPp of 310 pg/mL was: NT-proBNPp= 0.0226*(NT-proBNPp)1.1563.

100

101

Figure 4. Within-day courses for NT-proBNP in plasma, urine and the NT-proBNP urine/plasma ratio. Data are medians (interquartile range, IQR) relative to 08:00. Numbers underneath clock time refer to numbers of patient samples. #significance of 8:00 â&#x20AC;&#x201C; 18:00 timeperiod; *significance of two hours time-period; NT-proBNP, amino terminal pro-brain natriuretic peptide. All p-values are reported in Table 3.

102

Table 1. Baseline characteristics of the study group. Participants, numbers

Age, mean±SD, years

61±10

Gender (male/female), numbers

15/10

Body mass index mean±SD, kg/m2

28±6

eGFR, mean±SD, mL/min/1.73 m2

81±31

left ventricular ejection fraction, mean±SD, %

36±15

New York Heart Association classes, number (%) I

2 (8)

14 (56)

III

8 (32)

Medication, number (%) Beta-blocking agent

19 (76)

Angiotensin-converting enzyme inhibitor

10 (40)

Diuretics

22 (100)

Digoxin

5 (20)

NT-proBNPp, median (range) NT-proBNPp, pg/mL

205 (12-3,404)

NT-proBNPub, median (range) NT-proBNPu, pg/mL

19 (0-173)

NT-proBNPu/NT-proBNPp

2 (0-49)

NT-proBNPu/h, pg/h

1 (0-18)

NT-proBNPu/NT-proBNPp, median (range)

0.06 (0.00-1.86)

eGFR, estimated glomerular filtration rate; NT-proBNP, amino terminal pro-brain natriuretic peptide; p, plasma; u, urine;

Retrieved from medical files. Seven out of 25 LVEF data were

missing. One out of 25 NYHA classifications was missing. b full 24-h urine portions.

103

Table 2. Analytical variations (CVa), intra-individual variations (CVi), total variations (CVt) and reference change values (RCV), for plasma and urine NT-proBNP. Within-day

Day-to-day

Week-to-week

CVa, %

CVi, %

9 (0-21)

18 (5-90)

30 (6-215)

CVt, %

9 (3-22)

19 (6-90)

30 (7-215)

RCV, %

CVa, %

5 (0-15)

7 (0-14)

CVi, %

34 (0-71)

CVt, %

35 (0-71)

23 (0-60)

RCV, %

6 (2-15)

8 (2-15)

4 (2-87)

27 (0-112)

40 (0-180)

40 (0-200)

NT-proBNPp, pg/mL

NT-proBNPu, pg/mL

21 (0-60)

4 (0-87) d

28 (0-180)

29 (0-200)

NT-proBNPu/creatinineu, pg/nmol CVa, %

Cvi, %

29 (0-93)

CVt, %

29 (0-93)

27 (0-113)

RCV, %

113

CVa, %

5 (0-15)

7 (0-14)

4 (0-87) e

Cvi, %

47 (0-174)

NT-proBNPu/hour, pg/h

31 (0-67)

CVt, %

48 (0-174)

32 (0-67)

RCV, %

132

33 (0-180)

34 (0-200)

Data are medians (ranges). For within-day urine amino-terminal pro-brain natriuretic peptide (NT-proBNPu) we used spontaneously voided urine portions; for day-to-day and week-to-week 104

NT-proBNPu we used the full 24-h urine portions.

Within-day CVis of NT-proBNPu, NT-

proBNPu /creatinineu, and NT-proBNPu/h were higher than the CVi of NT-proBNPp (p=0.003 for pg/mL; p=0.002 for pg/nmol and p<0.001 for pg/h). Within-day CVi of NT-proBNPu/h (pg/h) was higher than that of NT-proBNPu (pg/mL; p=0.031).

Day-to-day CVi of NT-

proBNPu/h (pg/h) was higher than that of NT-proBNPu (pg/mL; p=0.044). e

Analytical variation (CVa) increased with decreasing NT-proBNP outcome; the upper-limit of

the presented CVa was determined by one patient.

105

Table 3. Within-day time-period differences for NT-proBNP in plasma, urine and the urine/plasma ratio. p-values

WSR

time-period

08-18

08-10

NT-proBNPp

0.006f

0.000f

NT-proBNPu

0.064

NT-proBNPu/creatu

10-12

12-14

14-16

16-18

0.015

0.019

0.327

0.586

0.557

0.261

0.100

0.650 1.000

0.701

0.790

0.638

0.286

0.036

0.861

0.347

0.028

0.657

NT-proBNPu/h

0.695

0.054

0.088

0.152

0.006g 0.007g 0.286

NT-proBNPu/NT-

0.091

0.215

0.009f

0.152

0.790

proBNPp

0.480

0.445

W, Wilcoxon test; F, Friedman test; WSR, Wilcoxon Signed Ranks test using p<0.01 for significance; NT-proBNP, amino-terminal pro-B-type natriuretic peptide; f significant at p<0.05; g

significant at p<0.01.

106

Supplemental Table 1. Analytical variations (CVa), intra-individual variations (CVi), total variations (CVt) and reference change values (RCV), for plasma and urine NT-proBNP in the subgroup with NTproBNPp>310 pg/mL. Within-day

Day-to-day

Week-to-week

18 (9-90)

21 (6-66)

NT-proBNPp, pg/mL CVa, %

CVi, %

12 (7-21)

CVt, %

12 (7-22)

19 (10-90)

21 (7-66)

RCV, %

2.55 (2.55-8)

2.55 (2.55-9)

2.55 (1.76-8)

36 (1-60)

36 (19-101)

NT-proBNPu, pg/mL CVa, % CVi, %

48 (36-71)

CVt, %

49 (39-71)

36 (4-60)

36 (19-101)

RCV, %

135

100

3 (3-9)

3 (3-8)

27 (4-86)

38 (22-134)

NT-proBNPu/creatinineu, pg/nmol CVa, %

Cvi, %

32 (10-93)

CVt, %

32 (11-93)

27 (5-86)

38 (22-140)

RCV, %

107

CVa, %

2.55 (2.55-8)

2.55 (2.55-9)

2.55 (1.76-8)

Cvi, %

56 (38-107)

26 (14-66)

55 (26-94)

NT-proBNPu/hour, pg/h h

CVt, %

56 (39-107)

28 (14-66)

55 (26-94)

RCV, %

157

154

Data are median (range). For within-day urine amino-terminal pro-brain natriuretic peptide (NTproBNPu), spontaneously voided urine portions were used; for day-to-day and week-to-week NT-proBNPu, full 24-h urine portions were used.

The within-day CVis of NT-proBNPu were 107

higher than the CVi of NT-proBNPp (p=0.018 for pg/mL; p=0.028 for pg/nmol and p=0.018 for pg/h). i

The week-to-week CVis of NT-proBNPu were higher than the CVi of NT-proBNPp (p=0.013 for

pg/mL; p=0.020 for pg/nmol and p=0.028 for pg/h). Analytical variation, CVa; intra-individual variation, CVi; total variation, CVt; p, plasma; u, urine; reference change value, RCV.

108

109

110

Chapter 3

BNP AND NT-PROBNP, PREDICTORS OF 1-YEAR MORTALITY IN NURSING HOME RESIDENTS. Barents M, Hillege HH, van der Horst IC, de Boer RA, Koster J, Muskiet FA, de Jongste MJ.

J Am Med Dir Assoc. 2008;9(8):580-5. Epub 2008 7.

111

ABSTRACT OBJECTIVES. To investigate on one-year mortality prediction of B type natriuretic peptide (BNP) and N terminal-proBNP (NT-proBNP) in institutionalized elderly with multiple-morbidity. METHODS. A prospective, cross-sectional study in one nursing home was designed. Participants: 93 residents (mean age 81 Âą 3 years, 66% female). Residents with serious cognitive impairments, aphasia or metastatic cancer were excluded. Measurements: clinical assessment, immobilization, medical history, electrocardiogram (ECG), echocardiogram, blood samples. One general geriatrician assessed non-cardiovascular diseases, a cardiologist panel established the diagnosis of chronic heart failure (CHF). Subjects were tracked for 1 year as far as status of death. RESULTS. 18/93 enrolled individuals died. BNP was significant higher in non-survivors compared to survivors (138 (49-753) versus 87 (27-162) P 0.029), NT-proBNP was higher but did not reach significance 1382 (193-5683) versus 335 (175-900) pg/mL (interquartile range (IQR), P 0.059). The adjusted value on one-year mortality, of six predefined chronic diseases, immobilization, age, sex, NT-proBNP and BNP was estimated by means of Cox proportional hazard regression analyses. Finally, both for NT-proBP and BNP, a mutually adjusted multivariate Cox proportional hazard analysis with the covariates presented that BNP and NT-proBNP predicted one-year mortality significantly (HR 1.16 and P 0.003, HR 1.02 and P 0.001, respectively).The mortality risk increased at rising BNP and NT-proBNP levels. CONCLUSION. BNP and NT-proBNP are predictors of one-year mortality independently of age, gender and morbidity. The mortality risk increases at elevating natriuretic peptide concentrations. We postulate that plasma levels of BNP and NT-proBNP are also of use to predict prognosis in institutionalized elderly with multiple morbidity.

112

INTRODUCTION Elderly with multiple morbidity admitted to a nursing home are faced with questions concerning their health, living conditions and social life. Information about the prognosis of their diseases is essential for them to make decisions on these questions and therefore of importance for the quality of their life. Aside existing chronic diseases and their progression, also body mass index (BMI), waist-hip ratio, biomarkers as albumin and natriuretic peptides predict 1-year mortality (1-11). Natriuretic peptides (BNP and NT-proBNP) have become available as tools for the diagnosis of heart failure and for the prognosis (4;7-9). Their prognostic role was evaluated independently of the presence of chronic heart failure and found to be of importance. However, most of these data are obtained from specialized clinics treating relatively younger patients with heart failure without comorbidity. In addition, BNP levels also increase in noncardiac conditions such as old age, being female, pulmonary diseases (pulmonary hypertension or embolism, chronic obstructive pulmonary disease) and renal dysfunction (12-15). Although nearly all nursing home residents suffer from multiple morbidity, data on the value of use of natriuretic peptides in them are limited. We therefore investigated whether BNP and NT-proBNP plasma levels remain of prognostic value in elderly patients with multiple morbidity.

METHODS During the course of this study there were 140 persons in the departments for residents with diseases of somatic origin in NH â&#x20AC;&#x153;het Zonnehuisâ&#x20AC;?. Their impairments are based on cardiovascular, pulmonary, endocrine, neurodegenerative, skeletal muscle, renal dysfunction (16) and other disorders. Most of them receive long-term care. Some residents with skeletal muscle or cerebrovascular disorders were reactivated, one-third of whom could be discharged (short stay) but who should remain care dependent in primary care. Both groups were invited to take part if they were 65 years of over. The long-term and short-stay residents were included if they understood

113

the impact of the study on themselves (competent) and if they agreed to participate by written informed consent. Persons with aphasia or a cognitive impairment measured by the mini-mental state estimation (MMSE) test ≤20 were excluded. The MMSE test contains 30 questions and a MMSE ≤20 is suspect for cognitive impairments (17). Persons with metastatic cancer who stayed in a department specialized in palliative terminal care were excluded. Persons were also excluded if the echocardiographic frames could not be sufficiently visualized or if they refused to have their blood sample taken. The study was approved of by the Medical Ethical Committee in Groningen, University Medical Center Groningen, The Netherlands (METc number 2004.107).

Diagnostics In this cross-sectional study, all data were collected anonymously and within one week (questionnaire, neurohormone sampling, ECG, echocardiography). One physician collected data of the patients’ medical history and (chronic) diseases of the cardiovascular, pulmonary, endocrine, neurodegenerative and skeletal muscle system, of their symptoms and medication. He also examined blood pressure, height and weight and performed full physical examination (18). Aside number and type of chronic disorder, we registered immobilization defined as being wheelchair dependent or bedridden. We regarded immobilization as a consequence of the chronic diseases on the individual and therefore as a parameter of progression of the present chronic diseases. The diagnosis of chronic heart failure (CHF) was made by two experienced cardiologists. A third colleague decided in cases of disagreement. Of note, the cardiologists were unaware of resident’s BNP and NT-proBNP levels. A 12-lead electrocardiogram (ECG) was made with the electrocardiograph Cardioline delta three plus (Cardioline, Milan, Italy, www.cardioline.it), with the patient in a horizontal position. The left-ventricular ejection fraction (LVEF) was assessed semi-quantitatively by the twodimensional visual estimate method (19). A LVEF ≤ 45% was considered to be a left114

ventricular systolic dysfunction (LVSD) or diminished LVEF(dLVEF). The hand held cardiograph ‘Opti Go’ (Philips, Eindhoven, The Netherlands, www.philips.com) was used. One blood sample per resident (12 mL) was taken if one was fasting and at rest. Assessments of creatinine, haemoglobin and mean corpuscular volume (MCV) were performed in the NH laboratory. Since the neurohormone levels may vary about 100% we choose not to repeat the neurohormone sampling. For determination of levels of (NT-pro)BNP, a 5 μl aprotinine solution was adjusted to the 250 μl plasma samples in EDTA and to 250 μl serum samples. At the UMCG Clinical Chemical Laboratory (CCL) both were frozen at -20˚C and stored in batches for a maximum of 10 months. The assays were run in one go for both NT-proBNP and BNP. Both NT-proBNP and BNP were measured by immunoassays (Elecsys®1010/2010/modular analytics 2004 Roche diagnostics Indianapolis IN US, and AXSYM system® BNP 2003 Axis–shield diagnostics LTD ABBOT Wiesbaden Germany). NT-proBNP and BNP had coefficients of variation of 3.3% and 7.8%(ranges of 5-35000 pg/mL and 0-3465 pg/mL respectively). Renal function has been defined as estimated glomerular filtration rate (eGFR) measured by the Cockcroft -Gault formula in mL/min. Anemia was defined as haemoglobin (Hb) ≤ 7.5 mmol/l ≈12 g/dL (7). The status of death or alive of each resident was recorded after one year following initial data collection.

Statistical analysis To address as to BNP and NT-proBNP predict one-year mortality in a cohort of old age with multiple chronic diseases we used different statistic tests. Differences in basic characteristics of survivors and non-survivors were analysed using Students ’t test, X 2 test or the Fisher exact test for categorical data and the Mann-Whitney test for non-parametric continuous data, as appropriate. The association between underlying diseases or immobilization and NT-proBNP and BNP levels divided into tertiles, was tested with X2 statististics. The adjusted value of the six predefined chronic diseases, immobilization, NT-proBNP and BNP, unadjusted and adjusted for age and sex were estimated by means of Cox proportional 115

hazard regression analyses. The six chronic diseases were CHF all cases (or CHF with dLVEF or CHF pLVEF), COPD, diabetes, renal dysfunction, neurodegenerative diseases, skeletalmuscle system diseases (Table 3). Finally, both for NT-proBP and BNP, a mutually adjusted multivariate Cox proportional hazard analysis was made with the covariates in the equation. The quantitative relationship between NT-proBNP and BNP levels and one-year mortality was expressed in Figure 1 and 2, respectively. Statistical analyses were performed using SPSS 12.0.1 software (SPSS Inc., 233 S. Wacker, 11th Floor, Chicago IL US). All statistical comparisons were two-tailed, and a p value <0.05 was considered to be statistically significant.

RESULTS A total of 93 (62%) individuals out of 140 residents were included. Individuals were excluded because of cognitive impairment [30 (21%)], missing ECG data [2 (1.4%)], missing laboratory values [1 (0.7%)], and unwillingness to participate [14 (10%)]. The mean age of the studied population was 81 ± 3 years and approximately 66% of the residents were females (Table 1). After a follow-up period of one year, 18/93 individuals died. Patients who survived did not differ in age from non-survivors (80 ± 7 years and 81 ± 9 versus, p 0.796). In comparison with survivors, non-survivors had more often CHF (all cases), diabetes and higher BNP levels. Out of the 10 non-survivors with CHF, two subjects had symptoms and eight had no symptoms of heart failure, 4/10 had pLVEF and 6/10 dLVEF. BNP and NT-proBNP were elevated in CHF compared to no-CHF, but only BNP differed significantly (median BNP in CHF 194, IQR 92-460 versus BNP in no-CHF 87, IQR 28-187 pg/mL (p<0.001) and NT-proBNP in CHF 1871, IQR 539-4262 versus NT-proBNP in no-CHF 324, IQR 163-1146 pg/mL (p< 0.001)).

116

Renal dysfunction (p 0.046 and p 0.004), CHF all cases (p 0.014 and p 0.004) and CHF with dLVEF (p 0.018 and p 0.001) were more observed in the highest BNP and NT-proBNP tertiles (p-values respectively) (Table 2). CHF all cases (P=0.014 and P=0.004) and CHF with dLVEF (P=0.018 and P=0.001) were observed more in the highest BNP and NT-proBNP tertiles (P values respectively). BNP and NT-proBNP predicted 1-year mortality significantly after adjustment for age, sex, the 6 chronic diseases, and immobilization (with CHF all cases, the analysis had a hazard ratio [HR] of 1.67 and P=0.000, HR 0.60 and P=0.000; with CHF with pLVEF HR 1.76 and P=0.000, HR 2.06 and P=0.000; with CHF with dLVEF HR 1.03 and P=0.000, HR 0.93 and 4.86, respectively) (Table 3). The mortality risk was increasing at rising NT-proBNP and BNP levels. An increase of 10 pg/mL BNP was related to a 2.2% rise of the HR of the mortality risk. Moreover, an increase of 10 pg/mL NT-proBNP was associated with a 0.27% rise of the HR of mortality risk (Figures 1 and 2, Table 3).

DISCUSSION The major finding of this study is that in NH-residents BNP and NT-proBNP are independent predictors of mortality, after adjustment for age, sex, chronic diseases and immobilization. One of the main characteristics of NH-residents is the presence of multiple morbid conditions. Residents have an average of four chronic diseases each. Moreover, CHF all cases and diabetes are more frequently represented in those who died, compared to those who did survive (Table 1). We had to consider the skewed distribution of BNP and NT-proBNP caused by extended morbidity in a relatively small population. Therefore, medians instead of means of neurohormones were used and an adjustment for all confounders of neurohormone levels was made as is presented in Table 2 and 3. After adjustment for all confounders, the neurohormones predicted one-year mortality significantly in NH-residents. Another aspect of morbidity, aside of number and type is the progression of each chronic disease. In this study however we did not register the progression of chronic diseases. To address this shortcoming we determined immo117

bilization as parameter of consequences of morbidity (22-24). Aside mortality risks as age and sex, we had to count with chronic diseases and their consequences influencing individuals. So, after adjustment for age, sex and morbidity, BNP and NT-proBNP still remained predictors of one-year mortality in nursing home residents (Table 3). Moreover, the quantitative relation between natriuretic peptides and mortality revealed high levels of natriuretic peptides required for prognosis compared to their use for diagnosis (Figure 1 and 2 ). For instance, an increase of BNP or NT-proBNP with 10 pg/mL is related to an increase of mortality risk of 2,2% or 0.27%, respectively. To date, there is a scarcity of reports on the value of BNP and NT-proBNP in elderly in nursing homes. Comparisons with other populations should therefore only be made, taken these flaws into account. In NH residents BNP and NT-proBNP are of prognostic value at higher levels, compared to a non-elderly population (25;26). In a study of ambulatory CHF patients (mean age 76 Âą11 years) with preserved LVEF and with readmission or death for cardiac reasons as endpoints, Valle et al. found predictive values of BNP at 200-499 and >500 pg/mL (HRs of 2.2 and 5.8) after six months follow-up (27). Though there were differences in design and population between Valleâ&#x20AC;&#x2122;s and the present study, both studies suggest that BNP levels for prognostic use were found to be two- to fourfold higher than advised for diagnostic purpose. Bibbins et al. found NT-proBNP to be a marker of long-term mortality, independently of other prognostic markers. In subjects with stable coronary heart disease with a mean age of 67 to 72 Âą 9 year, they also found increasing mortality risks at incremental NT-proBNP levels (28). McKie et al. studied a younger community-based cohort without heart and renal failure (mean age of 62 years), and found BNP and NT-proBNP to be biomarkers for mortality at much lower levels (biosite assay of BNP was 63 vs. 22 and NT-proBNP 206 vs. 63 pg/mL median levels, nonsurvivors vs. survivors) (29). Regarding a different range, the relationship between increasing mortality risk and elevating natriuretic peptide levels was comparable to our results (30). Altogether, we have demonstrate in this study that BNP and NT-proBNP are of prognostic value also in care dependent elderly with multiple morbidity. However, the cut-off points for prognostic 118

purpose of neurohormones seem higher and should be identified in future studies keeping track on comorbidity. A limitation of the study is the small population of 93 nursing home residents, in one centre. On the other hand, the population studied is representative of other nursing home populations in terms of distribution of age, gender, diabetes, CHF, the use of ACE-inhibitor therapy and number of chronic diseases, although not representative in terms of the presence of hypertension and renal dysfunction (31-33).

CONCLUSION In a cohort of NH-residents, BNP and NT-proBNP levels are independent predictors of one-year mortality, after adjustment for age, sex, chronic morbidity and immobilization. In the respect of the prognosis, mortality risk increases with elevated natriuretic peptide levels. We postulate that plasma levels of BNP and NT-proBNP are also of use to predict prognosis in institutionalized elderly with multiple co-morbidities.

ACKNOWLEDGEMENTS All authors had no conflict of interest. Fund: This study was subsidised by the â&#x20AC;&#x153;Vereniging het Zonnehuisâ&#x20AC;?, Soesterberg, The Netherlands. Financial disclosure: All contributors have no conflict of interest to disclose.

119

Table 1. Baseline characteristics of NH residents, survivors compared to non-survivors. Survivors

Non-survivors

Number (% of 93)

75 (81)

18 (19)

Age (year) mn±SD

80 ± 7

81 ± 9

0.796

Female

49 (53)

12 (13)

0.740

BP diastolic mmHg mn±SD

80 ± 12

74 ± 11

0.050

BP systolic mmHg mn±SD

142 ± 26

133 ± 23

0.155

Hemoglobin mmol/l mn±SD

7.7 ± 1

7.4 ± 1

0.352

Creatinine umol/l mn±SD

76 ± 27

102 ± 78

0.175

Body mass index kg/m2 mn±SD

26 ± 6

23 ± 3

0.780

LVEF % mn±SD

52 ± 8

49 ± 10

0.254

CHF all cases

13 (14)

10 (10)

0.007

CHF reduced ejection fraction

10 (11)

6 (6)

0.053

CHF preserved ejection fraction

3 (3)

4 (4)

0.190

chronic obstructive pulmonary d.

17 (18)

7 (7)

0.161

Diabetes mellitus type 2

16 (17)

9 (9)

0.038

renal dysfunction

11 (12)

7 (7)

0.066

neurological disorders

39 (41)

7 (7)

0.231

skeletal-muscle disorders

35 (37)

4 (4)

0.060

immobilization

26 (27)

9 (9)

0.232

87 (27 -162)

138 (49 – 753)

0.029

335 (175-900)

1382 (193-5683)

0.059

BNP pg/mL md (IQR) NT-proBNP pg/mL md (IQR)

P value

Data are numbers (% of the 93 subjects) unless otherwise indicated. BP, blood pressure; immobilization, subjects in wheelchair or bed; md IQR, median with interquartile range; NTproBNP, N-terminal-pro B-type natriuretic peptide; BNP, B-type natriuretic peptide.

120

Table 2. BNP and NT-proBNP concentrations divided in tertiles and linear related to chronic diseases and immobilization. Tertiles

BNP

NT- BNP

total

III

% of 93

pro

III

% COPD

0.276

0.507

diabetes

0.433

0.254

0.046

0.004

0.089

0.443

0.904

0.800

CHF all

0.014

0.004

CHFpEF

0.559

0.586

CHFrEF

0.018

0.001

Immobilization

0.870

0.434

BNP, B-type natriuretic peptide; NT-proBNP, N-terminal-proBNP; COPD, chronic obstructive pulmonary disease; RD, renal dysfunction; MSD, musculoskeletal disorder; ND, neurological disorders; CHF, chronic heart failure; CHFpEF, CHF with preserved ejection fraction; HFrEF, CHF with reduced.

121

Table 3. Association of BNP and NT-proBNP with 1-year mortality after adjustment for mortality risk factors. Neuro-

BNP

NT-proBNP

Hormone, NH

95% CI

NH, unadjusted

1.22

1.13-1.32

0.000

1.03

1.02-1.04

0.000

NH, age and sex adj.

1.23

1.13-1.33

0.000

1.03

1.02-1.04

0.000

CHF all*

1.18

1.08-1.29

0.000

1.02

1.01-1.04

0.000

CHFpEF*

1.20

1.10-1.30

0.000

1.03

1.01-1.04

0.000

CHFrEF*

1.19

1.09-1.30

0.003

1.03

1.01-1.04

0.001

diabetes mellitus type2 1.20

1.09-1.31

0.000

1.03

1.01-1.04

0.000

COPD

1.18

1.08-1.29

0.000

1.02

1.01-1.04

0.000

Renal disorders

1.18

1.08-1.29

0.000

1.02

1.01-1.07

0.000

muscular-skeletal d.

1.17

1.07-1.30

0.001

1.02

1.01-1.06

0.000

neurological disorders

1.15

1.04-1.27

0.004

1.02

1.01-1.06

0.001

Immobilization

1.16

1.05-1.28

0.003

1.02

1.01-1.07

0.001

All diseases

1.16

1.05-1.28

0.003

1.02

1.01-1.07

0.001

NH, age and sex adj.+ each of the following added 1 at a time:

CHF, chronic heart failure; * only 1 of them is in the model; CHFpEF, CHF with preserved ejection fraction; CHFrEF, CHF with reduced EF; COPD, chronic obstructive pulmonary disease ; d. disorders; HR, hazard ratios; CI, confidence interval; BNP, B-type natriuretic peptide; NTproBNP, amino-terminal pro natriuretic peptide.

122

Fitted Harzard Ratio

6.0 5.0 4.0 3.0 2.0 1.0 0.0 0

100

200

300

400

500

600

700

800

BNP pg/ml

Figure 1. Hazard ratioâ&#x20AC;&#x2122;s (HR) of BNP and 1-year mortality risk. HR of B-type natriuretic peptide, BNP (X) and mortality risk (Y) with confidence intervals (Cl 3 and Cl 4) presenting the quantitative relation of mortality at increasing BNP. The BNP range is limited to fit in this graph.

Fitted Harzard Ratio

HR 1 is at BNP level 87 pg/mL.

6.0 5.0 4.0 3.0 2.0 1.0 0.0 0

1000

2000

3000

4000

5000

NT-proBNP pg/ml

Figure 2. Hazard ratioâ&#x20AC;&#x2122;s (HR) of N-terminal-proB-type natriuretic peptides, NT-proBNP and 1-year mortality risk. HR according to Nt-proBNP range from 0 to 5000 pg/mL are depicted. Maximum Nt-proBNP level is fitted to this graph. HR 1 is at Nt-proBNP 450 pg/mL. 123

References 1

Price GM, Uauy R, Breeze E, Bulpitt CJ, Fletcher AE. Weight, shape, and mortality risk in older persons: elevated waist-hip ratio, not high body mass index, is associated with a greater risk of death. Am J Clin Nutr 2006;84(2):449-60.

Sullivan DH, Roberson PK, Bopp MM. Hypoalbuminemia 3 months after hospital discharge: significance for long-term survival. J Am Geriatr Soc 2005;53(7):1222-6.

van Dijk PT, Mehr DR, Ooms ME, Madsen R, Petroski G, Frijters DH, et al. Comorbidity and 1 year mortality risks in nursing home residents. J Am Geriatr Soc 2005;53(4):660-5.

Wallen T, Landahl S, Hedner T, Nakao K, Saito Y. Brain natriuretic peptide predicts mortality in the elderly. Heart 1997;77(3):264-7.

Yan LL, Daviglus ML, Liu K, Stamler J, Wang R, Pirzada A, et al. Midlife body mass index and hospitalization and mortality in older age. JAMA 2006;295(2):190-8.

Taylor JA, Christenson RH, Rao K, Jorge M, Gottlieb SS. B-type natriuretic peptide and N-terminal pro B-type natriuretic peptide are depressed in obesity despite higher left ventricular end diastolic pressures. Am Heart J 2006;152(6):1071-6.

Richards M, Troughton R, Lainchbury J, Doughty R, Wright S. Health gains by using natriuretic peptides in diagnosis, prognosis and treatment. Scand J Clin Lab Invest Suppl 2005;240:129-37.

Valle R, Aspromonte N, Feola M, Milli M, Canali C, Giovinazzo P, et al. B-type natriuretic peptide can predict the medium-term risk in patients with acute heart failure and preserved systolic function. J Card Fail 2005;11(7):498-503.

124

Wang R, Mouliswar M, Denman S, Kleban M. Mortality of the institutionalized old-old hospitalized with congestive heart failure. Arch Intern Med 1998;158(22):2464-8.

Witham MD, Gillespie ND, Hutcheon SD, Struthers AD, McMurdo ME. B-type natriuretic peptide is associated with mortality in older functionally impaired patients. J Am Geriatr Soc 2005;53(11):1991-5.

deFilippi CR, Seliger SL, Maynard S, Christenson RH. Impact of renal disease on natriuretic peptide testing for diagnosing decompensated heart failure and predicting mortality. Clin Chem 2007;53(8):1511-9.

Redfield MM, Rodeheffer RJ, Jacobsen SJ, Mahoney DW, Bailey KR, Burnett JC, Jr. Plasma brain natriuretic peptide concentration: impact of age and gender. J Am Coll Cardiol 2002;40(5):976-82.

Kiely DG, Kennedy NS, Pirzada O, Batchelor SA, Struthers AD, Lipworth BJ. Elevated levels of natriuretic peptides in patients with pulmonary thromboembolism. Respir Med 2005;99(10):1286-91.

Bozkanat E, Tozkoparan E, Baysan O, Deniz O, Ciftci F, Yokusoglu M. The significance of elevated brain natriuretic peptide levels in chronic obstructive pulmonary disease. J Int Med Res 2005;33(5):537-44.

Robinson BE. Epidemiology of chronic kidney disease and anemia. J Am Med Dir Assoc 2006;7(9 Suppl):S3-S6.

Tombaugh TN, McIntyre NJ. The mini-mental state examination: a comprehensive review. J Am Geriatr Soc 1992;40(9):922-35.

Shamsham F, Mitchell J. Essentials of the diagnosis of heart failure. Am Fam Physician 20001;61(5):1319-28.

125

Quinones MA, Waggoner AD, Reduto LA, Nelson JG, Young JB, Winters WL, Jr., et al. A new, simplified and accurate method for determining ejection fraction with twodimensional echocardiography. Circulation 1981;64(4):744-53.

K/DOQl clinical practice guidelines for chronic kidney disease: Evaluation, classification, and stratification Am J Kidney Dis 2002;39:S1-266.

20 Izaks GJ, Westendorp RG, Knook DL. The definition of anemia in older persons. JAMA 1999;281(18):1714-7. 21

Badley EM. An introduction to the concepts and classifications of the international classification of impairments, disabilities, and handicaps. Disabil Rehabil 1993;15(4):161-78.

22 Johnston M, Pollard B. Consequences of disease: testing the WHO International Classification of Impairments, Disabilities and Handicaps (ICIDH) model. Soc Sci Med 2001;53(10):1261-73. 23 Porock D, Oliver DP, Zweig S, Rantz M, Mehr D, Madsen R, et al. Predicting death in the nursing home: development and validation of the 6-month Minimum Data Set mortality risk index. J Gerontol A Biol Sci Med Sci 2005;60(4):491-8. 24 Galasko GI, Lahiri A, Barnes SC, Collinson P, Senior R. What is the normal range for N-terminal pro-brain natriuretic peptide? How well does this normal range screen for cardiovascular disease? Eur Heart J 2005;26(21):2269-76. 25 Valle R, Aspromonte N, Feola M, Milli M, Canali C, Giovinazzo P, et al. B-type natriuretic peptide can predict the medium-term risk in patients with acute heart failure and preserved systolic function. J Card Fail 2005;11(7):498-503.

126

26 Valle R, Aspromonte N, Barro S, Canali C, Carbonieri E, Ceci V, et al. The NTproBNP assay identifies very elderly nursing home residents suffering from pre-clinical heart failure. Eur J Heart Fail 2005;7(4):542-51. 27 Bibbins-Domingo K, Gupta R, Na B, Wu AH, Schiller NB, Whooley MA. N-terminal fragment of the prohormone brain-type natriuretic peptide (NT-proBNP), cardiovascular events, and mortality in patients with stable coronary heart disease. JAMA 2007;297(2):169-76. 28 McKie PM, Rodeheffer RJ, Cataliotti A, Martin FL, Urban LH, Mahoney DW, et al. Amino-terminal pro-B-type natriuretic peptide and B-type natriuretic peptide: biomarkers for mortality in a large community-based cohort free of heart failure. Hypertension 2006;47(5):874-80. 29 Heckman GA, Misiaszek B, Merali F, Turpie ID, Patterson CJ, Flett N, et al. Management of heart failure in Canadian long-term care facilities. Can J Cardiol 2004;20(10):963-9. 30 Mair J, Friedl W, Thomas S, Puschendorf B. Natriuretic peptides in assessment of leftventricular dysfunction. Scand J Clin Lab Invest Suppl 1999;230:132-42. 31

Mann JL, Evans TS. A review of the management of heart failure in long-term care residents. Consult Pharm 2006;21(3):222-8.

32 Nygaard HA, Naik M, Ruths S, Kruger K. Clinically important renal impairment in various groups of old persons. Scand J Prim Health Care 2004;22(3):152-6.

127

128

Chapter 4

CHRONIC HEART FAILURE, RELATED TO HELP AT ACTIVITIES OF DAILY LIVING IN NURSING HOME RESIDENTS. Barents M, Hillege HH, Muskiet FA, de Jongste MJ.

Activities, Adaptation & Aging 2011;35(2):98-110.

129

ABSTRACT It is unknown whether CHF is associated with ADLâ&#x20AC;&#x201C;functioning in nursing home residents therefore we studied that association. METHOD. 103 residents were screened on age, sex, comorbidity, and CHF. ADL-dependence and ADL-help were scored with the use of Minimum Data Set (MDS)-items of the Resident Assistant Instrument. Logistic multivariate regressions of ADL-dependence and ADL-help were performed with age, sex, CHF and comorbidity as independent variables. RESULTS. age 78 Âą 11 years. 24/103 had CHF (23%). CHF was associated with ADL-help (OR 4.68 and p=0.015) after multivariable adjustment. CONCLUSION. CHF is associated with increase of ADL-help given to nursing home residents. CHF is of importance for both residents (loss of autonomy and life satisfaction) and organization (workforce planning).

130

INTRODUCTION Chronic heart failure (CHF) is a major public health care problem in the developed world because it prevails mostly at high age and takes up quite a part of public health care expenses. Chronic heart failure is an underestimated diagnosis.

In the United States (US), nearly 5 million patients are suffering of CHF (1). CHF is primarily a disease of the elderly. In the US and Europe, CHF affects about 1% of persons in their 50s and rises progressively with age to afflict 10% of persons in their 80s (2). In Canadian and Dutch nursing homes, the CHF prevalence is higher than in independent living elderly and valued at 15-23% (3;4). In North America, CHF is responsible for large health care costs (1;4;5). In developed countries, health care expenses on CHF consumes 1-2% of the entire health care budget. In Sweden and The Netherlands, the costs of public health care are increasing gradually (5). Health care expenses consist primarily of hospitalization costs followed by costs of care.

As noted, CHF is often undetected or incorrect diagnosed for several reasons. The key symptoms fatigue and dyspnea are unspecific. Older adults attribute fatigue to their age or chronic diseases in case of multimorbidity. Fatigue and effort intolerance are gradually different symptoms. Oxygen poverty related to the bodyâ&#x20AC;&#x2122;s need causes fatigue and effort intolerance. Dyspnea is often taken for chronic obstructive pulmonary disease (COPD) (7). So the lack of symptom specificity and the contribution of symptoms to aging lead often to under diagnosis of heart fail131

ure. Another problem is that diagnostic tests (B-type natriuretic peptide (BNP) exclude but does not determine CHF in older adults with multimorbidity (3;8). At last, echocardiography is not easy accessible.

From literature, we know that CHF patients need care at activities of daily life (ADL) because decreased effort tolerance (9;10). In independent elderly ADL-ability is vital in providing for life satisfaction (11). Thus, care givers need to know whether persons suffer from CHF not only from treatment but also from ADL-ability and quality of life point of view.

ADL-ability can be measured by the Resident Assisted Instrument (RAI). RAI is a core set of assessments developed to provide a picture of, under other things, each nursing home residentâ&#x20AC;&#x2122;s ADL-ability (12). Care attendants observe and score specific ADL-settings. They also fill in standardized diagnoses. RAI is widely used in American and European nursing homes as aid for staff and care attendants (13). Remarkably, RAI does not contain CHF as standardized diagnosis so nursing home staffs which use RAI do not take CHF into account for care planning and might overestimate ADL-ability of residents with CHF.

In The Netherlands, three questions gave rise to a study of CHF in nursing home residents (residents): first, we expected under- and incorrect CHF diagnoses also in residents according to literature in primary care (7). Second and third, the CHF prevalence in residents and the CHF consequences on ADL-ability were unknown. Results of the first two questions studied 132

were described elsewhere (3;8). Third, we knew that Dutch nursing home staffs didn’t take CHF into account for care level calculations. Consequently residents with CHF might be undervalued concerning ADL-functioning and the care planning might be underestimated concerning residents suffering CHF. We didn’t know whether CHF is associated with ADL–functioning in residents. Subsequently, we sought to study the association of CHF to ADL-functioning in this nursing home. We screened residents of one nursing home on CHF, scored ADL-ability with RAI and compared residents with and without CHF on ADL-ability.

METHODS

Design This mono-centre study had a non-randomized, prospective screening study design. The Medical Ethical Commission of the University Medical Centre of Groningen (UMCG) approved the study under number 2004.107.

Subjects The study population consisted of 150 residents who stayed in nursing home the “Zonnehuisgroep Noord” in Zuidhorn, Netherlands between September 2004 and May 2005. The inclusion criteria were competency (the resident understood the impact of study participation on himself), age 65 years and over and signing the informed consent. The Mini Mental State Estimation (MMSE 30) was administered with a required MMSE score of at least 21 when the resident's 133

competency was in doubt (14). The exclusion criteria were age under 65 years, serious aphasia, MMSE < 21, metastatic carcinomas and absence of echocardiogram or blood sample.

Data collection Data of all subjects consisted of a) medical history from medical records, b) standardized physical examination, c) electrocardiogram and echocardiogram, d) blood tests and e) ADLscores.

We collected all data of each person within one week. One physician-investigator summarized medical history except CHF and performed physical examinations. We categorized the chronic diseases (comorbidity) on the basis of six organ systems to limit the number of variables. The six disease categories were chronic obstructive pulmonary diseases (COPD), diabetes, neurological diseases (ND), muscular-skeletal diseases (MSd), chronic renal dysfunction (CRD) and blindness. ND comprised of cerebrovascular accidents (CVA), multiple sclerosis, amyotrophic lateral sclerosis, Parkinsonâ&#x20AC;&#x2122;s disease, ataxia, spasms, neuropathies and paresis /paralysis. MSd included artrosis, hip artroplasty and all fractures in the past six months, (rheumatism) arthritis, deformations of body and extremities and myopathies. CRD was defined as creatinineclearance â&#x2030;¤ 50 mL/min (Cockgroft-Gault formula). A panel of cardiologists diagnosed CHF on the basis of medical history, symptoms, electrocardiogram, echocardiogram and blood tests. This diagnostic method was the golden standard for CHF.

134

Standardized physical examination b) comprised examination of the cardiac pulmonary and skeletal muscle system, height and weight. c) Physician assistants made 12-lead-electrocardiograms (ECG) and echocardiographists echocardiograms on location. The echocardiogram [Opti Go (www.philips.com)] estimated the left ventricular ejection-fraction (LVEF, percentages) two-dimensionally and visually. We defined a LVEFâ&#x2030;¤ 45% as left ventricular systolic dysfunction (LVSD) (15). Blood tests d) consisted of a one-time-assessment of haemoglobin (Hb), mean corpuscular volume (MCV) and creatinine. Care attendants scored e) ADL divided in ADL-dependence (ADL-D) and ADL-help (ADL-H). They observed dependence (D) and help (H) during four specified ADL-settings as toilet using, eating, performing personal hygiene and walking on the ward. They scored dependence level (D) of the resident and amount of help (H) given to the resident during these four ADL-settings in the last seven days. ADL-D had a six-point scale so sum scores varied from 0 to 24. ADL-H had a three-point scale so sum scores varied from 0 to 12. Higher scores indicated more dependence and help (Table 1). Afterwards, we condensed ADLD and ADLâ&#x20AC;&#x201C;H scores each to ranges from 0 to 6, as advised by experts (www.prismant.nl). In this way, a data set arose which is known as Minimal Data Set of the Resident Assistant Instrument (RAI) (16-18). We selected the RAI for ADL measurements because its relative reliability and validity in nursing homes. And, because the care attendants of this nursing home had experience with RAI (8-10;13;19;20). Blood samples d) were taken from residents at rest and

135

in fasting condition, stored at -20 C° and analyzed in one go by the Laboratory Noord (www.laboratoriumnoord.nl) to optimize validity.

Statistics Statistical analyses were performed using SPSS 15.0.01. The question was whether CHF was associated to more ADL-dependence and ADL-help compared to residents without CHF. Differences in base characteristics of residents with and without CHF were calculated using Student’s t, chi-square and the Mann-Whitney test for nonparametric continuous data, as appropriate (Table 2). Correlations between variables were calculated by Spearman’s correlation coefficient ro (Table 3). Univariate and multivariate logistic regression analyses were made of ADL-D and ADL-H. Age, sex, CHF and 6 chronic diseases were put in the analyses. We used the Entermethod since this method matched theory testing best (Table 4 and 5). All statistical comparisons were two-tailed, and a p value <0.05 was considered to be statistically significant.

RESULTS In total, 150 residents were recruited of whom 47 were excluded. Of the 47 excluded, 14 refused participation, 30 had cognitive impairment or aphasia, 2 had inaccessible echocardiogram frames and 1 refused to let take blood (1). Thus, 103 residents were recruited of whom 39 men and 74 women. The distribution of age and sex did not differ between the included and excluded residents (age 78 ± 11 versus 79 ± 12, p=0.703 and sex male/female 37/63 versus 136

30/70, p=0.340 respectively). Each resident had 2 Âą 1 chronic diseases in average with a distribution of 0-6 not including CHF. Two percentages residents had no other chronic disease (only CHF), 18% had one chronic disease, 40% had two, 28% had three and 12% had four to six chronic diseases. The CHF group had more often chronic renal dysfunction as compared to the non-CHF group (p=0.032). CHF involved more ADL-D and ADL-H compared to residents without CHF (p=0.005 and p=0.009)(Table 2). For significant correlations see Table 3.

In univariate analyses, CHF was significantly associated with ADL-H after adjustment for sex and age [odds ratio (OR)=4.44 confidence interval (CI)1.45 â&#x20AC;&#x201C; 13.60; p=0.009 and OR=4.46 (CI 1.44-13.88); p=0.015, respectively]. In multivariate analyses, CHF was significant associated with ADL-H after adjustment for sex, age and 6 chronic diseases [OR=4.68 (CI 1.35- 16.17); p=0.015](Table 5).

DISCUSSION The question of this study is whether chronic heart failure is associated with ADL-dependence and ADL-help in nursing home residents. We found that CHF is clearly associated with ADLhelp after adjustment for age, sex and comorbidity [OR=4.68 (CI 1.35- 16.17), p=0.015]. We also found that CHF is not associated with ADL-dependence. The somatic chronic diseases examined are neither associated with ADL-dependence nor ADL-help in this population.

137

What does an increase of ADL-help imply for resident and organization? At first, reception of ADL-help implies loss of autonomy, also for residents. Second, ADL-ability is a determinant for life satisfaction probably also in residents. Therefore, for the resident point of view, each increase of ADL-help makes the difference between an autonomic or assisted performance of ADL-activities, such as use of toilet. Thus, a resident will experience each increase of ADL-help as loss of autonomy and life satisfaction. Each increase of ADL-help is therefore clinically most relevant (11;12). For the organization point of view, an increase of ADL-help means that the resident need one or two additional care attendants. This affects the workforce planning. Thus, an increase of ADL-help is most important for resident and organization.

Why are the six chronic diseases examined not associated with ADL-dependence and ADL-help while some of them occur frequently? The disease category might be underestimated when residents have more than one disease within the category. In such cases the disease merely count as one chronic disease. Moreover, the categorization of chronic diseases used does not represent the clientâ&#x20AC;&#x2122;s disease burden. Generalization for the sake of categorization may have led to the absence of significant associations of each of the six chronic diseases with ADLdependence and ADL-help.

How does the result that ADL-help is determined by CHF relate to literature? In his primary care study on ADL and motor skills, Norberg finds that senior citizens with CHF are not de138

pendent on their personal hygiene but are in 75% of the cases dependent and get help with putting on and taking off their compression stockings and shoes (21). Chen finds an increase in ADL-help at 7 ADL-items from MDS, in institutionalized CHF residents (mean age 91 years), in their final year of life (22). Scilley and Owsley prove the relation between impaired eyesight and ADL-dependence in nursing home residents (23). Altogether, for the greater part, the results of this study are supported by other studies. Is it possible that the etiological categorization of chronic disorders influences its results? Is one of the chronic disorders related to CHF or are they interrelated by a common pathophysiology such as vascular disorders? Kriegsman shows indeed that, in the case of several chronic disorders, a common etiology leads to less loss of function and a different etiology leads to more loss of function (9).This mono-centre study is limited in its small population (# 103). However, when compared with two other Dutch nursing home populations, our study population was representative in terms of distribution of age, sex, hypertension and diabetes but not in terms of renal dysfunction (24;25). A bias on the part of one physician-investigator who collected the data is possible. But this seems unlikely since she did not assess electrocardiograms. Moreover, she was blinded for the final diagnoses made by the cardiologistâ&#x20AC;&#x2122; panel. Another possible limitation of the study might be that depression and psychotropic drug use are left out of the analysis. The potential influence of these variables on ADL-help has to be taken in consideration in further studies. The relevance of this study is that CHF is associated to increase of ADL-help in this nursing home population. Therefore, the diagnosis of CHF is of importance both for resident (loss of autonomy and life satisfaction) and organization (workforce planning). Further studies on CHF in relation to ADL are needed in lar139

ger NH populations. Also, subgroup analysis of CHF related to ADL are needful because it is unknown whether only CHF with decreased LVEF is related to more ADL-help and not CHF with preserved LVEF.

CONCLUSION CHF is associated with increase of ADL-help given to nursing home residents. CHF is of importance for both residents (loss of autonomy and life satisfaction) and organization (workforce planning).

ACKNOWLEDGEMENTS The “Vereniging het Zonnehuis”, Bilthoven, The Netherlands granted the study. This study was subsidised by the “Vereniging het Zonnehuis”, Bilthoven, The Netherlands. The authors have no conflict of interest.

140

Table 1. Scores of ADL-dependency (ADL-D) and ADL-help (ADL-H) of MDS-items (Minimum Data Set) from the Resident Assistant Instrument (RAI). scores definition Activities of daily living-

independent

care attendance, relatively independent

some help, mild restrictions

extensive help (with lifting) by 1 person

extensive help (with lifting) by 2 persons

dependent

total dependent

no help, no care needed

care attendance, verbal support

physical help by 1 person

physical help by 2 or more persons

dependence

Activities of daily living-help

Activities of daily living is scored from 0-6; Activities of daily living-help is scored from 0-3.

141

Table 2. Baseline characteristics of nursing home residents (n=103). variables

all

CHF+

% 103

CHF-

% 103

103

(100)

(23)

(77)

age years, mean±SD

(11)

(9)

(11)

0.351

sex male (%)

(37)

(46)

(23)

0.358

COPD (%)

(24)

(25)

(24)

0.925

diabetes m. type 2(%)

(25)

(38)

(22)

0.114

neurological d. (%)

(50)

(47)

0.181

muscular-skeletal(%)

(43)

(29)

(47)

0.096

Renal dysfunction (%)

(32)

(12)

(20)

0.032

(3)

(0)

(4)

0.083

tension systolic mn±SD

141

(22)

143

(27)

140

(21)

0.534

tension diastolic mn±SD

(13)

0.590

BMI kg/cm2, mean±SD

(5)

0.897

7.7

(1.0)

7.8

(1.3)

7.7

(0.9)

0.834

ADL-D median, iqr

(1-4)

(2-5)

(0-4)

0.005

ADL-H median, iqr

(3-7)

(3-8)

(2-7)

0.009

number

Comorbidity

blind (%)

Physical examination

Blood tests Hb mmol/L, mean±SD

ADL scores

Characteristics in numbers (percentage) unless otherwise indicated. COPD, chronic obstructive pulmonary disorders; ND, neurological disorders; MD, musculoskeletal disorders; BMI, body mass index; ADL, activities of daily living; ADL-D, ADL-dependence; ADL-H, ADL-help.

142

Table 3. Significant inter-variable correlations of age, sex, chronic disorders and CHF. variable with

variable

correlation r

chronic heart failure

renal dysfunction

0.21

0.031

chronic heart failure

ADL-dependence

0.28

0.004

chronic heart failure

ADL-help

0.29

0.030

renal dysfunction

age

0.42

0.000

neurological disorders

ADL- dependence

0.21

0.032

neurological disorders

musculoskeletal disorders

0.36

0.000

neurological disorders

sex

0.25

0.010

ADL- dependence

ADL-help

0.80

0.000

sex

0.23

0.020

ADL, activities of daily living.

143

Table 4. Associations of ADL-dependence (ADL-D) and age (univariate), sex (univariate), CHF (age and sex adjusted) and each of the six chronic diseases (age and sex adjusted).

ADLD

uni

age

va-

sex

riate

adj. OR

multi variate

95% CI

age

1.00

0.96- 1.06

0.808

sex

0.72

0.26- 2.01

0.527

CHF

2.55

0.86- 7.54

0.092

2.45

0.82-7.35

COP

0.87

0.26- 2.93

0.823

0.81

T2DM

1.63

0.54- 4.88

0.387

CRD

1.95

0.60-6.35

0.93

MSd blind

95% CI

1.00

0.95-1.06

0.861

1.07

0.32-0.59

0.913

0.110

2.43

0.73-8.14

0.149

0.24-2.80

0.744

0.85

0.22-3.30

0.813

1.64

0.54-4.94

0.383

1.49

0.45-4.93

0.516

0.267

1.84

0.51-6.70

0.352

1.04

0.23-4.65

0.958

0.32-2.75

0.897

1.12

0.32-3.87

0.862

1.93

0.47-7.93

0.361

0.32

0.10- 1.06

0.062

0.33

0.10-1.09

0.069

0.48

0.13-1.79

0.274

10.50

0.90-122.81

0.061

9.72

0.82-115.67

0.072

10.73

0.73-

0.084

158.83

CHF, chronic heart failure; COPD, chronic obstructive pulmonary disorders; T2DM, diabetes mellitus; CRD, chronic renal dysfunction; ND, neurological disorders; MD, musculoskeletal disorders; OR, odds ratio; CI, confidence interval; P, P-value.

144

Table 5. Associations of ADL-help (ADL-H) and age (univariate), sex (univariate), CHF (age and sex adjusted) and each of the six chronic diseases (age and sex adjusted). ADL-H

uni

age

multi

vari-

sex

vari-

ate

adj.

ate

95% CI

age

0.97

0.94-1.09

0.722

sex

1.07

0.23-4.51

0.982

95% CI

OR CI

1.01

0.95-1.07

0.736

0.88

0.24-3.21

0.848

4.68

1.35-16.17

0.015

CHF

4.44

1.45-13.60

0.009

4.46

1.44-13.88

0.010

COPD

0.68

0.18-2.62

0.577

0.65

0.17-2.53

0.534

0.60

0.14-2.62

0.497

T2DM

2.01

0.65-6.21

0.225

2.04

0.66-6.31

0.218

1.80

0.52-6.32

0.353

1.60

0.45-5.65

0.465

1.52

0.39-6.00

0.548

0.97

0.21-4.56

0.968

1.32

0.45-3.85

0.616

0.39

0.07-2.19

0.285

0.50

0.06-4.16

0.524

0.56

0.18-1.75

0.317

0.56

0.78-1.75

0.316

0.75

0.20-2.83

0.673

blind

2.83

0.24-3.24

0.407

2.82

0.23-34.16

0.415

6.73

0.41-110.65

0.182

CHF, chronic heart failure; COPD, chronic obstructive pulmonary disorders; T2DM, diabetes mellitus ; CRD, chronic renal dysfunction; ND, neurological disorders; MD, musculoskeletal disorders; OR, odds ratio; CI, confidence interval; P, P-value.

145

Figure 1. ADL-dependence of residents with and without chronic heart failure. M, median; IQR, interquartile range; CHF, chronic heart failure; ADL, activities of daily living. .

Figure 2. ADL-help of residents with and without chronic heart failure. M, median; IQR, interquartile range; CHF, chronic heart failure; ADL, activities of daily living.

146

References 1 Dickstein K, Cohen-Solal A, Filippatos G, McMurray JJ, Ponikowski P, Poole-Wilson PA, et al. ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure 2008: the Task Force for the Diagnosis and Treatment of Acute and Chronic Heart Failure 2008 of the European Society of Cardiology. Developed in collaboration with the Heart Failure Association of the ESC (HFA) and endorsed by the European Society of Intensive Care Medicine (ESICM). Eur Heart J 2008;29(19):2388-442. 2 Kelder JC, Cowie MR, McDonagh TA, Hardman SM, Grobbee DE, Cost B, et al. Quantifying the added value of BNP in suspected heart failure in general practice: an individual patient data meta-analysis. Heart 2011;97(12):959-63. 3 Wright SP, Doughty RN, Pearl A, Gamble GD, Whalley GA, Walsh HJ, et al. Plasma amino-terminal pro-brain natriuretic peptide and accuracy of heart-failure diagnosis in primary care: a randomized, controlled trial. J Am Coll Cardiol 2003;42(10):1793-800. 4 Cowie MR, Mosterd A, Wood DA, Deckers JW, Poole-Wilson PA, Sutton GC, et al. The epidemiology of heart failure. Eur Heart J 1997 Feb;18(2):208-25. 5 Hobbs RE. Guidelines for the diagnosis and management of heart failure. Am J Ther 2004;11(6):467-72. 6 Zaphiriou A, Robb S, Murray-Thomas T, Mendez G, Fox K, McDonagh T, et al. The diagnostic accuracy of plasma BNP and NTproBNP in patients referred from primary care with suspected heart failure: results of the UK natriuretic peptide study. Eur J Heart Fail 2005;7(4):537-41. 7 Izaks GJ, Westendorp RG, Knook DL. The definition of anemia in older persons. JAMA 1999;281(18):1714-7.

147

8 Achterberg WP, Holtkamp CC, Kerkstra A, Pot AM, Ooms ME, Ribbe MW. Improvements in the quality of co-ordination of nursing care following implementation of the Resident Assessment Instrument in Dutch nursing homes. J Adv Nurs 2001;35(2):268-75. 9 Holtkamp CC, Kerkstra A, Ooms ME, van CC, Ribbe MW. Effects of the implementation of the Resident Assessment Instrument on gaps between perceived needs and nursing care supply for nursing home residents in The Netherlands. Int J Nurs Stud 2001;38(6):619-28. 10 Morris JN, Murphy K, Nonemaker S. Interrai resident assesment instrument versie 2.0 gebruikershandleiding. Utrecht: SIG zorginformatie Mailiebaan 50 3508 SC Utrecht The Netherlands; 96 A.D. 11 Mutasingwa DR, Ge H, Upshur RE. How applicable are clinical practice guidelines to elderly patients with comorbidities? Can Fam Physician 2011;57(7):e253-e262. 12 Grundy SM, Cleeman JI, Daniels SR, Donato KA, Eckel RH, Franklin BA, et al. Diagnosis and management of the metabolic syndrome: an American Heart Association/National Heart, Lung, and Blood Institute Scientific Statement. Circulation 2005;112(17):2735-52. 13 Barents M, van der Horst IC, Voors AA, Hillege JL, Muskiet FA, de Jongste MJ. Prevalence and misdiagnosis of chronic heart failure in nursing home residents: the role of B-type natriuretic peptides. Neth Heart J 2008;16(4):123-8. 14 Voors AA, Walma EP, Twickler TB, Rutten FH, Hoes AW. [Multidisciplinary guideline 'Heart failure 2010']. Ned Tijdschr Geneeskd 2011;155:A2957.

148

15 Barents M, van der Horst IC, Voors AA, Hillege JL, Muskiet FA, de Jongste MJ. Prevalence and misdiagnosis of chronic heart failure in nursing home residents: the role of B-type natriuretic peptides. Neth Heart J 200;16(4):123-8. 16 Mason JM, Hancock HC, Close H, Murphy JJ, Fuat A, de BM, et al. Utility of biomarkers in the differential diagnosis of heart failure in older people: findings from the heart failure in care homes (HFinCH) diagnostic accuracy study. PLoS One 2013;8(1):e53560. 17 Oudejans I, Mosterd A, Bloemen JA, Valk MJ, van VE, Wielders JP, et al. Clinical evaluation of geriatric outpatients with suspected heart failure: value of symptoms, signs, and additional tests. Eur J Heart Fail 2011;13(5):518-27. 18 Fernhall B. Long-term aerobic exercise maintains peak VO(2), improves quality of life, and reduces hospitalisations and mortality in patients with heart failure. J Physiother 2013 ;59(1):56. 19 Dontje ML, van der Wal MH, Stolk RP, Brugemann J, Jaarsma T, Wijtvliet PE, et al. Daily Physical Activity in Stable Heart Failure Patients. J Cardiovasc Nurs 2013 Feb 14.

149

150

Chapter 5

PHYSICAL EXERCISE PERFORMED BY RESIDENTIAL ELDERLY OF HIGH AGE, IS SAFE HOWEVER DOES NOT AFFECT HEART FAILURE SYMPTOMS, NT-PROBNP AND CARDIOVASCULARMETABOLIC INDICES; A PILOT, SINGLE-BLINDED RCT. Barents M, Weening B, de Greef MH, Muskiet FAJ, Hillege HL, DeJongste MJ.

Submitted.

151

ABSTRACT Does exercise benefit care-home residents afflicted with cardiovascular-metabolic diseases who are leading a sedentary life? Exercise effects were investigated on chronic heart failure (CHF) and the metabolic syndrome. METHODS. For 16 weeks a twice weekly exercise was performed by the exercise group and a non-physical program by the control group. Cardiovascular-metabolic markers were measured [(N-terminal-pro B-type-natriuretic peptides(NT-proBNP), waist, HbA1c, diabetes mellitus type 2 (T2DM), plasma lipids and blood pressure]. RESULTS. Analysis of aged 85 years showed neither significant changes in exercise groups (24) compared to controls (22) nor lasting injury from falls. Only exercise group attendance is inversely related to non-fasting triglycerides (p=0.024). CONCLUSION. In care-home residents, guided physical exercise is safe in the short term but may be less effective in reducing chronic heart failure symptoms, NT-proBNP concentrations and the metabolic syndrome. We have explained the neutral results by the high age (85 year on average) accompanied by low muscle endurance, and by a training schedule that might be not enough personalized.

152

INTRODUCTION The Dutch guideline for cardiovascular risk management (CVRM) states that the 10-year risk of cardiovascular diseases (CVD) of 70+ elderly should exceed 20% and that the elderly should be eligible for medication when lifestyle changes fail (1). Care home elderly (residents) are even more at cardiovascular (CV) risk when compared to 70+ elderly because the first group (of elderly) are sedentary most often and they are suffering from multimorbidity. Examples of multimorbidity are diabetes mellitus type 2 (T2DM), CV diseases, notably heart failure (CHF), cerebrovascular ischemic attack (CVA) (2). In the daily practice of geriatric medicine, these risks are addressed pharmacologically rather than by a change of life style such as physical activity (1;3;4). Changes of lifestyle require motivation to adjust behaviour while drug intake barely takes extra effort. This is one of the reasons why residential elderly remain sedentary. Their daily physical activity is three to eleven minutes. However, this inactivity of residents is inconsistent with the recommendations of the CVRM guideline (5-9). This inconsistency is also based on perspectives of a limited lifespan and fear of falling. These anxieties prevent residents from becoming physically active. Chronic heart failure (CHF) is defined as a syndrome resulting from myocardial muscle dysfunction with usual exposure of characteristic symptoms (10). Natriuretic peptides, notably B-type natriuretic peptide (BNP) and amino-terminal pro-B-type NP (NT-proBNP) are well established CHF biomarkers. They are widely used for CHF diagnosis, follow-up and prognosis (11), while they also provide opportunities for biomarker-guided treatment optimization (12). The metabolic syndrome was defined as the presence of at least three out of five of the following symptoms (13): central obesity, compromised glucose homeostasis or diagnosed T2DM, atherogenic dyslipidemia and hypertension (14;15). Both CV and metabolic derailments have atherosclerosis as a central pathophysiological phenomenon that is eligible for lifestyle changes as physical exercise. In the present study we sought to investigate two aspects of physical exercise with/in residents, its safety and changes of CHF symptoms and a marker, and markers of the metabolic syn153

drome. We aimed to study whether participation in a predefined exercise program is safe and whether symptoms and a marker of CHF and markers of the metabolic syndrome, will change after such a program. x

METHODS This pilot study was designed as a multicenter single-blinded randomized controlled trial. The end points were CHF symptoms, a marker, and the metabolic syndrome. This study was part of the Benefit study (). The Benefit study took place in 14 care homes. The care home residents in the present study participated in the same intervention program as those in the Benefit study. This program consisted of either physical exercise or a program without exercise. The Benefit and present study differed in size (Benefit: 14 care homes, present study: 4 care homes) and end points. The Benefit study end points consisted of physical fitness, functional performance, activities of daily life and quality of life (16).

Subject selection. From March 2010 to December 2011, we invited all elderly of four residential homes in Groningen to participate in the study. The inclusion criteria were: no dementia, the ability to walk at least ten meters, understanding the impact of the study. Potential participants had to sign an informed consent. Exclusion criteria were age <70 years, disturbed cognition (Mini-Mental State Examination test <20; range 0-30), serious aphasia, heart failure New York Heart Association (NYHA)-class IV, terminal phase, and metastatic cancer. To the best of our knowledge there was no literature available about effects of exercise programs on patients with CHF and CV-metabolic indices among the elderly with multimorbidities. Since a meaningful estimation of the number of participants needed for a power analysis was debatable we decided to perform a pilot study.

154

Endpoint definitions (Table 1) Chronic heart failure was defined as a syndrome resulting from myocardial muscle dysfunction with the usual exposure of characteristic symptoms such as shortness of breath (dyspnoea) and fatigue, both in rest, during activity and during exertion (in three conditions)(10). Symptoms of CHF were estimated as changed if dyspnea (scores 0-3) and or fatigue (scores 0-3) decrease with at least one score point. The metabolic syndrome was defined as the presence of at least three out of five of the following symptoms (17): central obesity (waist circumference ≥102 males or ≥88 females, compromised glucose homeostasis (increased glycated hemoglobin >41 mmol/mol or the diagnosis T2DM), atherogenic dyslipidemia (high triglycerides >150 mg/dL, low HDL-cholesterol <40 mg/dL for males or <50 for females) and increased systolic blood pressure (>130 mm Hg) (14;15). Smoking is not included in this enumeration. Participation in the exercise program was defined as safe if falls or unfavourable events during exercise sessions occurred in <4% [16 out of 32 followed sessions x 24 (number of) exercising residents=384 occasions/100%=around 4% accidents]. The chosen changes of endpoint variables are represented in Table 1. Motives for non-participation. As expected (17), a relatively low participation was taken into account. Participants were randomly assigned to the two groups at each location. Intending to analyse motives for not participating, we inquired those who refused participation about their reasons. In order to compare characteristics of non-participants to those of participants, biometric data of the first group (age, weight, height, heart rate and blood pressure) were collected by the investigator. Medical history and medication were obtained from general practitioners. The reasons for refusal were grouped into intrinsic motives (‘no realization of benefits of exercise’ and ‘does not want to adhere to the program’) and extrinsic motives. The latter were defined as ‘too busy with other activities’, illness, hospital admission, moving and impending death.

155

Subjects During two visits before and after the intervention, all participants were subjected to a structured interview, specific physical examination and biometric measurements. Blood samples were taken within two hours after breakfast and lunch. Participants were physically examined for cardiopulmonary symptoms, blood pressure (Heine gamma-G5ÂŽ; www.Heine.com), symptoms or complaints of the abdomen, joints, back, visual ability and hearing. Multimorbidity was scored using the Charlson Comorbidity Index (18;19). Blood samples were taken two hours after breakfast or lunch. For logistic reasons, it was not feasible to collect fasting blood samples. The Medical Ethical Committee of the UMCG approved of the study (number 2010.178).

Interventions The exercise program included moderate to high-intensity training sessions (16). In each session, the participants performed progressive resistance training of upper and lower extremities and trunk, static and dynamic balance training, and functional training. Elastic bands (20) were used for resistance training. The following muscle groups were trained: biceps, triceps, shoulder, back, abdominal and hip muscles, quadriceps femoris, gastrocnemic, peroneus and tibialis as well as foot muscles. The volume of resistance training was gradually increased from one set of eight repetitions in the first five weeks to two sets of eight repetitions in weeks six to ten and finally to three sets of eight repetitions in weeks eleven to sixteen. Balance training was adjusted to individual abilities. The level of difficulty was established by observing the participants performing a balance task, e.g. walking on a fitness mat. The participantsâ&#x20AC;&#x2122; balance was progressively challenged by increasing the level of difficulty every two weeks. Functional training comprised exercises like chair stands, walking, turning, walking on a course with obstacles. Frequency and duration of both group sessions were one hour twice weekly/two times a week during sixteen weeks, 32 in total. The controls followed a non-physical social program with information, games and videos in the same weeks. The participants were not allowed to participate in other physical activity programs except current physiotherapy. 156

Clinical chemical analyses Hemoglobin, estimated glomerular filtration rate (eGFR), HDL-cholesterol, triglycerides and HbA1c were measured by standard procedures. NT-proBNP was measured by immune-assay (www.roche-diagnostics.com).

Statistics All statistical analyses were performed with SPSS version 18.0 for Windows. Results were expressed as means ± SD or medians [interquartile range (IQR)] for parametric and nonparametric data, respectively. Available data of non-participants were compared to data of participants. Characteristics of exercise and control groups were compared at baseline. Differences within groups between end-point minus baseline of each variable were tested by a paired t-test or Wilcoxon-signed-ranks-test. Differences across both groups were tested by unpaired tests (ttest or Mann-Whitney U test). A sub-analysis was made, aimed at those participants who completed 16 or more (≥50%) of all sessions, who were defined as ‘as treated’. In case of insignificant differences in the outcomes of the ‘intention-to-treat’ and ‘as treated’ statistical analyses, we only report outcomes of the ‘intention-to-treat’ analysis. All statistical comparisons were twotailed. A p<0.05 was considered statistically significant.

RESULTS

Subjects’ characteristics The number of subjects in four care homes consisted of 434 residents. Of these, 163 (38%) were eligible to participate. There were 52/434 (12%) subjects willing to join the study who were included (participants), and 111/434 (26%) who refused to take part (non-participants). Of the non-participants, 47/111 (42%) were willing to provide us some biometric data only once. They also explained their motives not to take part in the study:23/47 repulsed participation because of intrinsic and 24/47 because of extrinsic motives. Fifty-two participants were 157

assessed at baseline, of which 28/52 were randomized to the exercise group and 24/52 to the controls. During the study, 6/434 (1%) participants dropped out because of not exerciserelated reasons: one died after a CVA, two moved to other institutions and three discontinued their participation for intrinsic reasons. Their data were excluded from the final analysis. Out of 52 participants, 46 completed the program, 24/28 of the exercise group and 22/24 of the controls. No differences were observed between characteristics of the interviewed non-participants (n=47) and participants (n=46), except that the non-participants had a more disturbed cognition compared to the participants (MMSE 30 22 vs. 25; p=0.011). The participants were 85 years on average and 65% of them were female (Table 1). T2DM occurred in 17% with a mean HbA1c of 43 Âą 8 mmol/mol. Half of them used antihypertensives. Nine (20%) had CHF. More females than males and more residents with CHF than without CHF were enrolled in the exercise group compared to controls, respectively (79% vs. 50%; p=0.041 and 8 vs. 1; p=0.028). In the exercise group, median NT-proBNP (446 vs. 251 pg/mL), systolic blood pressure (143 versus 127 mm Hg) and antihypertensive use (83% versus 55%) respectively, were significantly higher compared to controls. The exercise group differed from the control group neither in fatigue nor in dyspnea. One fall without permanent injury was reported during an exercise session (Table 3). The usual physical activity consisted of one walk a day to an indoor restaurant. Residents seldom visited the garden outside. For distances longer than 50150 m a wheelchair was used.

Comparison of exercise and control groups before and after intervention Within the exercise group, the female waist changed significantly [-8(-15- -2) cm]. The triglycerides changed insignificantly [-6(-19-8)] within the exercise group and significantly [-8(-170)] in the controls. Between-group comparison revealed relevant changes in female waist and

158

systolic BLOOD PRESSURE, though the findings were not statistically significant. The other variables did show neither significant within-group nor between-group differences (Table 4). Sub-analysis. In the sub-analysis of participants who attended ≥16 of the sessions (‘as treated’) we found a decrease of female waist within the controls [-11(-22-0)cm, p=0.049)], and within the exercise group a decrease of triglycerides [-11(-19--1) mg/dL, p=0.032] and an increase of NT-proBNP 81(8-155), p=0.33], both insignificant when compared between exercise and control group (supplemental Table 4a). There were no other significant or worthwhile betweengroup differences detected in this ‘as treated’ sub-analysis. Attendance and endpoints. The number of attended sessions was not associated with age, gender and comorbidity. However, it was inversely related with changes of triglycerides (p=0.024) in the exercise group only.

DISCUSSION In this study, we questioned whether participation in a predefined exercise program was safe and effective to change symptoms, a marker of CHF, and markers of the metabolic syndrome. Besides, we regarded participation rates in the interventions. In this pilot study, we found the applied exercise program to be safe for residents (1 fall, no lasting injuries). We observed that neither metabolic markers nor a marker and symptoms of CHF changed after completing a one hour moderate-to-high intensity exercise program, followed twice weekly over a period of 16weeks, when compared to controls. It was found that the number of attended sessions in the exercise group was inversely related to the change of non-fasting triglycerides from baseline to study end. This relation was however not confirmed by a triglycerides decrease after exercise. The change in triglycerides suggest that some metabolic change occurred, but this was at most genuinely minor. Of the eligible residents, we found one-third (32%) motivated, one-third not willing for intrinsic and extrinsic reasons and one-third unmotivated to participate with unknown motives .

159

To explain the results, we assessed population characteristics, exercise frequency, safety, participation and attendance. At last, the results of the Benefit study were added to our outcome. Firstly, plasma levels of Hb, HbA1c, triglycerides, HDL-cholesterol, and systolic blood pressure were all within reference limits before the intervention. The renal function was only slightly affected. Therefore, it was estimated that participants were properly set up on medications for CVD, CHF and metabolic syndrome (e.g. the majority used antihypertensives (Table 3). It was noted that doses of antidiabetics, antihypertensives and digitalis remained unchanged during the study period. Secondly, we also found that participants had fewer complaints. Only 27% of the exercise and 41% of the control group complained of fatigue and or dyspnea. Thirdly, the high age of participants might explain partly why symptom and marker changes did not occur. In general, people aged 50 yrs and older, lose muscle strength (mass) and muscle endurance (mitochondria), especially when they lead a sedentary life (21). Endurance training, which is the best exercise to increase or maintain mitochondrial concentration when aging, has generally resulted in relatively small functional benefits for care home residents (22;23). This implies that the high-age related changes in muscular and CV physiology, augmented by the CHF disease state in some of our participants, limited the exercise capacity and its corresponding benefits. Surprisingly, the studied population was stable thanks to medication. Fourthly, the high intraindividual variation of NT-proBNP (24) explained the lack of results, which precluded the detection of minor changes, if any, in a small study group. Although the exercise program used was specifically developed for our study population (16), it was questioned whether a (three-times weekly) frequency of three times a week might have provided more changes. However, we preferred a twice weekly frequency for fear of more dropouts and based on promising results of others (25). Measures taken to ensure safety consisted not only of the individual exertion level but also of guidance of the training and the own environment where the training sessions were held. Till now, effective management of (the) fear of falling is lacking (26).

160

The participation rate of our study consisting of 32% inclusion and 68% refusal, was compared with that of another group of 65 years old residents with 49% inclusion and 6% refusal (17). This comparison suggested that the higher age of our participants was due to the lower participation rate. Another reason for the found lower participation rate might be interference of other social activities with our interventions. In addition, the reported non-participants were related to those with a lower cognition. This suggested that a certain level of cognition may be required to participate in a study design as is applied in our study. One positive aspect was noted. Most participants remained in the program. Their motivation appeared from fewer drop-outs (12%) compared to the number of drop-outs (>20%) found in another group of 72 year old exercising residents (17). In this study, the attendance (up to 60%) was lower compared to that reported by others (31). Accordingly, we analyzed the relations between attendance and all outcome variables in a subgroup of participants with the best attendance ( â&#x2030;Ľ50% attendance). Compared to the entire group, this subgroup did not exhibit more changes in metabolic markers and symptoms and a marker of CHF (supplemental Table 4a). It is therefore unlikely that our results are based on lower attendance. Other reasons must account for this. Using the same exercise program in comparable cohorts, it is rational to summarize the conclusions of the present study and the Benefit study (20). Both studies conclude that the exercise program is ineffective in reducing disability, care-dependency and cardiovascular-metabolic endpoints in care home residents. However, on the basis of these two studies, the question whether we should advise physical exercise to residents in order to promote health and functionality or not, cannot be answered. As a consequence, we still do not know whether the CVRM recommendations on physical exercise are applicable to care home residents. Subsequently, studies in care home residents are warranted. The limitations of the present study were the skewed distribution of CHF and gender following randomization and the non-fasting triglycerides measurement, which could, unfortunately, not be avoided because of logistical reasons.

161

CONCLUSION In care-home residents, guided physical exercise is safe in the short term but may be less effective in reducing chronic heart failure symptoms, NT-proBNP concentrations and the metabolic syndrome. We have explained the neutral results by the high age (85 year on average) accompanied by low muscle endurance and by a training schedule that might be not enough personalized.

ACKNOWLEDGEMENTS The “Hanze University of Applied Sciences” and “Stichting de Hoven”, both in Groningen, The Netherlands, granted the study. Conflict of interest statement: none.

162

References 1 Wiersma T, Smulders YM, Stehouwer CD, Konings KT, Lanphen J. [Summary of the multidisciplinary guideline on cardiovascular risk management (revision 2011)]. Ned Tijdschr Geneeskd 2012;156(36):A5104. 2 Marengoni A, Rizzuto D, Wang HX, Winblad B, Fratiglioni L. Patterns of chronic multimorbidity in the elderly population. J Am Geriatr Soc 2009;57(2):225-30. 3 Mancia G, De BG, Dominiczak A, Cifkova R, Fagard R, Germano G, et al. 2007 ESHESC Practice Guidelines for the Management of Arterial Hypertension: ESH-ESC Task Force on the Management of Arterial Hypertension. J Hypertens 2007;25(9):1751-62. 4 Kroon AA. [Dutch guideline for cardiovascular risk management: a consensus model]. Ned Tijdschr Geneeskd 2012;156(36):A5196. 5 Nelson ME, Rejeski WJ, Blair SN, Duncan PW, Judge JO, King AC, et al. Physical activity and public health in older adults: recommendation from the American College of Sports Medicine and the American Heart Association. Med Sci Sports Exerc 2007;39(8):1435-45. 6 Jans M.P. De ontwikkeling van een beweegnorm voor ouderen in verpleeg- en verzorgingshuizen. Vreede de P., Tak E.C.P.M., Meeteren van N.L.U., editors. TNO kwaliteit van leven. 046[KvL/B&G], 9-15. 1-9-2008. www.tno.nl, TNO.

7 Christmas C, Andersen RA. Exercise and older patients: guidelines for the clinician. J Am Geriatr Soc 2000;48(3):318-24. 8 Taylor AH, Cable NT, Faulkner G, Hillsdon M, Narici M, Van Der Bij AK. Physical activity and older adults: a review of health benefits and the effectiveness of interventions. J Sports Sci 2004;22(8):703-25. 163

9 Mutasingwa DR, Ge H, Upshur RE. How applicable are clinical practice guidelines to elderly patients with comorbidities? Can Fam Physician 2011;57(7):e253-e262. 10 Mosterd WL, Rosier PF. [Guideline 'Chronic heart failure']. Ned Tijdschr Geneeskd 2004;148(13):609-14. 11 McMurray JJ, Adamopoulos S, Anker SD, Auricchio A, Bohm M, Dickstein K, et al. ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure 2012: The Task Force for the Diagnosis and Treatment of Acute and Chronic Heart Failure 2012 of the European Society of Cardiology. Developed in collaboration with the Heart Failure Association (HFA) of the ESC. Eur Heart J 2012;33(14):1787-847. 12 Morello AM, Januzzi JL. Amino-terminal pro-brain natriuretic peptide: a biomarker for diagnosis, prognosis and management of heart failure. Expert Rev Mol Diagn 2006;6(5):649-62. 13 Grundy SM, Cleeman JI, Daniels SR, Donato KA, Eckel RH, Franklin BA, et al. Diagnosis and management of the metabolic syndrome: an American Heart Association/National Heart, Lung, and Blood Institute Scientific Statement. Circulation 2005;112(17):2735-52. 14 Executive Summary of The Third Report of The National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, And Treatment of High Blood Cholesterol In Adults (Adult Treatment Panel III). JAMA 2001;285(19):2486-97. 15 Brunzell JD, Davidson M, Furberg CD, Goldberg RB, Howard BV, Stein JH, et al. Lipoprotein management in patients with cardiometabolic risk: consensus statement from the American Diabetes Association and the American College of Cardiology Foundation. Diabetes Care 2008;31(4):811-22.

164

16 Weening-Dijksterhuis E, de Greef MH, Scherder EJ, Slaets JP, van der Schans CP. Frail institutionalized older persons: A comprehensive review on physical exercise, physical fitness, activities of daily living, and quality-of-life. Am J Phys Med Rehabil 2011;90(2):156-68. 17 Nyman SR, Victor CR. Older people's recruitment, sustained participation, and adherence to falls prevention interventions in institutional settings: a supplement to the Cochrane systematic review. Age Ageing 2011;40(4):430-6. 18 Charlson M, Szatrowski TP, Peterson J, Gold J. Validation of a combined comorbidity index. J Clin Epidemiol 1994;47(11):1245-51. 19 Charlson ME, Charlson RE, Peterson JC, Marinopoulos SS, Briggs WM, Hollenberg JP. The Charlson comorbidity index is adapted to predict costs of chronic disease in primary care patients. J Clin Epidemiol 2008;61(12):1234-40. 20 Baum EE, Jarjoura D, Polen AE, Faur D, Rutecki G. Effectiveness of a group exercise program in a long-term care facility: a randomized pilot trial. J Am Med Dir Assoc 2003;4(2):74-80. 21

Rogers MA, Evans WJ. Changes in skeletal muscle with aging: effects of exercise training. Exerc Sport Sci Rev 1993;21:65-102.

Fiatarone MA, O'Neill EF, Ryan ND, Clements KM, Solares GR et al. Exercise training and nutritional supplementation for physical frailty in very elderly people. N Engl J Med 1994;330(25):1769-1775.

23 Tseng BS, Marsh DR, Hamilton MT, Booth FW. Strength and aerobic training attenuate muscle wasting and improve resistance to the development of disability with aging. J Gerontol A Biol Sci Med Sci 1995;50 Spec No:113-9.

165

24 Bruins S, Fokkema MR, Romer JW, Dejongste MJ, van der Dijs FP, van den Ouweland JM, et al. High intraindividual variation of B-type natriuretic peptide (BNP) and aminoterminal

proBNP

patients

with

stable

chronic

heart

failure.

Clin

Chem

2004;50(11):2052-8. 25 Chin APM, van Poppel MN, Twisk JW, van MW. Effects of resistance and all-round, functional training on quality of life, vitality and depression of older adults living in longterm care facilities: a 'randomized' controlled trial [ISRCTN87177281]. BMC Geriatr 2004;4:5. 26 Parry SW, Deary V, Finch T, Bamford C, Sabin N, McMeekin P, et al. The STRIDE (Strategies to Increase confidence, InDependence and Energy) study: cognitive behavioural therapy-based intervention to reduce fear of falling in older fallers living in the community - study protocol for a randomised controlled trial. Trials 2014;15:210. 27 Rydwik E, Gustafsson T, Frandin K, Akner G. Effects of physical training on aerobic capacity in frail elderly people (75+ years). Influence of lung capacity, cardiovascular disease and medical drug treatment: a randomized controlled pilot trial. Aging Clin Exp Res 2010;22(1):85-94.

166

Table 1. Endpoints At risk for

Primary endpoints

all

waist circumference (cm)a triglycerides (mg/dL)

Female

>102

>88

<40

<50

>150

high-Density Lipoprotein-cholesterol (mg/dL)a systolic blood pressure (mm Hg)

Male

diastolic blood pressure (mm Hg)a non-fasting glucose (mg/dL)a glycated Hemoglobin (mmol/mol)

>130 >95 <140 >41

Secondary endpoints fatigue in 3 activity levels (1 â&#x20AC;&#x201C; 3)

dyspnea in 3 activity levels (1 â&#x20AC;&#x201C; 3)

N-terminal pro B-type natriuretic peptide( pg/mL)b a

>450

Based on definition of the metabolic syndrome. The severity of fatigue and dyspnea are based

on the definition of the NYHA classification: class 1: no symptoms; class 2: symptoms occur at ordinary activity; class 3: symptoms occur ordinary activity. b NT-proBNP cut-off value for the diagnosis of heart failure (11).

167

Table 2. Sample description. Participants are compared to non-participants at baseline. Variable

Participants

Non-participants

N = 46

N=47

male/female

16/30

9/38

0.151

age (years)

85 ± 6

84 ± 8

0.658

body mass index kg/m2

27 ± 5

25 ± 65

0.258

mini-mental state estimation 30

25 ± 4

22 ± 4

0.011

heart rate (beats/min)

73 ± 11

71 ± 25

0.522

smoking median (interquartile range)

3 (1-4)

heart failure n (%)

9 (%)

Cerebro-vascular accident n (%)

16 (%)

comorbidity

11 (%)

residents with antidiabetics n (%)

6 (13)

residents with antihypertensives n(%)

32 (70)

Data are presented as means ± SD. Comorbidity, Charlson Comorbidity Index (sum of current chronic diseases out of 17 chronic diseases).

168

Table 3. Characteristics of exercise and control groups of care home residents. Variable

Exercise group

Controls

Ex vs. C

(Ex) n=24

P-value

age years mn SD

85±5

84±6

0.298

male/female number/number

5/19

11/11

0.041

waistm cm mn SD

107±15

113±28

0.692

waistf cm mn SD

105±21

97±12

0.690

26±3

25±4

0.421

0.273

mini-mental state examination mn SD smoking number fatigue n (range 0-3)

1±1

0.757

dyspnea n (range 0-3)

1±1

0.859

heart rate (beats/min)

71±12

76±9

0.154

diastolic blood pressure (mm Hg) mn SD

85±12

81±12

0.263

systolic blood pressure (mm Hg) mn SD

143±9

127±18

0.005

comorbidity number md IQR

3(0-4)

3(1-4)

0.705

cerebrovascular accidents number

0.353

diabetes mellitus type 2 number

0.183

% residents using antidiabetics

0.099

chronic heart failure number

0.028

% residents using antihypertensives

0.037

digoxin %

0.671

446 (260-967)

251 (93-664)

0.024

58±22

70±27

0.098

13±2

0.871

glucose mg/dL mn SD

118±38

101±20

0.061

HbA1c mmol/mol mn SD

43±10

43±6

0.844

triglycerides mg/dL mn SD

122±56

131±66

0.633

HDL-cholesterol male mg/dL mn SD

40±6

51±16

0.139

HDL-cholesterol female mg/dL mn SD

51±11

54±22

0.690

attended sessions%

60±37

58±34

0.840

0.852

NT-proBNP pg/mL median IQR 2

eGFR mL/min/1.73 m mn SD hemoglobin mg/dL mn SD

fall without injury number

IQR, interquartile range; NT-proBNP, amino-terminal proBNP; eGFR, estimated glomerular filtration rate. 169

Table 4. Changes (Dif) in the exercise and control groups from baseline to the study end, together with between-group differences of these changes. Variable Difference

Dif

Exercise

Control

Ex vs. C

(Ex)

(C)

Between

Within group

Within

group

adjusted

Mn (95%CI)

(95%CI) Dif waist male cm

(95%CI)

-1 (-4 – 1)

0.135

-3 (-1 – 2)

0.155

-2 (-6 – 2)

0.381

Dif waist female cm

-8 (-15- -2)

0.018

-2 (-6 – 3)

0.405

-4 (-10 – 1)

0.076

Dif triglycerides mg/dL

-6 (-19 – 8)

0.394

-8 (-17- 0)

0.033

-6 (-11- -1)

0.271

Dif HDL-C male mg/dL

0 (-3 – 3)

0.951

-6 (-18– 6)

0.274

-3 (-11- 5)

0.773

Dif HDL-C female mg/dL

1 (-3 – 5)

0.577

2 (0 – 5)

0.086

2 (-1- 4)

0.948

Dif DBP mm Hg

-4 (-7 – 0)

0.029

-3 (-10– 4)

0.378

-4 (-8 – 1)

0.935

Dif SBP mm Hg

-1 (-14 – 11)

0.830

2 (-7 – 11)

0.681

-4 (-7- 6)

0.059

Dif glucose mg/dL

-13 (-27–2)

0.078

7 (-5 – 20)

0.228

-7(-26– 13)

0.127

Dif HbA1c mmol/mol

3 (0 – 6)

0.086

1 (-1 – 3)

0.329

0 (0 – 0)

0.579

Dif fatigue n

0 (-1 – 0)

0.137

0 (-1 – 0)

0.261

0 (-1- 0)

0.930

Dif dyspnea n

0 (-1 – 0)

0.328

0 (0 – 0)

0.162

0 (-1- 0)

0.696

Dif NT-proBNP

334(-117– 785)

0.139

67(7–126)

0.030

22(-9– 140)

0.948

pg/mL*(md IQR)

mn, means; SD, standard deviation; * md, median with IQR, interquartile range; MMSE30, mini-mental state examination 30; comorbidity, Charlson Comorbidity Index (sum of current chronic diseases out of 17 chronic diseases); CVA , cerebrovascular accidents; DBP, diastolic blood pressure; SBP, systolic blood pressure; HDL-C, HDL–cholesterol; n, number; NTproBNP, amino-terminal proBNP, this variable is tested non-parametrically; eGFR, estimated glomerular filtration rate; attendance, percentage attended sessions.

170

Supplemental Table 4a. Subgroup of attendance 竕･16 sessions: changes (Dif) in exercise and control groups from baseline to the study end, together with between-group differences of these changes. Variable Differences

Exercise

Control

Within-group

Ex vs. C Between-

Withingroup

group adjusted (Dif)

Mn(95%CI)

Dif waist m cm

-4(-10-3)

0.270

-1(-4-1)

Dif waist f cm

-2(-7-3)

0.408

-11(-22-0)

0.135

2(-6-10)

0.601

0.049

-9(-22-5)

0.190

Dif triglycerides

-11(-19--1)

0.032

-5(-12-2)

0.134

5(-6-17)

0.563

Dif HDL-C m

-7(-20-7)

0.277

0(-3-3)

1.00

8(-10-25)

0.351

Dif HDL-C f

25(-29-78)

0.321

2(-1-4)

0.2100

0(-6-6)

0.986

2(-10-13)

0.762

0(-14(-14)

0.993

-2(-19-16)

0.540

Dif HbA1c %

0.1(-0.1-0.2)

0.332

0.1(0.1-1.0)

0.281

0.1(-0.2-0.3)

0.736

Dif fatigue n

0(0-0)

0.543

0(0-0)

0.136

0(-1-0)

0.543

Dif dyspnea n

0(0-0)

0.332

0(0-0)

0.229

0(-1-0)

0.670

Dif NT-proBNP

81(8-155)

0.033

367(-245-981)

0.222

280(-354-914)

0.852

Dif SBP

Mn(95%CI)

Data are means ﾂｱSD; *median with interquartile range (md, iqr); m, male; f, female; DBP, diastolic blood pressure; SBP, systolic blood pressure; HDL-C, HDL窶田holesterol; n, number; NTproBNP, amino-terminal proBNP this variable is tested non-parametrically.

171

172

Chapter 6

SUMMARY, DISCUSSION, RECOMMENDATIONS AND FUTURE PERSPECTIVES

173

Chapter 6.1 SUMMARY A drawback of the improved survival from cardiovascular diseases (CVD) (1) is that cardiovascular morbidities have grown. As we have emphasized in the introduction of this thesis, the number of people living with heart failure (HF) has increased impressively in conjunction with a tremendously

improved

life

expectancy

the

Dutch

population

(www.cbs.nl;

www.nationaalkompas.nl), during the last century. Since HF is mainly a problem of advanced age, subjects diagnosed with chronic HF (CHF) often concern elderly. However, there is a wide lack of knowledge with regard to data on HF when age progresses and elderly become frail and more care dependent. The most care dependent elderly are committed to a care or nursing home. In this thesis, those elderly persons are indicated as residential elderly. The shortage of (research) data is mainly caused by excluding residential elderly from studies because of high age and multimorbidities, such as CVD. Since there is a deficiency of data, no specific guidelines are available to diagnose and to treat HF in residential elderly. From this perspective, there is debate on the applicability of existing HF guidelines developed for CHF patients in general, to residential elderly with CHF. However, guidelines for residential elderly with CHF are sorely missing because diagnosing CHF is notoriously difficult: elderly tend to subscribe complaints to aging instead of CHF or other diseases. Subsequently, CHF treatments are often incorrectly applied, or not applied at all. The consequences of these caveats may be loss of quality of life and increase of costs. Therefore, we sought to study CHF in the residential elderly.

In Chapter 2.1 we have explored the epidemiological scale of CHF, by determining the prevalence of CHF in residential elderly. Based on the recent literature, when general guidelines on CHF (2;3;4) were applied to residential elderly, diagnosing CHF was less accurate due to poor predictive values of signs and symptoms, medical history, and electrocardiogram (ECG).

174

From medical files of residential elderly in a single nursing home in Groningen, we inventoried signs, symptoms, medical history and registered the number of subjects with CHF. Then we performed a physical examination and an ECG of all residents. Next we estimated the accuracy of both, former CHF diagnostic (natriuretic peptides (NPs) not used) and current CHF diagnostic procedures (including NPs), since, in contrast to the broad recognition of NPs as a screening test for CHF, their use as biomarkers in residential elderly has remained limited. The restricted use in elderly is caused by lack of validation of cut-off values for NPs in this group of subjects. Finally, we studied ECGs of all residents. A panel of cardiologists decided on the presence or absence of CHF by assessment of NPs and the result of echocardiography, successively. Echocardiographic investigations were feasible in 98%. As a result of the present study, CHF was established in 24/103 (23%) residential elderly. Fifteen (15/24) residents were not previously detected with CHF. Before the study, 22 were identified with CHF. Out of the 22 residents with CHF before the study, in only 9 residents CHF was confirmed and of 13 residents with CHF the diagnosis was rejected. The diagnostic accuracy of NT-proBNP at 450 pg/mL was 0.71 sensitivity, 0.67 specificity, 0.42 positive predictive value (PPV) and 0.91 negative predictive value (NPV). The diagnostic accuracy of BNP at 100 pg/mL was: 0.71 sensitivity, 0.70 specificity, 0.41 PPV and 0.88 NPV. In brief, the most striking results of the present study were that more than half the CHF diagnoses were missed (i.e. 15/24) or incorrectly made (i.e. 13/22). Hence we concluded that the overall accuracy of identifying CHF is limited. However, given the high NPVs, the use of natriuretic peptides as additional diagnostic instrument seems promising, even in nursing home residents but requires further evaluation. The CHF prevalence in residential elderly was estimated at 23%. In Chapter 2.2 we studied whether the outcomes of chapter 2.1 were consistent with the CHF prevalence in Aruban nursing homes. Aruban nursing homes were chosen since we were interested in a comparable population with Dutch nationality living in a different environment and 175

having another lifestyle. The same design as in Chapter 2.1 was utilized to study the prevalence of CHF in Aruban residential elderly. The other aim was to validate whether employment of natriuretic peptides could improve identification of CHF in residential elderly in another part of the Dutch Kingdom. The main difference with the chapter 2.1 study was that we were unable to obtain echocardiograms of all residential elderly, due to logistic reasons. Therefore, we could not meet the diagnostic standards for CHF (2-4). As a result, 51 out of 235 elderly Aruban residents were included with a mean age of 78±8 years. According to the medical files 7/51 residents were acknowledged with CHF. However, two out of the 7 residents did not have CHF. Furthermore, CHF was established in 16 out of 51 (31%) residents. Out of the 16 residents with CHF, 11 were not previously diagnosed with CHF and of 5/16 residents identified with CHF, the diagnosis was confirmed. In brief, the general guidelines for HF are applied to Aruban residential elderly, infrequently. When compared to residential elderly in Groningen, the prevalence of CHF may be at least as high in Aruban residential elderly. In conclusion, recognition of CHF appears to be severely underestimated in Aruban residential elderly. The detection of CHF will be improved by implementing an appropriate guideline, including the determination of BNP.

In Chapter 2.3 we addressed the problem of the large intra-individual variations [individual coefficient of variation (CVi)] of NT-proBNP in plasma, since large CVis limit the applicability of NT-proBNP for among others, guided therapy optimization in individual patients with CHF (5). In search of more reliable CVis we compared concentrations of urine NT-proBNP (NT-proBNPu) to concentrations of plasma NT-proBNP (NT-proBNPp), in HF patients living in Curaçao. Urine and blood samples were taken on a single day (“within-day”): six blood samples every 2 hours and spontaneously voided urines during 24-hours. On five consecutive days (“day-today”): five blood samples, five enforced urine samples, and five full 24-h urines were taken. On the same day of six consecutive weeks (“week-to-week”): one blood sample, one enforced

176

urine sample and one full 24-h urine was collected. Out of these blood and urine samples the total CVs (CVts), CVis and reference change values (RCVs) were calculated. In this study 25 CHF patients were included with a mean age of 61 (range 36-80) years, 60% was male and the average left ventricular ejection fraction was 36Âą15%. Median CV is for NTproBNPp were 9% (within-day), 18% (day-to-day) and 30% (week-to-week). For NT-proBNPu the CVis were 34%, 21% and 28%, respectively. The reference change value (RCV) of NTproBNPu was higher than, or equal to, the RCV for NT-proBNPp, even after correction for creatinine and the time period (in hours) in which the urine accumulated in the bladder. Our data of day-to-day and week-to-week samples suggest that (immunoreactive) concentrations of NT-proBNP in urine do not unequivocally correlate with (immunoreactive) NT-proBNP concentrations in plasma, up to an NT-proBNPp threshold of about 310 pg/mL (Figure 1). In addition, we found that beyond this threshold the higher NT-proBNP concentrations, as expressed in Figure 1, are not explained by an increase in blood pressure, during the day (results not shown). We hypothesise that filtration of natriuretic peptides may be considered as the dominant factor beyond the threshold and that at lower NT-proBNP levels filtration and reabsorption are more dynamically interacting. In conclusion, measurements of NT-proBNP in urine have no advantages in stable patients with chronic heart failure.

177

Figure 1. Relation between NT-proBNP in plasma and NT-proBNP in the enforced urine voidings collected in the day-to-day and week-to-week protocols. Data derive from 22 patients and in total 182 comparisons of NT-proBNP in plasma and urine. Day-to-day and week-to-week samples were used. For urine we have used the enforced voidings.

In Chapter 3 we have investigated the prognostic value of natriuretic peptides on one-year mortality in the same cohort of residential elderly as studied in chapter 2.1. The clinical relevance is that if natriuretic peptides are related to prognosis of residential elderly with CHF, these biomarkers may be exploited for advance care planning. Advance care planning is related to quality of life and becomes important when life expectancy is restricted (www.Verenso.nl). To investigate one-year mortality, survival and death of residential elderly was tracked, during one year. Since ten residential elderly could not be followed up, because they moved to unknown destinations, we studied the remaining 93 residential elderly.

178

Eighteen out of 93 residents (mean age 81 Âą 3 years, 66% female) died within one year (nonsurvivors). A mutually adjusted Cox proportional hazard regression analysis was performed. Adjustments were made for six predefined chronic diseases, immobilization, age, sex, NT-proBNP and BNP. The results found were that both natriuretic peptides (NT-proBNP and BNP) significantly predicted one-year mortality (HR 1.02 and p=0.001, and HR 1.16 and p=0.003, respectively). In addition, the one-year mortality risk increases exponentially in concert with higher BNP and NTproBNP levels. For instance, an NT-proBNP increase up to 2,000 pg/mL was associated with a 60% higher risk of dying, within one year. In conclusion, both BNP and NT-proBNP are independent predictors of one-year mortality for residential elderly with CHF. In addition, the mortality risk increases at natriuretic peptide concentrations elevated far beyond the diagnostic cut-off values of the guidelines (2;3;4).

In Chapter 4 we studied the relation between CHF and the support that is needed for residential elderly with CHF during their activities of daily life (ADL). To this end, the same cohort of residential elderly was examined as in chapter 2.1. The rationale for this study question was to draw attention of care teams to the need of support among those residents with CHF. For this study the same group of residential elderly was assessed as in chapter 2.1. To collect data on ADL the Minimal Data Set (MDS)-items of the Resident Assistant Instrument was used. The MDS consists of systematic observations on dependency and delivered support and aims to quantify dependence and aid (6). The MDS instrument consists of a software program for entering the observations of the individual residential elderly, during predefined activities of daily living (ADL). An example of such an activity is walking. The care taker observes and scores (quantifies) whether the individual is able to walk predefined distances independently and needs help during walking. Of 103 residents with and without CHF, the ADL-dependence and ADLhelp scores were compared and adjusted for differences between the two groups by means of

179

logistic multivariate regression. CHF was associated with ADL-help (OR 4.68 and p=0.015) however, CHF was not related to ADL-dependence. In conclusion, CHF is associated with a necessary increase of ADL-help for nursing home residents. In this regard, CHF is of importance for both residents (loss of autonomy and life satisfaction) and organization (workforce planning).

In Chapter 5 we assessed a physical exercise program in residential elderly with CHF. The rationale for this study was the discrepancy between the sedentary lifestyle adhered to by residential elderly and the recommendations for physical activity from HF guidelines (2;3;4). Also family members are stimulating their parents in care homes to become physically active. However, for a variety of reasons, residential elderly prefer to remain sedentary. Furthermore, because of fear of falling, also caretakers prefer residential elderly to maintain their sedentary lifestyle. On the other hand, a sedentary life style is related to an increased risk on (progression of) atherosclerosis and CV-metabolic diseases, such as CHF and metabolic syndrome (7). To investigate whether regular exercise is safe and has beneficial effects on CV-metabolic indices for residential elderly, we evaluated effects of exercise on CHF symptoms, NT-proBNP and metabolic syndrome. A pilot single-blinded, randomized controlled feasibility trial in four care homes was performed. In collaboration with other investigators (8), we studied whether a standardized exercise program affects CHF symptoms, NT-proBNP and the metabolic syndrome. The other investigators (8) studied whether the same standardized exercise program affects functional endpoints in comparable groups. The exercise intervention included a guided moderate-to-high intensity training of one hour twice weekly, during 16 consecutive weeks. Exercise consisted of progressive resistance training of upper and lower extremities, and of the trunk, static and dynamic balance training, and functional training. The control group intervention contained a guided non-physical social program of similar frequency, duration, length and location. The study endpoints consisted of 180

symptoms of CHF (fatigue, dyspnea), NT-proBNP, and body measurements and laboratory results, together known as parameters of the â&#x20AC;&#x153;metabolic syndromeâ&#x20AC;?. The syndrome is defined as the presence of three out of five measurements and laboratory results. These measurements include waist, blood pressure, fasting glucose, triglycerides and high-density lipoprotein (HDL)cholesterol. We substituted glucose for glycated hemoglobin (HbA1c), since fasting blood collection was not feasible for logistic reasons. As a result, out of 434 residential elderly 163 were eligible. Unfortunately, only 52 (12%) were willing to participate in the study. Of these 52, six dropped out during the study for reasons not related to the exercise. The remaining 46 participants were 85Âą6 years, 65% females and all had been randomized to exercise (n=24) or control groups (n=22). The intention-to-treat analysis showed no significant change of any of a variable in the exercise groups compared to controls. The as-treated group completed half or more of the 32 training sessions. The females of the latter group had a significant reduction in waist size [-8 (range -15 - -2) cm]. The controls showed a significant triglycerides decrease [-8 (-17-0) mg/dL] and an NT-proBNP increase [67 (7-126) pg/mL]. The attendance of the residents was inversely related to the nonfasting triglycerides in the exercise group (p=0.024). We concluded that over a short period of time guided physical exercise, performed by residential elderly, had neither beneficial effects on symptoms and a biomarker of CHF, nor on cardiovascular-metabolic indices. We have explained the neutral results with the high age (85 years on average) accompanied by loss of muscle endurance, and with a training offer that is not enough personalized.

181

In brief, the outcomes of this thesis are: 9 the prevalences of CHF in residential elderly are 23% in Groningen, which is consistent with literature (9), and >30% in Aruba (chapters 2.1 and 2.2).

9 in Dutch residential elderly, undetected diagnoses of CHF often occur. The use of natriuretic peptides, as an additional test for CHF, needs to be further studied. In residential elderly in Groningen, the incorrect diagnoses of CHF can be reduced by applying natriuretic peptides (chapter 2.1 and 2.2).

9 determination of NT-proBNP concentrations in the urine of CHF patients is not more favourable compared to NT-proBNP plasma levels. This finding results from the large intraindividual variabilities of urine NT-proBNP which is comparable to that in plasma NTproBNP (chapter 2.3).

9 in residential elderly, natriuretic peptide testing has prognostic values [HR of NT-proBNP 1.02 and BNP 1.16 (vs. HR (10) of BNP is 2.2)], of which the outcome of BNP is consistent with the literature (10) (chapter 3).

9 residential elderly with CHF need more ADL-help, when compared to those without CHF (chapter 4).

9 a short-term period of guided exercise performed by residential elderly has neither effect on CHF nor on cardiovascular indices (i.e. the metabolic syndrome). There are no studies to compare with (chapter 5).

182

183

184

Chapter 6.2

DISCUSSION

185

We assessed our hypothesis (Introduction) according to which general HF guidelines (2-4) are not applicable to residential elderly with regard to diagnostics. Therefore, we applied natriuretic peptides (NPs) threshold values, derived from HF guidelines (2-4) to the residential elderly cohort in Groningen (chapter 2.1). It should be noticed that both, the ESC and Multidisciplinary HF guidelines (2;3) advise to use the same cut-off values (NT-proBNP 125 and BNP 35 pg/mL), which values are lower compared to the NICE HF guideline (i.e. NT-proBNP 400 and BNP 100 pg/mL) (4). Using these cut-off values from the guidelines, we calculated the predictive values for the CHF diagnosis in residential elderly of a single nursing home in Groningen (Table 1). We also calculated the predictive values at higher thresholds such as at NT-pro BNP 900 pg/mL and BNP 100 pg/mL. Years after our study, Mason et al. (11) published a paper on predictive values of NPs for CHF. The cut-off values of their study in British residential elderly were slightly different from those chosen in our study. In accordance with the NICE HF guideline (4) Mason et al. validated their findings for NT-proBNP and BNP cut-off points (Table 1). The Mason group (11) and we not only evaluated isolated natriuretic peptides but also combinations of medical history, symptoms and signs, ECG, and natriuretic peptides to predict the presence of CHF. We both found that any combination of studied parameters did not improve the predictive values, as determined by natriuretic peptides either (Table 1). The outcomes of Masonâ&#x20AC;&#x2122;s study, specifically with regard to PPV, correspond well with our findings.

186

Table 1. Predictive values of NT-proBNP and BNP for the heart failure diagnosis in two populations of residential elderly (RE): the Groningen (NL) (Chapter 2.1) and the British population (11). Cut-off value pg/mL

sensitivity

specificity

PPV

NPV

Groningen (NL) RE NT-proBNP1

125

1.00

0.28

0.29

1.00

NT-proBNP

450

0.71

0.67

0.42

0.91

NT-proBNP

900

0.67

0.85

0.56

0.89

BNP1

0.88

0.39

0.30

0.91

BNP

0.88

0.63

0.31

0.89

BNP2

100

0.71

0.70

0.41

0.88

NT-proBNP2

400

0.56

0.69

0.35

0.84

NT-proBNP

760

0.62

0.75

0.42

0.87

BNP

115

0.67

0.68

0.38

0.88

British RE (11)

Residential elderly, RE; PPV, positive predictive value; NPV, negative predictive value; NTproBNP, N-terminal amino pro B-type natriuretic peptide; B-type natriuretic peptide. 1 ESC HF guideline (2) and multidisciplinary HF guideline (3) 2 NICE HF guideline (4).

In addition, both groups report that when applying HF guidelines (2-4) to Dutch and British residential elderly (11), more than half of the initial CHF diagnoses are missed (too many false negatives). Also lowering the threshold values (2;3), when compared to the NICE (4) HF guideline, does not provide any substantial improvements. We and Mason et al. further estab-

187

lished that in residential elderly, the calculated predictive values are not reliable for ruling-in but only for ruling out CHF. In this thesis, we thus adapt the algorithm of the Multidisciplinary HF guideline (3) to residential elderly (Figure 1). These modifications include the use of NPs for excluding CHF only, the use of echocardiography for ruling in CHF, the use of NICE cut-off values (4), and utilization of echocardiography in primary care laboratories (www.Certe.nl). However, Oudejans et al. (12) demonstrated in geriatric outpatients with a prevalence of >40% CHF, that natriuretic peptide testing in combination with other diagnostic tests, are very accurate for both, ruling out and ruling in CHF. Therefore, the PPV of CHF in our group might have been lower due to a lower a priori chance on having CHF based on the lower CHF prevalence (23%) when compared to the a priori chance of CHF in outpatients (CHF prevalence >40%). In contrast to the results of the Oudejans group, but in concert with those of Mason and our study group [Chapter 1.2 and reference (11)], we have revised our recommendations for residential elderly. We had to reject our hypothesis in part, since CHF cannot be diagnosed in residential elderly by making use of HF guidelines. In addition, our hypothesis was fair in part because CHF can be ruled out using natriuretic peptides according to the HF guidelines.

Based on our new insights, we advise to make an echocardiogram for diagnosing CHF in residential elderly. Theoretically, NPs are recommended to rule out CHF although we realize that excluding CHF does not comply with daily practice. Further, We processed these recommendations into an algorithm for CHF diagnostics (i.e. non-acute HF) for residential elderly (Figure 1).

188

189

190

Chapter 6.3

RECOMMENDATIONS AND FUTURE PERSPECTIVES

191

In the discussion, we have concluded that general HF guidelines cannot be used for diagnosing CHF in residential elderly. Furthermore, echocardiography is warranted to establish the diagnosis of CHF in the individual residential elderly. To that end, we have customized the Dutch diagnostic algorithm for CHF (3) to that particular group (Figure 1). In the future, it is crucial to assess this adjusted CHF diagnostic algorithm on accuracy and costs. We therefore recommend to validate the customized CHF diagnostic algorithm in new populations of residential elderly. In the introduction, the paragraph on HF diagnostics, we emphasized that access to echocardiography was limited. Since recently, echocardiography has become available for primary care patients in The Netherlands, that barrier to accurately diagnose CHF seems to be removed. We subsequently argue for the option of on-site echocardiography for those (residential) elderly who are immobile.

To editors of the Multidisciplinary HF guideline (3), we give in consideration to adapt the algorithm (3) to residential elderly. The adaptations concern non-acute HF exclusively. They include the use of echocardiography for ruling in CHF, the use of NPs only for excluding CHF, and referral for echocardiography also in primary care laboratories (www.Certe.nl).

In Chapter 3, we showed the prognostic values of NPs for one-year mortality of residential elderly with an established diagnosis of CHF. We therefore advise to add natriuretic peptide testing to â&#x20AC;&#x153;advance care planningâ&#x20AC;? for those (residential) elderly facing a limited life expectancy. In Chapter 4, we demonstrated the relation between CHF and the needs for more ADL-help, compared to residential elderly without CHF. This relation made us to advise care teams to recognise higher requirements for ADL-help, in residential elderly with CHF. In Chapter 5, we observed that residential elderly who performed exercise during four months, changed neither their CHF nor their CV-metabolic indices. We have explained the absence of improvements to the high age and subsequent loss of muscular endurance. In addition, the 192

training frequency and duration were too modest contributing to the absence of improvements as well. Based on these findings, we recommend in future to study residential elderly with CHF during and after a personalized exercise program of three times a week in the long-run. An example of a personalized program is the option for the aged to select his type of training himself. Such programs may be more effective compared to programs imposed (13). Besides, we emphasize to pay attention regularly to the motivation of residential elderly to exercise, since we found low participation rates in our study.

As a consequence of our results partly processed in the diagnostic CHF algorithm specific for residential elderly (Chapter 6.2), we recommend the following: To editors of continuing medical education (e.g. www.CME-scholing.nl), we suggest to record our new insights into CHF in education programs for physicians elderly medicine (SO). The insights include the found CHF prevalence of at least 23% in residential elderly. Those also include that CHF can be better diagnosed according to the diagnostic CHF algorithm that we customized to residential elderly (Figure 1). Furthermore, that advance care for residential elderly, can be improved by the determination of natriuretic peptides also. Importantly, neither CHF symptoms and a marker nor CV-metabolic indices may not improve in residential elderly, after exercise during a four months period. However, in the long-run, beneficial effects of exercise in residential elderly may not be impossible, when using personalized conditions. This includes that the resident chooses himself whether he starts walking, home-training, swimming or visits the fitness club.

193

Residential elderly with complaints and symptoms suspected for non-acute heart failure (HF) in a long-term care facility or primary care medicine

Patient diagnosed with HF

Patient not diagnosed with HF

Physical examination, ecg, lab.

Physical examination, ecg, NT-

consider comorbidity or HF de-

proBNP or BNP (Hb, eGFR, TSH,

cline consider further steps

K+ glucose)

elderly medicine in primary care

Yes

NT-proBNP<400 pg/mL or BNP <100 pg/mL, ecg undisturbed No

HF is unlikely. Consider other causes

Further diagnostics by the SO/GP

Neg.

Echocardiography Pos

HFrEF and or HFpEF is likely.

HF outpatientsâ&#x20AC;&#x2122; clinic Cause analysis/ treatable HF

Cause analysis /no treatable

cause.

cause for HF.

Consultation with SO /GP

Referral back to SO/GP with

Treatment

treatment advice.

Figure 1. Diagnostic algorithm for non-acute heart failure in residential elderly.

194

BNP, B-type natriuretic peptide; NT-proBNP, nitrogen-terminal-proBNP; SO, specialist ouderengeneeskunde; GP, general practitioner.

References 1

Leening MJ, Siregar S, Vaartjes I, Bots ML, Versteegh MI, van Geuns RJ, et al. Heart disease in The Netherlands: a quantitative update. Neth Heart J 2014;22(1):3-10.

McMurray JJ, Adamopoulos S, Anker SD, Auricchio A, Bohm M, Dickstein K, et al. ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure 2012: The Task Force for the Diagnosis and Treatment of Acute and Chronic Heart Failure 2012 of the European Society of Cardiology. Developed in collaboration with the Heart Failure Association (HFA) of the ESC. Eur Heart J 2012;33(14):1787-1847.

Voors AA, Walma EP, Twickler TB, Rutten FH, Hoes AW. [Multidisciplinary guideline 'Heart failure 2010']. Ned Tijdschr Geneeskd 2011;155:A2957.

Mant J, Al-Mohammad A, Swain S, Laramee P. Management of chronic heart failure in adults: synopsis of the National Institute For Health and clinical excellence guideline. Ann Intern Med 2011;155(4):252-259.

Bruins S, Fokkema MR, Rรถmer JW, DeJongste MJ, van der Dijs FP, van Ouweland JM, et al. High intraindividual variation of B-type natriuretic peptide (BNP) and aminoterminal proBNP in patients with stable chronic heart failure. Clin Chem 2004;50(11):2052-8. 195

Grebe C, Brandenburg H. [Resident assessment instrument. Application options and relevance for Germany]. Z Gerontol Geriatr 2015;48(2):105-13.

Goff DC Jr, Lloyd-Jones DM, Bennett G, Coady S, D'Agostino RB, Sr, Gibbons, et al. 2013 ACC/AHA guideline on the assessment of cardiovascular risk: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. J Am Coll Cardiol 2014 ;63(25 Pt B):2935-2959.

Weening-Dijksterhuis E, de Greef MH, Krijnen W, van der Schans CP. Group exercise has little effect on ADL, physical fitness, and care dependency in frail institutionalised elderly people: a randomized controlled trial. Thesis 2014; www.Hanze.nl.

Hancock C, Close H, Mason JM, Murphy JJ, Fuat A, Singh R, et al. High prevalence of undetected heart failure in long-term care residents: findings from the Heart Failure in Care Homes (HFinCH) study. Eur J Heart Fail 2012;15(2):158-65.

Bibbins-Domingo K, Gupta R, Na B, Wu AH, Schiller NB, Whooley MA. N-terminal fragment of the prohormone brain-type natriuretic peptide (NT-proBNP), cardiovas cular events, and mortality in patients with stable coronary heart disease. JAMA 2007;297:169â&#x20AC;&#x201C;176.

Mason JM, Hancock HC, Close H, Murphy JJ, Fuat A, de Belder M, et al. Utility of biomarkers in the differential diagnosis of heart failure in older people: findings from the heart failure in care homes (HFinCH) diagnostic accuracy study. PLoS.One 2013;8(1); e53560.

196

Oudejans I, Mosterd A, Bloemen JA, Valk MJ, van Velzen E, Wielders JP, et al. Clinical evaluation of geriatric outpatients with suspected heart failure: value of symptoms, signs, and additional tests. Eur J Heart Fail 2011;13(5):518-27.

Beckers PJ, Denollet J, Possemiers NM, Wuyts K, Vrints CJ, Conraads VM. Maintaining physical fitness of patients with chronic heart failure: a randomized controlled trial. Eur J Cardiovasc Prev Rehabil 2010;17(6):660-7.

197

198

Chapter 7

SAMENVATTING, DISCUSSIE, AANBEVELINGEN EN TOEKOMSTPERSPECTIEVEN

199

Chapter 7.1 SAMENVATTING Een nadelig gevolg van de verbeterde overleving van cardiovasculaire ziektes (CVD) (1) is dat de cardiovasculaire morbiditeit is toegenomen. Zoals in de introductie van dit proefschrift wordt benadrukt, is het aantal mensen met hartfalen (HF) aanzienlijk toegenomen, terwijl tegelijkertijd de levensverwachting in de Nederlandse populatie sterk is verbeterd gedurende de laatste eeuw (www.cbs.nl; www.nationaalkompas.nl). Omdat HF zich veelal op hogere leeftijd manifesteert, behoren patiënten met chronisch hartfalen (CHF) meestal tot de categorie ouderen. Er bestaat echter een groot tekort aan kennis over HF bij toenemende leeftijd en bij kwetsbare ouderen die afhankelijker worden van zorg. Ouderen die het meest afhankelijk zijn van zorg, worden opgenomen in een verzorgings- of verpleeghuis. Deze ouderen worden in deze thesis aangeduid als residentiële ouderen. Het gebrek aan onderzoeksgegevens en resultaten is voornamelijk ontstaan doordat residentiële ouderen geweerd worden uit studies vanwege hun hoge leeftijd en multimorbiditeit, zoals CVD. Vanwege een gebrek aan gegevens zijn er geen richtlijnen beschikbaar om HF bij residentiële ouderen te diagnostiseren en te behandelen. Vanuit dit gezichtspunt bestaat er discussie over de toepasbaarheid van bestaande HF-richtlijnen bij residentiële ouderen die ontwikkeld zijn voor CHF patiënten in het algemeen. Richtlijnen voor residentiële ouderen met CHF worden node gemist omdat de CHF diagnose lastig te stellen is: ouderen neigen ertoe om klachten toe te schrijven aan het ouder worden in plaats van CHF of andere ziektes. Daardoor wordt de behandeling van CHF vaak ten onrechte of helemaal niet uitgevoerd. Als gevolg van deze ongewenste situatie gaat er kwaliteit van leven verloren en nemen de kosten toe. Deze bevindingen vormen de aanleiding om CHF bij residentiële ouderen aan nader onderzoek te onderwerpen. In Hoofdstuk 2.1 wordt de schaalgrootte van CHF verkend door de prevalentie van CHF onder residentiële ouderen te bepalen. Wanneer de algemene HF richtlijnen (2-4) worden toegepast, blijkt uit recente literatuur de HF diagnostiek minder accuraat te zijn ten gevolge van de laagvoorspellende waarden van klachten en symptomen, medische voorgeschiedenis en elektrocar200

diogram (ECG). Uit de medische gegevens van residentiële ouderen van één verpleeghuis in Groningen werden de klachten, de symptomen en de medische voorgeschiedenis verzameld en werd het aantal ouderen met CHF geïnventariseerd. Vervolgens werden alle ouderen onderworpen aan een lichamelijk onderzoek en maakten we een ECG van elk van hen. Daarna bepaalden we hoe accuraat de eerdere CHF diagnostiek [zonder gebruik van natriuretische peptiden (NP’s)] en de huidige CHF diagnostiek (met NP’s) waren, omdat in tegenstelling tot de brede erkenning van NP’s als screening test van CHF, deze NP’s nog steeds weinig gebruikt worden als biomarker bij residentiële ouderen. Het beperkte gebruik bij ouderen wordt veroorzaakt doordat de afkapwaarden van NP’s in deze groep onvoldoende gevalideerd zijn. Tot slot werden de ECG’s van alle residentiële ouderen beoordeeld.

Een cardiologenpanel stelde de diagnose CHF vast of sloot deze uit door, achtereenvolgens de uitslagen van de NP’s en resultaten van echocardiografie te beoordelen. Echocardiografisch onderzoek bleek mogelijk in 98% van de ouderen. Het resultaat van de huidige studie was dat CHF werd bevestigd in 24/103 (23%) van de residentiële ouderen. Voorafgaand aan de studie, werden 22 ouderen geïdentificeerd met CHF. Bij deze 22 ouderen met CHF, vastgesteld voorafgaand aan de studie, werd CHF bevestigd bij 9 ouderen en bij 13 ouderen werd de diagnose CHF verworpen. De diagnostische accuratesse van NT-proBNP bij 450 pg/mL bestond uit een sensitiviteit van 0,71, specificiteit van 0,67, positief voorspellende waarde (PPV) van 0,42 en negatief voorspellende waarde (NPV) van 0,91. De diagnostische accuratesse van BNP bij 100 pg/mL toonde een 0,71 sensitiviteit, 0,70 specificiteit, 0,41 PPV en 0,88 NPV. In het kort waren de meest opvallende resultaten van de huidige studie dat meer dan de helft van de CHF diagnoses werd gemist (namelijk 15/24) of onjuist werd gesteld (namelijk 13/22). Daarom kan worden geconcludeerd dat het identificeren van CHF inaccuraat is. Maar uitgaande van de hoge NPV, lijkt het gebruik van NP’s als additionele diagnostische test veelbelovend, zelfs bij residentiële ouderen. Het laatste moet echter nog nader worden geëvalueerd. De CHFprevalentie onder residentiële ouderen werd vastgesteld op 23%. 201

In Hoofdstuk 2.2 is onderzocht of de uitkomsten van hoofdstuk 2.1 consistent zijn met de CHFprevalentie in Arubaanse verpleeghuizen. Er werd voor Arubaanse verpleeghuizen gekozen vanwege onze interesse in een vergelijkbare onderzoekspopulatie met die in Nederland, die echter woont in een andere omgeving en er een andere levensstijl op na houdt. Er werd gebruik gemaakt van hetzelfde studieontwerp zoals gehanteerd in hoofdstuk 2.1 om de prevalentie van CHF bij Arubaanse residentiële ouderen te berekenen. Een ander doel was het maken van een inschatting of de toepassing van NP’s voor het opsporen van CHF bij residentiële ouderen in een ander deel van het Nederlands Koninkrijk verbeterd kan worden. Het belangrijkste verschil met de studie uit hoofdstuk 2.1 was dat er geen echocardiogram van alle residentiële ouderen kon worden verkregen vanwege logistieke belemmeringen. Daarom kon niet worden voldaan aan de diagnostische standaard voor CHF (2-4). Als resultaat werden 51 van de 235 Arubaanse ouderen met een gemiddelde leeftijd van 78±8 jaar in de studie opgenomen. Afgaand op de medische gegevens, werd CHF vastgesteld bij 7/51 ouderen. Echter twee van de zeven ouderen hadden geen CHF. Verder werd CHF bevestigd bij 16 van de 51 (31%) ouderen. Van de 16 ouderen met CHF, waren 11 niet eerder gediagnostiseerd met CHF en van 5/16 ouderen bij wie CHF eerder was vastgesteld, werd de diagnose bevestigd. Kort gezegd, de algemene richtlijnen voor HF worden laagfrequent toegepast bij Arubaanse residentiële ouderen. Vergeleken met residentiële ouderen uit Groningen, is de prevalentie van CHF waarschijnlijk minstens zo hoog onder de Arubaanse residentiële ouderen. Als conclusie blijkt de herkenning van CHF ernstig onderschat te worden onder Arubaanse residentiële ouderen. De opsporing van CHF kan worden verbeterd door een geschikte richtlijn te implementeren, inclusief de bepaling van BNP.

Hoofdstuk 2.3 behandelt het probleem van de grote intra-individuele variaties [individuele variatie coëfficiënt (CVi)] van NT-proBNP in plasma, aangezien grote CVis de toepasbaarheid van NT-proBNP beperken voor o.a. de NT-proBNP-geleide therapieoptimalisatie bij individuele pati202

enten met CHF (5). Op zoek naar meer betrouwbare CVis werden de concentraties van urineNT-proBNP (NT-proBNPu) vergeleken met die van plasma-NT-proBNP (NT-proBNPp) bij patienten met CHF woonachtig op Curaçao. Op één enkele dag ("op-1-dag") werden urine en bloed monsters afgenomen: zes bloedmonsters elke 2 uur en de spontaan uitgeplaste urines gedurende 24 uur. Op vijf opeenvolgende dagen ("dag-tot-dag") werden verzameld: vijf bloedmonsters, vijf geforceerd uitgeplaste urine monsters en vijf volledige 24-uurs urines. Op dezelfde dag van zes opeenvolgende weken ("week-tot-week") werden verzameld: één bloedmonster, één geforceerd uitgeplast urinemonster en een volledige 24-uurs urine. Uit deze bloed- en urinemonsters werden de totale CVs (CVts), de CVis en de coëfficiënten van het percentage verandering van de referentiewaarden (RCV) berekend. In de studie werden 25 patiënten met CHF opgenomen met een gemiddelde leeftijd van 61 jaar (spreiding 36-80), 60% was man en de gemiddelde linker ventrikel ejectiefractie (LVEF) was 36±15%. De mediane CVis voor NT-proBNPp waren 9% (“op-1-dag”), 18% (“dag-tot-dag”) en 30% (“week-totweek”). De CVis van NT-proBNPu waren respectievelijk 34%, 21% en 28%. De RCV van NT-proBNPu was hoger dan of gelijk aan de RCV van NT-proBNPp, zelfs na correctie voor creatinine en de periode (in uren) waarin de urine zich in de blaas had verzameld. De gegevens van de “dagtot-dag” en “week-tot-week” monsters doen vermoeden dat (immunoreactive) concentraties van NT-proBNPu in de urine niet ondubbelzinnig correleren met de (immunoreactive) NT-proBNP concentraties in plasma, tot aan de NT-proBNPp drempel van ongeveer 310 pg/mL (Figuur 1). Daarnaast werd opgemerkt dat boven deze drempel de hogere NT-proBNP concentraties, zoals weergegeven in Figuur 1, niet verklaard kunnen worden door een verhoging van de bloeddruk gedurende de dag (resultaten niet weergegeven). Veronderstelt kan worden dat filtratie van natriuretische peptiden kan worden beschouwd als dominante factor voorbij de drempelwaarde en dat bij lagere NT-proBNP-waarden, filtratie en terugresorptie meer dynamisch interacteren. Geconcludeerd kan worden dat het bepalen van NT-proBNP in urine geen voordelen biedt bij patiënten met stabiel chronisch hartfalen. 203

Figuur 1. De relatie tussen NT-proBNP in plasma and NT-proBNP in geforceerd uitgeplaste urines verzameld volgens de “dag-tot-dag” en “week-tot-week” protocollen.

De data zijn afkomstig van 22 patiënten en in totaal 182 corresponderende plasma en urine monsters. De “dag-tot-dag”- en week-tot-week”-bloedmonsters werden gebruikt. Voor de urine werden de geforceerd uitgeplaste urinemonsters gebruikt. In hoofdstuk 3 werd de prognostische waarde onderzocht van NP’s op de sterfte in één jaar bij hetzelfde cohort residentiële ouderen dat werd bestudeerd in hoofdstuk 2.1. De klinische relevantie hiervan is dat als NP’s gerelateerd worden aan de prognose van residentiële ouderen met CHF, deze biomarkers kunnen worden benut voor “advance care planning”. Deze vorm van 204

zorgplanning is gerelateerd aan de kwaliteit van leven en wordt belangrijk wanneer de levensverwachting beperkt is (www.Verenso.nl). Om de sterfte in één jaar te onderzoeken, werden gedurende één jaar de overleving en het overlijden van residentiële ouderen bijgehouden. Omdat tien residentiële ouderen niet konden worden vervolgd doordat zij naar onbekende bestemmingen

verhuisden

onderzochten

resterende

residentiële

ouderen.

Achttien uit 93 bewoners (gemiddelde leeftijd 81 ± 3 jaar, 66% vrouw) stierven binnen één jaar(niet-overlevenden). Een wederzijds aangepaste Cox proportionele risico regressieanalyse werd uitgevoerd. Er vonden correcties plaats voor zes vooraf gedefinieerde chronische ziekten, immobilisatie, leeftijd, geslacht, NT-proBNP en BNP. De gevonden resultaten lieten zien dat beide NP’s (NT-proBNP en BNP) de één-jaars sterfte significant voorspelden (respectievelijk, HR 1,02 en p = 0,001 en HR 1,16 en p = 0,003). Bovendien nam het risico op overlijden in één jaar exponentieel toe parallel aan de stijging van BNP en NT-proBNP in de hogere concentraties. Bijvoorbeeld, een toename van NT-proBNP tot 2.000 pg/mL was geassocieerd met een 60% hoger risico op overlijden binnen één jaar. Concluderend zijn zowel BNP als NT-proBNP onafhankelijke voorspellers van de één-jaars mortaliteit onder residentiële ouderen met CHF. Daarbij neemt het sterfterisico toe bij NP concentraties die verhoogd zijn tot ver boven de diagnostische afkapwaarden van de richtlijnen (24). In hoofdstuk 4 is de relatie bestudeerd tussen CHF en de steun die nodig is voor residentiële ouderen met CHF tijdens hun activiteiten van het dagelijks leven (ADL). Daarvoor werd hetzelfde cohort residentiële ouderen onderzocht als in hoofdstuk 2.1. Het doel van deze studievraag was aandacht te trekken van zorgteams voor de behoefte aan steun onder residentiële ouderen met

CHF.

Voor deze studie werd dezelfde groep residentiële ouderen onderzocht als in hoofdstuk 2.1. Om gegevens over ADL te verzamelen, werden de “Minimal Data Set (MDS)” items van het “Resident-Assistant Instrument” gebruikt. De MDS bestaat uit systematische observaties van de af205

hankelijkheid en van de geboden ondersteuning en heeft als doel om de afhankelijkheid en de hulp te kwantificeren (6). Het MDS instrument is in feite een softwareprogramma voor de invoer van observaties van individuele residentiële ouderen tijdens vast omschreven ADL. Een voorbeeld van een dergelijke activiteit is lopen. De verzorgende observeert en scoort (kwantificeert) of het individu in staat is om bepaalde afstanden onafhankelijk te lopen en of hij hulp nodig heeft bij het lopen. Van de 103 residentiële ouderen met en zonder CHF, werden de scores van ADL-afhankelijkheid en ADL-hulp vergeleken en gecorrigeerd voor de verschillen tussen beide groepen door middel van multivariabele logistische regressieanalyse. CHF was geassocieerd met ADL-hulp (OR 4,68 en p=0,015) echter, CHF was niet gerelateerd aan de ADLafhankelijkheid.

Geconcludeerd kan worden dat CHF bij residentiële ouderen geassocieerd is met de behoefte aan meer ADL-hulp. In dit verband is CHF van belang voor zowel de residentiële ouderen (verlies van autonomie en levenssatisfactie) als ook voor de organisatie (personele inzet).

In hoofdstuk 5 wordt een fysiek oefenprogramma onderzocht onder residentiële ouderen met CHF. De reden voor deze studie is inzicht te verkrijgen in de gevonden discrepantie tussen de sedentaire levensstijl van residentiële ouderen en de gedane aanbevelingen voor fysieke activiteit van de HF-richtlijnen (2-4). Ook stimuleren familieleden hun ouders in verzorgingshuizen om lichamelijk actief te worden. Echter, om diverse redenen blijven residentiële ouderen liever sedentair. Bovendien, uit angst voor vallen, geven ook verzorgers er de voorkeur aan dat residentiële ouderen zoveel mogelijk sedimentair zijn.

Een sedentaire levensstijl is gerelateerd aan een verhoogd risico op toename van atherosclerose en CV-metabole ziekten, zoals CHF en het metabool syndroom (7).Onderzocht werd of lichamelijke training een gunstige invloed had op symptomen van CHF, NT-proBNP, het metabool syndroom en CV-metabole indices bij residentiële ouderen. Een pilot, enkelvoudig206

geblindeerde, gerandomiseerde en gecontroleerde haalbaarheidstudie werd uitgevoerd in vier verzorgingstehuizen. In samenwerking met andere onderzoekers (9) werd bestudeerd of een gestandaardiseerd trainingsprogramma invloed heeft op symptomen van CHF, NT-proBNP en het metabool syndroom. De andere onderzoekers (8) onderzochten of hetzelfde gestandaardiseerde oefenprogramma de, door hen gekozen, meer functionele eindpunten beïnvloedt in vergelijkbare groepen. De trainingsinterventie omvatte een begeleide training van matig-tot-hoge intensiteit van één uur, twee keer per week gedurende 16 opeenvolgende weken. De training bestond uit progressieve weerstandstraining van bovenste en onderste extremiteiten en van de romp, uit statische en dynamische evenwichtstraining en uit functionele training. De interventie van de controle groep bestond uit een begeleid, niet-fysiek, sociaal programma van dezelfde frequentie, duur, lengte en locatie. De studie eindpunten bestonden uit symptomen van CHF (vermoeidheid, kortademigheid), en NT-proBNP, uit lichaamsmetingen en uit laboratoriumresultaten, die als combinatie bekend zijn als parameters van het "metabool syndroom". Het syndroom is gedefinieerd als aanwezigheid van drie uit vijf lichaamsmetingen en laboratoriumresultaten. Deze metingen omvatten de taille, de bloeddruk, de nuchtere glucose, de triglyceriden en het high-density lipoproteïne (HDL)-cholesterol. Glucose wordt vervangen door geglyceerd hemoglobine (HbA1c), aangezien een nuchtere bloedafname logistiek niet haalbaar bleek te zijn. Als resultaat, kwamen van de 434 residentiële ouderen, 163 in aanmerking voor deelname. Helaas, waren slechts 52 (12%) bereid om deel te nemen aan de studie. Van deze 52, vielen er zes af tijdens de studie vanwege redenen die niet waren gerelateerd aan de training. De resterende 46 deelnemers waren 85±6 jaar, 65% was vrouw en allen werden gerandomiseerd voor trainingsgroep (n=24) of controle groep (n=22). De “intention-to-treat analyse” toonde geen significante verandering van welke variabele dan ook bij de trainingsgroep vergeleken met de controles. De groep die geanalyseerd werd “as-treated” voltooide de helft of meer van de 32 trainingssessies. De vrouwen van de laatstgenoemde groep bleken een afname van taille omvang te hebben [-8 (spreiding -15--2) cm]. De controles toonden een significante verlaging van triglyceriden [-8 (-17-0) mg/dL] en een stijging van NT-proBNP [67 (7-126) pg/mL]. 207

Het aantal bijgewoonde sessies was omgekeerd gerelateerd aan de niet-nuchtere triglyceriden bij de trainingsgroep (p = 0,024). Wij concludeerden dat fysieke training onder begeleiding, uitgevoerd door residentiĂŤle ouderen gedurende een korte periode, geen effect had op symptomen en een biomarker van CHF, noch op CV-metabole indexen. Een mogelijke verklaring voor de neutrale resultaten is de hoge leeftijd van de residentiĂŤle ouderen (gemiddeld 85 jaar), die gepaard gaat met een verlies van spieruithoudingsvermogen en door een trainingsaanbod dat onvoldoende is afgestemd op het individu.

208

Kort samengevat zijn de resultaten van deze thesis: 9 de prevalenties van CHF bij residentiële ouderen zijn 23% in Groningen hetgeen strookt met de literatuur (9), en >30% op Aruba (hoofdstukken 2.1 en 2.2).

9 bij Nederlandse residentiële ouderen komt CHF vaak voor. Het gebruik van natriuretische peptiden, als aanvullende test voor CHF, moet verder bestudeerd worden. Bij residentiële ouderen in Groningen, kunnen onjuiste diagnoses van CHF worden teruggebracht door diagnostisch natriuretische peptiden toe te passen (hoofdstuk 2.1 en 2.2).

9 bepaling van NT-proBNP concentraties in de urine van patiënten met CHF heeft geen voordelen vergeleken met plasma-NT-proBNP concentraties. Deze bevinding komt voort uit de grote intra-individuele variabiliteit van urine-NT-proBNP die vergelijkbaar is met die in plasma-NT-proBNP (hoofdstuk 2.3).

9 bij residentiële ouderen, zijn bepalingen van natriuretische peptiden van prognostische betekenis (HR van NT-proBNP 1,02 en BNP 1,16 (HR van BNP is 2.2 (10)), waarbij het resultaat van BNP overeenkomt met dat in de literatuur (10) (hoofdstuk 3).

9 CHF bij residentiële ouderen is gerelateerd aan de behoefte aan meer ADL-hulp vergeleken met ouderen zonder CHF (hoofdstuk 4).

9 een korte periode van training uitgevoerd door residentiële ouderen heeft geen effect op CHF noch op CV indices (zoals b.v. het metabool syndroom). Er zijn geen studies voorhanden om dit te kunnen vergelijken (hoofdstuk 5).

209

210

Hoofdstuk 7.2

DISCUSSIE

211

We toetsten onze hypothese (Introductie) die stelt dat de diagnostiek van niet-acuut HF, zoals die aanbevolen wordt in de algemene HF richtlijnen (2-4), niet toegepast kan worden op residentiële ouderen. Daartoe, gebruikten we de afkapwaarden van de natriuretische peptiden (NP’s) zoals aanbevolen door de HF richtlijnen (2-4) in het cohort Groninger residentiële ouderen (hoofdstuk 2.1). Het is van belang te vermelden dat zowel de ESC als de Multidisciplinaire HF richtlijn (2;3) dezelfde afkapwaarden adviseren (NT-proBNP 125 and BNP 35 pg/mL) en dat de laatste waarden lager zijn vergeleken met die van de NICE HF richtlijn (namelijk NTproBNP 400 and BNP 100 pg/mL) (4). Gebruikmakend van deze afkapwaardes uit de richtlijnen, berekenden we de voorspellende waarden voor de diagnose CHF, in residentiële ouderen van één Gronings verpleeghuis (Tabel 1). Tevens bepaalden we de voorspellende waardes ook bij hogere afkapwaarden zoals bij NT-pro BNP 900 pg/mL en BNP 100 pg/mL. Jaren na onze studie publiceerde Mason et al. (11) een artikel over voorspellende waarden van NP’s voor CHF bij Engelse residentiële ouderen. De afkapwaarden van de studie van Mason et al. verschilden weinig van die uit onze studie. De laatstgenoemde studie groep valideerde haar resultaten met de NT-proBNP en BNP afkapwaarden volgens de NICE guideline (4)(Tabel 1).

212

Tabel 1. Voorspellende waarden van NT-proBNP en BNP voor de diagnose hartfalen in twee populaties residentiële ouderen: de Groninger (NL) (hoofdstuk 2.1) en de Britse populatie (11). afkapwaarde pg/mL

sensitiviteit

specificiteit

PPV

NPV

Groninger RE (NL) NT-proBNP1

125

1.00

0.28

0.29

1.00

NT-proBNP

450

0.71

0.67

0.42

0.91

NT-proBNP

900

0.67

0.85

0.56

0.89

BNP1

0.88

0.39

0.30

0.91

BNP

0.88

0.63

0.31

0.89

BNP2

100

0.71

0.70

0.41

0.88

NT-proBNP2

400

0.56

0.69

0.35

0.84

NT-proBNP

760

0.62

0.75

0.42

0.87

BNP

115

0.67

0.68

0.38

0.88

British RE (11)

Residentiële ouderen (residential elderly), RE; Nederland, NL; PPV, positief voorspellende waarde; NPV, negatief voorspellende waarde; NT-proBNP, N-terminal amino pro B-type natriuretic peptide; B-type natriuretic peptide. 1 ESC HF richtlijn (2) en multidisciplinaire HF richtlijn (3) 2 NICE HF richtlijn (4).

De groep van Mason (11) en wij onderzochten niet uitsluitend NP’s maar ook combinaties van medische voorgeschiedenis, klachten en symptomen, ECG, en NP’s met het doel om CHF aan te tonen. Wij beiden vonden dat geen van de mogelijke combinaties van parameters, de voor213

spellende waarden verbeterde vergeleken met die berekend met behulp van NP’s alleen. De resultaten van Mason’s studiegroep, vooral wat betreft de PPV, komen goed overeen met onze bevindingen. Beide onderzoeksgroepen rapporteren ook dat toepassing van de HF richtlijnen (2-4) op Nederlandse en Engelse residentiële ouderen, daarin resulteert dat meer dan de helft van de CHF diagnoses gemist worden (te veel fout negatieven). Ook verlaging van de afkapwaarden (2;3) ten opzichte van de NICE richtlijn (4) levert geen enkele verbetering op. Samen met de Mason groep, komen we tot de conclusie dat de gevonden voorspellende waarden ongeschikt zijn om CHF aan te tonen maar alleen geschikt zijn om CHF uit te sluiten.

Daarom besloten we om in dit proefschrift het algoritme van de Multidisciplinaire HF richtlijn (3) aan te passen voor residentiële ouderen. Deze aanpassingen houden in dat NP’s alleen gebruikt worden om CHF uit te sluiten, dat echocardiografie noodzakelijk is om CHF aan te tonen, dat de afkapwaarden uit de NICE richtlijn (4) gebruikt worden, en dat echocardiografie ook door laboratoria voor de eerstelijnsgezondheidszorg uitgevoerd worden (www.Certe.nl). Toch toonde Oudejans et al. (12) aan bij geriatrische poliklinische patiënten van wie >40% CHF had, dat de bepaling van NP’s in combinatie met andere diagnostische tests wel degelijk accuraat is om zowel CHF aan te tonen als om CHF uit te sluiten. Waarschijnlijk was de PPV voor CHF bij ons onderzoekscohort lager omdat de vooraf kans op CHF lager was door de lagere CHF prevalentie 23%, vergeleken met de vooraf kans bij de poliklinische patiënten (CHF prevalentie >40%). In tegenstelling tot de resultaten van de Oudejans groep (12) maar in overeenstemming met de resultaten van Mason et al. en onze studie groep [Chapter 1.2 en (11)], hebben we onze aanbevelingen voor residentiële ouderen bijgesteld. Onze hypothese is deels verworpen omdat de algemene HF richtlijnen ongeschikt zijn om CHF aan te tonen bij residentiële ouderen. Daarnaast werd onze hypothese deels aanvaard omdat CHF wel uitgesloten kan worden met behulp van NP’s volgens de HF richtlijnen.

214

Uitgaande van onze pas verworven inzichten wordt geadviseerd om een echocardiogram te laten maken om CHF aan te tonen. Theoretisch worden NPâ&#x20AC;&#x2122;s aanbevolen om CHF uit te sluiten alhoewel we ons realiseren dat uitsluiten van CHF niet volstaat in de dagelijkse praktijk. Deze adviezen hebben we opgenomen in een diagnostisch algoritme voor CHF (d.w.z. niet-acuut HF) voor residentiĂŤle ouderen (Figuur 1).

215

216

Hoofdstuk 7.3

AANBEVELINGEN EN TOEKOMST PERSPECTIEVEN

217

In de discussie stelden we vast dat algemene HF richtlijnen ongeschikt zijn om CHF te diagnostiseren bij residentiële ouderen. Ook is bij de individuele residentiële oudere, een echocardiogram essentieel om de diagnose CHF te stellen. Daartoe hebben we het Nederlandse diagnostische algoritme voor alle CHF patiënten (3), aangepast aan die specifieke groep (Figuur 1). In de toekomst is het van essentieel belang om dit aangepaste diagnostische CHF algoritme te toetsen op accuratesse en kosten. Daarom raden we aan om het aangepaste diagnostische CHF algoritme te valideren in nieuwe populaties residentiële ouderen.

In de introductie, het hoofdstuk over HF diagnostiek benadrukten we de beperkte mogelijkheden om een echocardiogram aan te vragen. Met de recente beschikbaarheid van echocardiografie voor eerstelijns patiënten in Nederland, lijkt die barrière, om CHF accuraat te diagnostiseren geslecht te zijn. Aansluitend pleiten we ervoor om een echocardiogram thuis of in de instelling te kunnen maken voor die residentiële oudere die immobiel zijn. We geven redacteuren van de Multidisciplinaire HF richtlijn (3) in overweging om de richtlijn aan te passen aan residentiële ouderen. De aanpassingen betreffen alleen niet-acuut HF. De wijzigingen bestaan daar uit dat CHF alleen kan worden aangetoond met een echocardiogram, dat CHF uitgesloten kan worden met NP’s en dat ook verwezen kan worden naar eerstelijns laboratoria voor echocardiografie (www.Certe.nl).

In Hoofdstuk 3 presenteerden we de prognostische betekenis van NP’s voor één-jaars overleving van residentiële ouderen met bewezen CHF. Op grond daarvan wordt geadviseerd om een NP bepaling toe te voegen aan de palliatieve zorgplanning voor (residentiële) ouderen met een beperkte levensverwachting. We laten in Hoofdstuk 4 de relatie zien tussen CHF en de behoefte aan meer ADL-hulp vergeleken met residentiële ouderen zonder CHF. Vanwege deze relatie raden we zorgteams aan om oog te hebben voor toegenomen behoeften aan ADL-hulp van residentiële ouderen met CHF.

218

Hoofdstuk 5 leert ons dat bij residentiële ouderen die trainden gedurende vier maanden, geen verandering gezien werd van CHF noch van CV-metabole indices. Het uitblijven van verbetering werd verklaard door de hoge leeftijd en het bijkomend verlies van spieruithoudingsvermogen. Bovendien was de frequentie van trainen te laag en de duur te kort hetgeen ook bijdroeg aan het uitblijven van effect. Op basis hiervan wordt aanbevolen om, in de toekomst, residentiële ouderen te bestuderen tijdens en na een individueel trainingsprogramma van drie maal per week gedurende een langere tijd. Een voorbeeld van een individueel programma vormt de mogelijkheid voor de oudere om zelf zijn trainingsvorm te kiezen. Dergelijke programma’s lijken effectiever te zijn dan vaststaande programma’s. Bovendien wordt sterk aangeraden om regelmatig aandacht te besteden aan de motivatie van residentiële ouderen om te trainen omdat we te maken hadden met een klein aantal deelnemers in onze studie (Hoofdstuk 5).

Uitgaande van onze resultaten die deels verwerkt zijn in het diagnostisch CHF algoritme specifiek voor residentiële ouderen (chapter 6.2), doen we de volgende aanbevelingen: Aan redacteuren van medische bijscholing (bv CME-scholing.nl), geven we de suggestie om onze nieuwe inzichten op het gebied van CHF te verwerken in bijscholingen voor specialisten ouderengeneeskunde (SO). Het gaat om de gevonden CHF prevalentie van tenminste 23% bij residentiële ouderen. Bovendien dat CHF bij voorkeur gediagnostiseerd wordt volgens het diagnostische CHF algoritme dat we aangepast hebben aan residentiële ouderen (Figuur 1). Verder dat palliatieve zorg voor de residentiële oudere beter gepland kan worden door gebruik te maken van een natriuretische peptide bepaling. Ook van belang is dat noch CHF symptomen en een CHF marker, noch CV-metabole indices bij residentiële ouderen verbeteren na een vier maanden durende training. Daarentegen is het niet onmogelijk dat fysieke training voordelen oplevert wanneer de training langere periode plaatsvindt en op individuele leest geschoeid is. Hieronder wordt verstaan dat de oudere zelf kiest of hij gaat lopen, fietsen, zwemmen of de fitnessclub bezoekt.

219

ResidentiĂŤle ouderen met klachten en symptomen verdacht voor niet-acuut hartfalen (HF) in een zorginstelling of eerstelijnsgezondheidszorg

Patient met de diagnose HF

Patient bij wie HF niet is vastge-

Lichamelijk onderzoek, ecg, NT-proBNP

Lichamelijk onderzoek, ecg, lab.

of BNP (Hb, eGFR, TSH, K+, glucose)

Overweeg comorbiditeit, toename HF en verwijzing cardioloog ouderengeneeskunde 1e lijn

NT-proBNP<400 pg/mL of BNP<100 pg/mL, afwijkend ECG Nee

HF is onwaarschijnlijk, overweeg andere oorzaken van klachten

Neg.

echocardiografie Pos.

Behandeling van HF door SO of huisarts (HA)

HFrEF / HFpEF is waarschijnlijk

HF polikliniek Onderzoek naar (behandelbare)

Analyse van oorzaak/ geen behan-

oorzaak HF

delbare oorzaak van HF

Overleg met SO/HA over gewenste

Verwijzing terug naar SO/HA met

behandeling. Behandeling

behandeladvies

Figuur 1. Diagnostisch algoritme voor niet-acuut hart falen bij residentiĂŤle ouderen. NT-proBNP, B-type natriuretic peptide; NT-proBNP, amino-terminaal-pro BNP; SO, specialist ouderengeneeskunde; HFrEF, hartfalen met afgenomen ejectie fractie; HFpEF, hartfalen met behouden ejectie fractie. 220

DANKWOORD Jaren geleden stond ik bij een verpleeghuisbewoner verdacht van hartfalen. Bij het raadplegen van de toen vigerende richtlijnen voor hartfalen, dacht ik direct dat de diagnostiek van hartfalen beter moest kunnen. Nu schrijf ik het dankwoord van mijn proefschrift waarin ik aantoon dat de kwaliteit van leven van verpleeghuisbewoners met hartfalen inderdaad verbetert als je de diagnose correct stelt. Veel is daaraan vooraf gegaan. Het is fantastisch dat dit gelukt is. De vele mensen die hieraan hebben bijgedragen wil ik graag bedanken.

Mijn eerste dank gaat uit naar de verpleeg- en verzorgingshuisbewoners, hun mantelzorgers en de patiënten. Zonder hen waren de onderzoeken niet mogelijk geweest. Prof. dr. F.A.J. Muskiet, beste Frits, van het begin af aan was je enthousiast om (NT-pro)BNP te gaan bepalen bij verpleeghuisbewoners in het kader van hartfalen diagnostiek. Na elke onderzoeksbespreking, gaf je “college” niet alleen over hartfalen, (patho)fysiologie, veroudering en insuline resistentie, maar ook over laag-gradige ontsteking o.a. in de hersenen, vitamine D en life-style. Ik heb gretig naar je geluisterd. Van jou heb ik geleerd op het scherpst van de snede te discussiëren. Prof. J.L. Hillege, beste Hans, je vroeg altijd eerst wat ik wilde bereiken met het onderzoek. Daarna kwam je een met een praktisch voorstel voor berekeningen. Je hebt bijgedragen aan de groei van mijn statistisch inzicht. Dr. M.J.L. de Jongste, beste Mike, je hebt me altijd positief gesteund bij het verwezenlijken van dit proefschrift. Bij het maken van onderzoeksdesigns, zei je dat je je tijd ver vooruit was. Dat bleek ok zo te zijn. Ons onderzoeksdesign werd later op grote schaal toegepast. Ook heb je mij goed begeleid bij het raadplegen van anderen op het juiste moment. Ik heb heel veel van je geleerd. Elk overleg gaf mij hernieuwde energie en richting. Daarvoor wil ik je hartelijk bedanken. Dr. I.C.C. van der Horst, beste Iwan, dank voor je opbouwende commentaar. Prof. dr. R.A. de Boer, beste Rudolf dank voor de prettige samenwerking.

221

De leden van de beoordelingscommissie wil ik hartelijk bedanken voor de tijd die ze hebben genomen om mijn proefschrift te beoordelen: prof. de. W.P. Achterberg (beste Wilco) en prof. dr. A.W. Hoes. Prof. dr. A.A. Voors, beste Adriaan, fijn dat je de tijd hebt genomen om mijn proefschrift te beoordelen en dat je de tijd gaat nemen om te opponeren.

In het verpleeghuis “het Zonnehuis-Noord”, Zuidhorn is mijn onderzoek begonnen. Samen met jullie, Mike en Frits, hebben we ons voorstel voor hartfalen diagnostiek met behulp van (NT)proBNP geïntroduceerd en konden we van start. Drs. J. Koster, beste Johan dank voor jouw inspanning om goede (98%) echografische afbeeldingen te maken. Dank aan de laboratorium medewerkers van het Zonnehuis, jullie waren altijd bereid om bloed af te nemen en ecgs te maken, en dank aan de administratieve medewerkers voor jullie hulp bij het regelen van afspraken. Heel prettig om met je samen te werken, Judith, in de (voormalige) Westerburcht. Ik bewaar goede herinneringen aan de tweejaarlijkse onderzoeksbesprekingen met de Zonnehuisgroep in Soesterberg. Dank Zonnehuisgroep voor de financiële steun. Dankzij de flexibiliteit van jullie, dr. E. Weening en dr. M.H.G. de Greef, konden we aanhaken aan het bewegingsonderzoek in vier verzorgingshuizen in Groningen. Ik heb de samenwerking met jou Betsie, en de gesprekken met jou, Mathieu als heel inspirerend ervaren. Het management team van De Hoven, hebben het ons financieel mogelijk gemaakt om ecgs en NT-proBNP bepalingen te laten doen. Mijn dank gaat uit naar de interesse voor ons onderzoek die de Wetenschappelijke Raad van De Hoven, Prof. dr. D.L. Knook en Meindert Bolt hebben getoond. Dank voor de leuke samenwerking met jullie, de bewegingsmedewerkers uit Appingedam.

Van het Universitair Medisch Centrum Groningen (UMCG), wil ik bedanken Eduard Heine, voor het bewaren van onze bloedmonsters op het laboratorium. Egbert Knol, je hebt statistische bewerkingen van de scorelijsten goed uitgelegd, bedankt. Alma Guikema, dank je voor je hulp bij het printen en verzenden van de boekjes. Eline, ik heb het als verrijking ervaren om samen met je te schrijven aan het onderzoek dat je op Aruba hebt uitgevoerd. Wat hebben we gestaag ge222

werkt aan ons stuk, Anneliene. Ik heb respect voor je nauwkeurigheid van data verwerking. Dank je voor onze prettige samenwerking op afstand.

Beste Corina, Jos, Meindert, Evert, Luch, Hilde, Nazir, Judith en Victoria, wat zijn onze toetsgroep bijeenkomsten van belang geweest. Daarin konden we veilig moeilijkheden en mogelijkheden bespreken in de ouderengeneeskunde. Hartelijk dank daarvoor. Christien, we hebben niet alleen gezocht naar goeie Engelse termen. Ik ervaar het als heel prettig om levensvragen met je te bespreken en daarna samen te lachen. Zo’n vriendschap geeft kracht. Ik ben zo blij dat je een van mijn paranimfen bent. Daphne, dank voor je Engelse woorden! Ria, dank voor de vele gesprekken over hoe onze doelen te bereiken. Feilloos wist je direct knelpunten te noemen en deze te relativeren met een lach. Laten we daarmee doorgaan. Lex, dank voor je bijdrage aan het Nederlands. Ik mag je niet bedanken Engbert, maar noem hier wel je bijdrage aan tekst en opmaak. Als laatste maar zo belangrijk, kom ik toe aan mijn familie. Lieve Els en Hanneke, jullie zijn als zussen een inspiratie bron voor mij hoe te gaan voor wat je wilt. Altijd heb ik steun gevoeld bij je in dit traject. Ik ben er trots op Els, dat je mijn paranimf bent. Lieve Philip, Erik, Wendelien en Sunna, er is in de wereld geen groter geluk dan jullie. Dankzij jullie, doe ik mee met de, voor mij “volgende generatie”. De warmte en humor die ik voel als we bij elkaar zijn, maken me tot een gelukkig mens. Patijn, Rick, Froukje en Danielle, dank je wel dat jullie ons zijn komen versterken. En Quinten, dankzij jou heb ik een nieuwe rol. Tenslotte, Willem, jou wil ik bedanken voor alle jaren continue steun. Steeds heb je meegedacht over mijn toekomst, welke stappen ik zou zetten. Je hebt vele uren alleen doorgebracht als ik achter de pc zat. Met al mijn liefde draag ik dit proefschrift aan je op. Je bent de belangrijkste in mijn leven die dit allemaal mogelijk heeft gemaakt. Je bent mijn basis van waaruit ik kan groeien. Ik kijk ernaar uit dat we weer meer tijd hebben voor elkaar.

223

CURRICULUM VITAE Maaike Barents obtained her medical degree in 1977 at the Rijks Universiteit Groningen (RUG). She started to investigate the diagnosis of cowâ&#x20AC;&#x2122;s milk allergy among infants while working for the Dutch Youth Health Organization in cooperation with the UMCG, Childrenâ&#x20AC;&#x2122;s Department (prof. dr. H.S.A. Heymans). In 1994, she acquired her specialty degree as physician-sexuologist at the Dutch Society of Sexuologists, NVVS. After her residency in the Academic Rehabilitation Centre in Haren (prof. Dr. J.H. Arendzen and drs. L.D.W. Vos) she decided to specialize in elderly care medicine at the Free University of Amsterdam in collaboration with nursing home het Zonnehuis in Zuidhorn. In the same time, she took the initial steps for the work as described in this thesis in collaboration with the UMCG (Thorax centre, Department of cardiology (dr. M.J.L. de Jongste) and Department of Clinical Chemistry (prof. F.A.J. Muskiet)). When settled as elderly care physician in Groningen and in addition to patient care, she completed her Ph D work at the last described Department. She is married to Willem Bohmers and they have two sons (Philip, 1978 and Erik, 1980) and two daughters (Wendelien, 1982 and Sunna, 1984).

224