Page 1

Right Atrial Size and Tricuspid Regurgitation Severity Predict Mortality or Transplantation in Primary Pulmonary Hypertension Miguel Bustamante-Labarta, MD, Sergio Perrone, MD, Ricardo Leon de la Fuente, MD, Pablo Stutzbach, MD, Ricardo Perez de la Hoz, MD, Augusto Torino, MD, and Roberto Favaloro, MD, Buenos Aires, Argentina

Primary pulmonary hypertension (PPH) is a fatal illness. In advanced stages only transplantation is able to increase survival. Echocardiography is useful for the assessment of these patients, but there is limited information about its prognostic value. With this goal, 25 consecutive patients, age: 36.7 ⴞ 12.7 years, were studied and followed up for a mean period of 29 months (range: 0.2-84). Eleven echocardiographic parameters of cardiac anatomy, function, and hemodynamics were assessed. Age and sex were also analyzed. Death and heart-lung transplantation were considered end-points. Thirteen events (Death: 8; transplantation: 5) occurred in the fol-

Primary pulmonary hypertension (PPH) is a fatal

illness with a dramatic and inexorable natural evolution to death in nearly all patients. Although spontaneous remission has been described, this is a very rare situation.1 After diagnosis, 5-year survival rate is about 21%, with a mean period of survival of 2.5 years.2,3 Despite its ominous prognosis, the natural evolution can be different among individuals. A subgroup of patients under treatment with vasodilators is able to show improvement in terms of clinical status and even survival.4,5 Nevertheless, in advanced stages of the illness, only lung or heartlung transplant increase survival.6,7 Unfortunately, there exists a dramatic imbalance between the number of patients waiting for being transplanted and the number of potential donors. Therefore, a careful analysis of clinical data and accessory information obtained from complimentary studies to detect prognostic predictors is crucial to identify the severity of PPH among patients waiting for a transplant. To date, a few clinical and hemodynamic parameters

From the Echocardiography Section and Intrathoracic Organ Transplantation Division, ICYCC-Fundacio´n Favaloro. Reprint requests: Dr Miguel H. Bustamante-Labarta, ICYCC, Fundacio´n Favaloro, Seccio´n Ecocardiografı´a, Av. Belgrano 1743, CP (1093), Buenos Aires, Argentina (E-mail: mbustamante@ Copyright 2002 by the American Society of Echocardiography. 0894-7317/2002/$35.00 ⫹ 0 27/1/123962 doi:10.1067/mje.2002.123962


low-up (11 of 13 in the first year). Kaplan-Meier estimated survival free from transplantation at 5 years was 40% (95% CI: 23%-70%). In the univariate analysis, RAA (HR: 1.1, P ⴝ .0004), TR (HR: 2.7, P ⴝ .02), and RVET (HR: 0.98, P ⴝ .02) showed statistically significant relation with survival free from transplantation. Multivariate analysis showed that RAS (HR: 1.10, 95% CI: 1.04-1.17, P ⴝ .001) and TR (HR: 2.52, 95% CI: 1.01-6.3, P ⴝ .047) were independent risk factors of transplantation and death. The use of these findings on the management of patients with PPH should be tested in larger studies. (J Am Soc Echocardiogr 2002;15:1160-4.)

have been described that are predictive of PPH survival.8 Among them, New York Heart Association (NYHA) functional class III or IV, presence of Raynaud phenomenon, elevated mean right atrial (RA) pressure, elevated mean pulmonary artery pressure, decreased cardiac index, and decreased diffusing capacity for carbon monoxide have been related to a more severe prognosis. Although echocardiography is a very useful tool in the assessment of PPH, very little information has been reported about its prognostic value in this illness.9,10 With this in mind, we have designed a study in which the primary goal was to assess the prognostic value of echocardiography in PPH.

METHODS Study Group From December 1993 to May 2000, 78 patients with pulmonary hypertension of diverse causes who were considered potential candidates for lung or heart/lung transplantation were prospectively and consecutively studied by the Echocardiography Section at the Institute of Cardiology and Cardiovascular Surgery-Favaloro Foundation. From this population, 25 patients (mean age: 37.6 ⫾ 12.7 years [range: 20-59 years], 19 female [76%] and 6 male [24%]) who fulfilled the following inclusion criteria entered the final study group: (1) diagnosis of PPH established by means of right and left hemodynamic study

Journal of the American Society of Echocardiography Volume 15 Number 10

and ventilation/perfusion lung centellography performed at the institution, after ruling out other causes of secondary pulmonary hypertension (SPH). (2) Follow-up monitored by the Intrathoracic Organ Transplantation Division of the institution. Fifty-three patients were excluded from the study for the following reasons: confirmed diagnosis of SPH (n⫽ 38), pulmonary hypertension of unclear cause (n ⫽ 4) and patients with PPH or SPH treated and followed up in a different institution (n ⫽ 11).

Echocardiographic Study

Bustamante-Labarta et al 1161

formed by protocol in every patient to define the cause of pulmonary hypertension.

Follow-up Most of the patients underwent a strict periodic clinical evaluation. In 22 patients the follow-up was performed in our institution. In the 3 remaining patients (living in other cities of the country) data were obtained by telephone interview conducted by trained personnel. Transplantation or cardiac death was defined as an end point (combined events). Transplantation was included as an end point to avoid statistical bias considering the cross-over of the more severely ill patients to the transplantation group. The follow-up period was estimated as the time lapse (in days) between the initial echocardiographic study and the last date of control, transplantation, or death.

Transthoracic echocardiography was performed in a standard manner with a 2.5-MHz phase-array transducer with Hewlett-Packard SONOS 1500, 2000 or 5500 (Hewlett Packard Co, Andover, Mass) or ATL HDI 5000 (Advanced Technology Laboratories, Inc, Bothell, Wash) machines in all the patients.11-13 The population analyzed in this study was also included in another study protocol previously reported by our group, designed to assess the use of transesophageal echocardiography in the etiologic diagnosis of patients with severe pulmonary hypertension as candidates to heart/lung transplantation.14 For that reason, all the patients underwent a transesophageal echocardiography study, which was useful for the assessment of pulmonary artery diameters. A 5-MHz biplane or multiplane transducer was used for transesophageal studies. The following echocardiographic parameters of cardiac anatomy, function, and hemodynamic status were prospectively analyzed in each patient: right ventricular (RV) diastolic diameter, measured from the apical 4-chamber view at the level of the tip of the tricuspid valve, left ventricular (LV) diastolic diameter, systolic RA size and left atrial size, measured as planimetric areas from the apical 4-chamber view; diastolic main pulmonary artery diameter (measured from the transesophageal approach), tricuspid regurgitation (TR) severity: classified in 3 degrees: mild (grade 1), moderate (grade 2), and severe (grade 3), on the basis of color Doppler regurgitant jet area/RA area ratio; acceleration time (AT) of the RV outflow tract flow spectral signal, obtained with pulsed Doppler with the sample volume positioned immediately below the pulmonic valve (parasternal short axis of the base); RV ejection time (RVET), obtained from the same spectral signal; AT/RVET ratio; RV first derivative of pressure (dP/dt), estimated from the rising slope of the continuous wave Doppler signal of the tricuspid regurgitation jet (change in pressure from 0.5-2 m/s); and systolic pulmonary pressure, estimated as the addition of RV-RA pressure gradient (obtained from the continuous wave Doppler of TR) to RA pressure (calculated after evaluation of diameter and behavior of inferior vena cava under respiratory maneuvers, from the subcostal approach).

Complete measurements were obtained in 22 patients. AT, RVET, and AT/RVET were not measured in 1 patient and dP/dt was not estimated in 2 other patients because of suboptimal Doppler spectral signals of RVOT flow and TR, respectively. Mean values and SD of analyzed parameters are depicted as follows: RV diastolic diameter: 40.92 ⫾ 11.28 mm; LV diastolic diameter: 32.80 ⫾ 7.4 mm; left atrial size: 13.34 ⫾ 4.01 cm2; RAS: 28.24 ⫾ 10.96 cm2; pulmonary artery diameter: 36.76 ⫾ 8.05 mm; AT: 61.88 ⫾ 19.15 msec; AT/RVET: 23.79 ⫾ 7.47; systolic pulmonary pressure: 88.48 ⫾ 20.4 mm Hg; RVET: 262.08 ⫾ 38 msec; dP/dt: 65.97 ⫾ 28.97 mm Hg/sec; and TR: 1.52 ⫾ 0.65. Table 1 summarizes values of the analyzed echocardiographic parameters and results from statistics. TR was present in all the patients and its severity was estimated as mild in 14 (56%), moderate in 9 (36%), and severe in 2 (8%) patients.

Other Studies

Follow-up Data

In addition to standard studies (ECG, laboratory, and chest x-ray examination) and echocardiography, a ventilationperfusion lung scan and a hemodynamic study was per-

Patients were prospectively evaluated and followed up for a mean period of 29 months (median: 12 months, range: 0.2-84). During the follow-up period,

Statistical Analysis Continuous variables are expressed as mean ⫾ 1 SD. The survival curve was calculated by Kaplan-Meier method, the standard error by Greenwood formula and the 95% CI was based on the logarithm transformation. The log rank test and Cox regression analysis were used to identify risk factors of death or transplantation. Values of P ⬍ .05 were considered statistically significant. In addition to the echocardiographic parameters described, age and sex were also considered in the prognostic evaluation.

RESULTS Echocardiographic Data

Journal of the American Society of Echocardiography October 2002

1162 Bustamante-Labarta et al

Table 1 Values of echocardiographic measurements and statistics. Patient


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 Mean/value SD


Age (y) RVDD

59 26 42 47 20 55 34 21 20 22 31 54 29 41 54 39 49 42 21 35 31 52 30 20 43 37.6 12.7

40 39 39 30 50 53 29 44 33 34 19 53 53 27 41 43 39 26 40 43 33 60 68 49 38 40.92 11.28


30 30 26 43 27 31 30 32 48 35 38 24 32 52 28 29 28 41 39 33 25 33 20 34 32 32.80 7.40






46 33 36 33 35 33 32 32 42 28 34 29 34 60 47 45 34 27 29 50 33 30 46 35 36 36.76 8.05

55 85 78 55 65 50 60 70 96 — 70 91 80 40 65 50 55 20 40 40 45 55 90 79 51 61.88 19.15

22 29 30 17 28 20 21 26 35 — 24 32 29 20 20 14 23 6 18 22 18 21 42 31 23 23.79 7.47

115 78 90 51 106 130 94 105 41 103 75 85 65 76 70 93 92 62 94 105 85 95 111 99 92 88.48 20.40


16 24 14 17 15 27 14 17 9 25 18 41 12,7 17,5 12 27 18 23 16,9 18,4 14 12 10 33 15 48 25 57 11 32 12 18 7 28 12 18 6,8 34 10 36 9,2 20 15 42 10 34 17 22 14 35 13.34 28.24 4.01 10.96



262 60 295 78,9 258 65,2 305 100 225 60 243 75 285 75 270 60 276 31 — 150 315 50 286 62 271 33 220 30 325 50 345 75 230 27,3 260 50 215 — 190 100 250 60 265 50 214 — 252 75 233 100 262.08 65.97 38.10 28.05


1 2 1 1 1 2 1 2 1 1 1 1 2 1 1 1 3 1 2 1 2 3 2 2 2 1.52 0.65


⫹ ⫹ ⫹ ⫹

⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹

RVDD, Right ventricular diastolic diameter; LVDD, left ventricular diastolic diameter; LAS, left atrial size; RAS, right atrial size; PAd, main pulmonary artery diameter; AT, right ventricular outflow tract spectral flow signal acceleration time; RVET, right ventricular ejection time; SPP, systolic pulmonary pressure; TR, tricuspid regurgitation.

Table 2 Echocardiographic parameters that evidenced an statistically significant relationship with survival free from transplantation Parameter




Mean values


Univariate analysis

yes no yes no yes no

13 12 13 12 13 11

34.42 cm2 21.53 cm2 1.77 1.25 247.54 s 279.27 s

10.77 cm2 6.43 cm2 0.73 0.45 28.92 s 41.64 s

HR: 1.1 P ⫽ .0004 HR 2.7 P ⫽ .02 HR: 0.98 P ⫽ .02

Multivariate analysis

HR: 1.1, 95% CI: 1.04–1.17, P ⫽ .001 HR: 2.52, 95% CI: 1.01–6.3, P ⫽ .047 NS

RAS, Right atrial size; TR, tricuspid regurgitation severity; RVET, right ventricular ejection time; HR, hazard ratio; CI, confidence interval; NS, not significant.

8 patients died and 5 were transplanted (combined events: 13 patients). Three of 5 transplanted patients died within the 24 hours of the surgery. Most of the events occurred in the first year of follow-up (7 death and 4 transplants). Kaplan-Meier estimated survival free from transplantation at 5 years was 40% (95% CI: 23%-70%). Prognostic value of echocardiographic parameters. In the univariate analysis, RAS (hazard ratio:

1.1, P ⫽ .0004), TR (hazard ratio: 2.7, P ⫽ .02), and RVET (hazard ratio: 0.98, P ⫽ .02) showed an statistically significant relationship with survival free from transplantation. Mean values and SD of these parameters, compar-

ing those patients with and those without events, are described in Table 2. Multivariate analysis (log-rank test and Cox regression analysis) showed that only RAS (hazard ratio: 1.10, 95% CI: 1.04-1.17, P ⫽ .001) and TR (hazard ratio: 2.52, 95% CI: 1.01-6.3, P ⫽ .047) were independent risk factors of transplantation or death. Nine of 11 patients with RAS greater to the median (27 cm2) were transplanted or died during the first year of follow-up and all the patients (n ⫽ 6) wit RAS greater than the 75th percentile (34 cm2), were transplanted or died within this period. Figure 1 shows the likelihood of survival free from transplantation, comparing patients with RAS equal or

Journal of the American Society of Echocardiography Volume 15 Number 10

Bustamante-Labarta et al 1163

Figure 1 Probability of survival free from transplantation on basis of analysis of right atrial size (RAS). Full line, right atrial area less than or equal to median (27 cm2). Dotted line, right atrial area greater than median.

lower than the median and RAS greater than the median.

DISCUSSION The prognostic value of echocardiographic parameters of cardiac anatomy, function, and hemodynamic state in patients with PPH was prospectively assessed in this study. In the multivariate analysis the RAS and the severity of TR were found to be predictive factors of mayor events (heart/lung transplantation and death) in the analyzed population (25 patients followed up by a mean period of 29 months). One of the determinants of the RAS is the pressure on the atrium. As previously mentioned, the mean RA pressure has been reported to have prognostic value on survival in PPH.8 Therefore, it seems reasonable to think that both closely related parameters (RA pressure and size) are predictive of mortality. It is more controversial to explain the fact that TR severity was also predictive of a severe prognosis in PPH. TR is a common finding in patients with severe pulmonary hypertension but its severity varies between patients depending on different determinants. In a previous study reported by our group, the relationship between the degree of TR and echocardiographic parameters of right cardiac anatomy and function was analyzed in a group of 44 patients with severe pulmonary hypertension (mean

pulmonary systolic pressure: 93.79 mm Hg).15 TR was present in all patients (mild in 47.7%, moderate in 31.8%, and severe in 20.5%). In the univariate analysis, the RV diastolic dimension (r ⫽ 0.51, 95% CI ⫽ 0.24 ⬍ R ⬍ 0.70), the diameter of the tricuspid annulus (r ⫽ 0.49, 95% CI ⫽ 0.22 ⬎ R ⬎ 0.69), and the RAS (r ⫽ 0.47, 95% CI ⫽ 0.17 ⬍ R ⬍ 0.69) were associated with the severity of TR. As those parameters are representative of the tricuspid apparatus structure, it seems reasonable to consider that the enlargement of the right atrium, tricuspid annulus, right ventricle, or a combination of these is accompanied by more severe TR. In the current study, it is probable that both RAS and TR shared the same physiopathologic mechanism. To our knowledge, the prognostic value of RAS and TR on PPH outcome has not previously been reported, at least from results obtained in prospective studies. Yeo et al,10 in a retrospective study, reported that TR severity had prognostic value on adverse outcome in patients with PPH, but this result was obtained only in univariate analysis. Another measurement that showed prognostic value in our study, but only in the univariate analysis, was the RVET. This parameter gives information about the RV function. The assessment of systolic RV function by means of noninvasive and even invasive studies is still a challenge. Echocardiography offers different parameters for its evaluation. Nevertheless, the complexity of the shape of the RV chamber (complicating volume calculations), the

1164 Bustamante-Labarta et al

relative value of indirect measurements (presystolic, systolic, and diastolic times), and the impossibility to avoid the influence of preload and afterload secondary to interactions such as TR or pulmonary pressure make all these estimations unreliable. In our study we selected RVET and RV dP/dt as repeatable and easily obtained parameters. To date, there is little information about the prognostic value of RV function indexes on pulmonary hypertension. Among the limited published data, Tei et al9 described the use of a new echocardiographic index of ventricular performance. This index, defined as “the sum of isovolumic contraction time and isovolumic relaxation time divided by ejection time,” was found to be an independent predictor of survival in PPH within the multivariate model.10 Our study was prospective and we were not able to evaluate the prognostic use of such index considering that it was reported just after our study started. Limitations As many other published studies on PPH prognosis, the study group was relatively small. Two combined events (death and heart/lung transplantation) were chosen to evaluate the prognosis of patients with PPH. As previously described, transplantation was included as an end point to avoid statistical bias, considering the crossover of the more severely ill patients to this treatment. In our opinion, it is valid to consider heart/lung transplantation as a true “hard event” because only patients with severe clinical deterioration who are unresponsive to other treatments choose this surgical option. Although it was only an echocardiographic study, lack of information about clinical status or response to treatment should be considered as another limitation. Nevertheless, it is noteworthy that every patient was, at least once in the follow-up period, in NYHA functional class III or IV.

CONCLUSIONS RAS and the severity of TR, 2 parameters easily obtained by echocardiography, were predictive variables of survival free from transplantation in patients with PPH. The impact of these findings on the therapeutic management of this fatal illness should be addressed in larger prospective studies.

Journal of the American Society of Echocardiography October 2002

REFERENCES 1. Bourdillon PD, Oakley CM. Regression of primary pulmonary hypertension. Br Heart J 1976;38:264-70. 2. Fuster V, Steele P, Edwards WD, Gersh BJ, McGoon MD, Frye RL. Primary pulmonary hypertension: natural history and the importance of thrombosis. Circulation 1984;70:580-7. 3. D’Alonzo GE, Barst RJ, Ayres SM, Bergofsky EH, Brundage BH, Detre KM, et al. Survival in patients with primary pulmonary hypertension: results from a national prospective registry. Ann Intern Med 1991;115:343-9. 4. Rich S, Brundage BH. High dose calcium-channel-blocking therapy for primary pulmonary hypertension: evidence for long-term reduction in pulmonary arterial pressure and regression of right ventricular hypertrophy. Circulation 1987;76: 135-4. 5. Rich S, Kaufman E, Levy PS. The effect of high doses of calcium-channel blockers on survival in primary pulmonary hypertension. N Engl J Med 1992:327:76-81. 6. Pasque MK, Trulock EP, Kaiser LR, Cooper JD. Single-lung transplantation for pulmonary hypertension: three-months hemodynamic follow-up. Circulation 1991:84:2275-9. 7. Higenbottam TW, Spiegelhalter D, Scott JP, Fuster V, DinhXuan AT, Caine N. Prostacyclin (epoprostenol) and heartlung transplantation as treatment for severe pulmonary hypertension. Br Heart J 1993;70:366-70. 8. Sandoval J, Bauerle O, Palomar A, Gomez A, MartinezGuerra ML, Beltran M, et al. Survival in primary pulmonary hypertension: validation of a prognostic equation. Circulation 1994;89:1733-44. 9. Tei C. New non-invasive index for combined systolic and diastolic ventricular function. J Cardiol 1995;26:357-66. 10. Yeo TC, Dujardin KS, Tei C, Mahoney DW, McGoon MD, Seward JB. Value of a Doppler-derived index combining systolic and diastolic time intervals in predicting outcome in primary pulmonary hypertension. Am J Cardiol 1998;81: 1157-61. 11. Sahn DJ, DeMaria A, Kisslo J, Weyman AE (the Committee on M-mode Standarization of the American Society of Echocardiography). Recommendations regarding quantitation of M-mode echocardiography: results of survey of echocardiographic measurements. Circulation 1978;58:1072-83. 12. Mitchell MM, Sutherland GR, Gussenhoven EJ, Taams MA, Roelandt JR. Transesophageal echocardiography. J Am Soc Echocardiogr 1988;1362-77. 13. Seward JB, Khandheria BK, Oh JK, Abel MD, Hughes RW Jr, Edwards WD, et al. Transesophageal echocardiography: technique, anatomic correlations, implementation and clinical applications. Mayo Clin Proc 1988:63:649-80. 14. Bustamante-Labarta MH, Stutzbach P, Guevara E, Perez de la Hoz R, Fernandez MO, Perrone S, et al. Diagno´ stico etiolo´ gico de la hipertensio´ n pulmonar mediante eco-transesofa´ gico en candidatos a trasplante cardiopulmonar. Rev Arg Cardiol 1999;67:327-31. 15. Bustamante-Labarta M, Stutzbach P, Perrone S, Perez de la Hoz R, Guevara E, Fernandez M, et al. Que factores se asocian con el grado de insuficiencia tricuspı´dea en la hipertensio´ n pulmonar severa? Rev Arg Cardiol 1997;65:170.