- Email: [email protected]
Prognostic importance of atrial natriuretic peptide in patients with chronic heart failure
JACC Vol.
13, No. 7 June 1989:15349
1534
Prognostic Importance of Atria1 Natriuretic Peptide in Patients With Chronic Heart Failure STEPHEN
S. GOTTLIEB,
DEBORAH
AHERN,
MD, FACC,” MARRICK
RN, MILTON
PACKER,
L. KUKIN,
MD, FACC,
MD, FACC
New York, New York
Several circulating neurohormones have been shown to have prognostic significance in patients with chronic heart failure, but the relation between plasma levels of atria1 natriuretic peptide and mortality in this disorder remains unknown. Plasma levels of immunoreactive atria1 natriuretic peptide were measured in 102 patients in whom left ventricular ejection fraction, ventricular arrhythmias on ambulatory electrocardiographic recording and plasma levels of norepinephrine, renin activity, aldosterone and arginine vasopressin were also measured. Compared with patients with atrial natriuretic peptide concentrations below the median value of 125 pg/ml, patients with higher levels of the peptide had a higher plasma renin activity (8.9 -C 1.8 versus 2.6 f 0.4 ng/ml per h) and plasma norepinephrine (858 + 116 versus 538 & 45 pglml), more frequent prema-
As cardiac performance declines in patients with chronic heart failure, various neurohormonal systems are stimulated in an attempt to maintain circulatory homeostasis. This neurohormonal activation becomes progressively more marked as the disease state advances (1,2) and may accelerate the deterioration of left ventricular function that occurs over time (3,4). Either because they are a marker of disease severity or play a direct pathogenetic role, circulating levels of many neurohormones strongly predict the long-term prognosis of patients with heart failure (4). Patients with the most marked activation of the sympathetic nervous system (as reflected by elevated plasma norepinephrine levels) and of
From the Division of Cardiology, Department of Medicine, Mount Sinai School of Medicine of the City University of New York, New York, New York. This study was supported by grants T32 HL 07347and ROl-HL-25055 from the National Heart, Lung, and Blood Institute, National Institutes of Health, Bethesda, Maryland. Dr. Packer is the recipient of a Research Career Development Award (KO4-HL01229)from the National Institutes of Health. Manuscript received August 8.1988; revised manuscript received November 22, 1988,accepted January 9, 1989. *Current address and address for reorints: Stephen S. Gottlieb, MD, Division of Cardiology, University of Maryland School of Medicine, 22 South Greene Street, Baltimore, Maryland 21201. 01989
by the
American College of
Cardiology
ture ventricular depolarizations (4,485 f 715 versus 2,004 f 495/day) and more advanced hemodynamic abnormalities (all p < 0.05). During the subsequent 13 to 25 months of follow-up, patients with high levels of atria1 natriuretic peptide had a significantly lower rate of survival than did those whose initial circulating peptide concentrations were normal or mildly increased (p = 0.01). These data indicate that, in patients with chronic heart failure, plasma atrial natriuretic peptide provides important prognostic information. This may relate to the ability of the hormone to reflect the interplay of several pathophysiologic factors that contribute to mortality in this disease. (.I Am Co11Cardiol1989;13:1534-9)
the renin-angiotensin system (as reflected by hyponatremia) have the highest mortality rate (5,6). These neurohormonal measurements may provide more powerful prognostic information than do the invasive and noninvasive variables that are conventionally used to assess left ventricular function (536). Previous studies that have evaluated the relation between neurohormones and survival in chronic heart failure have focused on the activation of vasoconstrictor systems. Patients with advanced left ventricular dysfunction, however, also have elevated levels of endogenous vasodilators (for example, prostaglandins and atria1 natriuretic peptide) (710); yet, it is not known whether these dilator hormones have prognostic significance. If, like the vasoconstrictor hormones, circulating levels of these vasodilators increase in parallel to the severity of the underlying disease, the release of prostaglandins and atria1 natriuretic peptide might be expected to be greatest in patients with the highest mortality. On the other hand, if the vasodilator hormones act to limit the deleterious effects of the vasoconstrictor systems, high circulating levels of endogenous vasodilators might be associated with a favorable long-term outcome. To distinguish 0735-1097/89/$3.50
JACC Vol. 13, No. 7 June 1989: 1534-9
ATRIAL NATRIURETIC
between these two possibilities, we evaluated the relation between plasma atria1 natriuretic peptide and long-term survival in 102 patients with chronic heart failure.
Methods Study patients. Our study group consisted of 102 patients with chronic congestive heart failure recruited over a 12 month period. There were 81 men and 21 women, aged 34 to 90 years (mean 64). The cause of heart failure was dilated cardiomyopathy in 53 patients, ischemic heart disease in 43 patients and primary mitral or aortic regurgitation, or both, in 6 patients. All patients had a left ventricular ejection fraction ~40% by radionuclide ventriculography. Thirty-two patients were in New York Heart Association functional class II, 39 in class III and 31 in class IV. All patients were clinically stable on constant doses of digoxin and diuretics; any previous therapy with vasodilator drugs was withdrawn for 220 half-lives
Study Design Patients underwent the following evaluations over 2 to 5 days: 1) hormonal measurements (all 102 patients); 2) 24 h ambulatory electrocardiographic (ECG) monitoring (98 patients); and 3) right heart catheterization (34 patients). Hormonal analysis. After the patient had been recumbent in a quiet room for ~30 min, blood was collected from an indwelling catheter for the following analyses: I) plasma renin activity, aldosterone and arginine vasopressin (by radioimmunoassay): 2) plasma norepinephrine, epinephrine and dopamine (by radioenzymatic assay); and 3) plasma atria1 natriuretic peptide (by radioimmunoassay). For the last assay, blood was collected in prechilled ethylenediaminetetraacetic acid (EDTA) tubes with aprotinin, and the plasma was separated immediately and stored at -70” C until extracted and analyzed as previously described (11). Blood was obtained in duplicate (on 2 sequential days) for all assays in 88 of the 102 patients. There was a strong correlation between hormonal determinations performed on the 2 days: for example, atria1 natriuretic peptide on day 1 was closely correlated with atrial natriuretic peptide on day 2 (r = 0.91). Ambulatory electrocardiographic recordings. Twentyfour hour recordings were obtained in all patients within 24 h of hormonal determinations; 48 h recordings were obtained in 88 patients. These recordings were used to measure the frequency of premature ventricular depolarizations and episodes of ventricular tachycardia (23 premature ventricular depolarizations in a row at a rate >lOO/min). Hemodynamic measurements. Right heart catheterization and arterial cannulation were performed to measure right heart and systemic pressures, respectively. Patients were permitted to rest overnight to allow dissipation of the hemo-
GOTTLIEB ET AL. PEPTIDE AND SURVIVAL
1.535
dynamic effects related to intravascular instrumentation. On the next morning, after all cardioactive medications were withheld for 212 h, the following hemodynamic variables were measured repeatedly with the patient in the fasting state until hemodynamic stability was evident: mean arterial pressure, heart rate, left ventricular filling pressure, mean right atria1 pressure and cardiac output. The methods we used have been described in detail previously (12). Mean systemic pressures were determined by electronic filtration. Derived hemodynamic variables were calculated as follow: cardiac index (CI) = CO/body surface area (liters/ min per m’): stroke volume index (SVI) = CUHR (ml/m’); stroke work index (SWI) = SVI x (MAP - LVFP) x 0.0136 (g - m/m’); systemic vascular resistance = 80 X (MAP MRAP)/CO (dynes*s.cm-‘). where CO = cardiac output, MRAP = mean right atrial pressure, LVFP = left ventricular filling pressure, HR = heart rate and MAP = mean arterial pressure. Follow-up. Survival was assessed to the day of death or to June 1, 1988; follow-up ranged from 13 to 25 months. Cumulative survival curves were constructed by KaplanMeier survivorship methods (13), and the differences between survival curves were tested for significance by both Mantel-Cox log-rank and Wilcoxon-Breslow methods (14.15). Statistics. Qualitative and quantitative differences between subgroups of patientswere tested for significance with use of the chi-square statistic and the t test for paired data, respectively. Whenever hormonal and ECG data were obtained in duplicate, the mean value was used for all calculations. All statistical calculations were performed with Statistical Analysis System (SAS) computer programs. Group data are expressed as mean values t SEM.
Results Atria1 natriuretic peptide and survival. Plasma concentrations of atria1 nutriuretic peptide ranged from 11 to 650 pg/ml (median 125) in the 102 patients with chronic heart failure evaluated in this study. Eighty-nine patients had a concentration of the peptide >50 pg/ml (upper limit of normal for the laboratory assay). Patients with a plasma level greater than the median value of 125pgiml had a significantly higher mortality rate than did those with a lower circulating level (p = 0.01 by both Wilcoxon and Mantel-Cox) (Fig. 1). Atria1 natriuretic peptide concentration and severity of heart failure. Patients with a plasma level >125 pgiml had
clinical, neurohormonal and hemodynamic evidence for more severe heart failure than did patients with lower levels of the peptide. Patients with the higher concentrations had greater limitation of functional capacity, higher levels of plasma renin activity, aldosterone and norepinephrine and more severe hemodynamic abnormalities than did patients with lower levels (all p < 0.05) (Tables 1 and 2). Patients with
GOTTLIEB ET AL. ATRIAL NATRIURETIC
1536
JACC Vol. 13, No. 7 June 1989:1534-9
PEPTIDE AND SURVIVAL
Table 2. Hemodynamic Characteristics of 34 Patients Grouped According to the Median Value for Atria1 Natriuretic Pentide* ANP< 125 pg/ml
ANP >125 pgiml (n = 15) ANP> 125 pg/ml
01 0
1
I
I
I
I
I
I
3
6
9
12
15
18
21
I 24
MONTHS
Figure 1. Kaplan-Meier analysis of cumulative rates of survival in patients with heart failure stratified into two groups on the basis of median plasma concentration of atria1 natriuretic peptide (ANP) (125 pg/mU.
higher peptide levels also tended to have a lower left ventricular ejection fraction (p = 0.07) and more frequent
ventricular premature depolarizations (p < 0.05) and episodes of ventricular tachycardia, although the last difference was not statistically significant because of the large variance seen in patients with high peptide levels. When patients were grouped according to circulating levels of vasoconstrictor hormones, plasma levels of atria1 natriuretic peptide were most markedly elevated in patients with the highest levels of plasma renin activity and norepinephrine (Fig. 2).
Cardiac index (litersimin per m*) Stroke work index (g-m/m*) Mean arterial pressure (mm Hg) Left ventricular filling pressure (mm Hg) Mean right atrial pressure (mm Hg) Heart rate (beats/min) Systemic vascular resistance (dynesscm-‘)
2.2 ” 0.1 28 If:4 94 + 6 18 ‘- 3 6.5 2 1.3 87 2 3 1,857 2 140
pg/ml (n = 19)
Value ’
1.7 ? 17 c 82 t 26 +
co.05 co.05 co.05 co.05
0.1 2 3 2
9.4 2 2.0 79 + 3 2,198 ? 141
NS NS NS
*These patients were grouped according to the median value for atrial natriuretic peptide (ANP) for all 102patients studied. NS = not significant.
Prognostic importance of peptide levels. To determine whether the levels of atria1 natriuretic peptide seen in our patients with heart failure had independent prognostic significance or only reflected the importance of other factors, all of the variables that differed between the high and low peptide subgroups (including ejection fraction) were entered into a Cox regression model. Because only 33% of the patients had hemodynamic measurements, these were not entered into the analysis. According to the regression model, plasma atria1 natriuretic peptide provided independent prog-
Table 1. Demographic and Neurohormonal Characteristics of 102 Patients Grouped According to the Median Value for Plasma Atrial Natriuretic Peptide (ANP) ANP >I25 pglml (n = 51) Age (yr) Gender Male Female Cause of heart failure ICM PDC PVD NYHA functional class II III IV LV ejection fraction Plasma renin activity (&ml per h) Plasma aldosterone (ng/dl) Plasma norepinephrine (pg/ml) Plasma epinephrine (pglml) Plasma dopamine (pglml) Plasma arginine vasopressin (pglml) Premature ventricular depolarizations (day-‘) Ventricular tachycardia (events/day)
ANP <125
63 + 2
ANP <125 pglml (n = 51) 64+2
40 11
41 10
26 25 0
17 28 6
22 19 10 0.21 f 0.01 2.6 ? 0.4 23 ? 3 538 ? 45 1092 29 132 ? 21 6.2 ? 0.8 2,004 2 495 1.5 ‘- 0.5
10 20 21 0.18 t 0.01 8.9 + 1.8 43 + 7 858 + 116 124 2 16 169 ? 42 8.2 2 0.9 4,485 ? 715 106.6? 67
P Value NS NS
co.05
co.05
NS co.05 <0.05 co.05 NS NS NS co.05 NS
KM = ischemic cardiomyopathy; LV = left ventricular; NS = not significant; NYHA = New York Heart Association; PDC = primary dilated cardiomyopathy; PVD = primary valvular disease.
JACC Vol. 13. No. 7 June 1989: 1534-9
ATRIAL
300 PC.05
rr
<2
2-5
”
‘5
PLASMA RENIN ACTIVITY ( ng/ml/hr
1
t
c400 400-800 MOO PLASMA NOREPINEPHRINE
( pg/ml)
Figure 2. Plasma concentrations of atria1 natriuretic peptide (ANP) in patients with chronic congestive heart failure grouped according to the level of plasma renin activity and plasma norepinephrine. Patients with the highest levels of both neurohormones had significantly higher concentrations of atrial natriuretic peptide than did patients with normal levels of plasma norepinephrine and plasma renin activity.
nostic information (p = 0.004) as did the functional class (p = 0.003) and the frequency of premature ventricular depolarizations (p = 0.02). Once these variables were entered into the model, other hormonal variables failed to contribute to the prediction of survival.
Discussion Prognostic significance of atria1 natriuretic peptide. Our findings are the first to demonstrate that the measurement of atria1 natriuretic peptide is an important prognostic factor in patients with chronic heart failure. Patients with very high circulating levels of the peptide had a substantially greater mortality than did those with normal or only moderately increased levels of the hormone. Plasma concentrations of atria1 natriuretic peptide provided prognostic information at least as powerful as that provided by the measurement of vasoconstrictor neurohormones. Two factors may explain the close relation between survival and levels qf atria1 natriuretic peptide in our study.
Previous studies (5,6) that have observed a relation between vasoconstrictor hormones and mortality have suggested that neurohormonal activation may contribute directly to the progressive deterioration of left ventricular function that occurs in patients with chronic heart failure. Atria1 natriuretic peptide is not known to exert deleterious actions in these patients; however, its vasodilator and natriuretic properties might be expected to lower ventricular wall stress (16
NATRIURETIC
PEPTIDE
ET AL. GOTTLIEB AND SURVIVAL
1537
18) and its ability to diminish the release of norepinephrine (19,20) and renin (17,21) might reduce the arrhythmogenic potential of these hormonal systems. All of these effects should have a beneficial effect on survival. Consequently, it is likely that the prognostic importance of atria1 natriuretic peptide in chronic heart failure is related to its ability to accurately gauge the severity of chronic heart failure and to reflect the interplay of the multiple factors that contribute to mortality in this disease. Relation of atria1 natriuretic peptide to symptoms. Patients with very high levels of the peptide and the most unfavorable prognosis also had the most limited functional capacity. This observation is consistent with previous reports that circulating levels of the atrial peptide rise in parallel with the severity of symptoms (22.23) and that the functional status of patients with heart failure is an excellent predictor of mortality (24). The survival rate of our patients with very high peptide concentrations was similar to that previously reported in patients with class III and 1V heart failure (24,25). whereas patients with lower levels of the peptide had a mortality rate comparable with that of patients with only mild to moderate symptoms (26). Relation of atria1 natriuretic peptide to cardiac performance. The patients in our study with very high levels of
atrial natriuretic peptide also demonstrated the most abnormal cardiac performance; these patients had a lower cardiac index and higher ventricular filling pressures than did patients with normal or mildly elevated levels of the peptide. Similar findings have been reported by other investigators (8,271.Because patients with severe left ventricular dysfunction have an extremely high mortality rate (38), the unfavorable outcome of patients with markedly increased concentrations of atria1 natriuretic peptide in our study may be related to the association of high levels of the hormone with advanced hemodynamic abnormalities. We could not confirm the very close relation between plasma atrial natriuretic peptide and mean right atria1 pressure (8,27) and left ventricular ejection fraction (29) observed in other investigations. In these previous studies, however. values for right atria1 pressure and ejection fraction varied widely because patients without heart failure were included in the analysis. In contrast, in our study, we evaluated only patients with heart failure. Consequently, there was only a narrow range of values for these two variables, and this may have limited our ability to discern a relation between plasma levels of atria1 natriuretic peptide and either right atrial pressure or ejection fraction. Relation of atria1 natriuretic peptide to vasoconstrictor neurohormones. Patients with the highest concentrations of
atria1 natriuretic peptide in our study also had the highest circulating levels of norepinephrine and renin activity. Such parallel activation of both vasodilator and vasoconstrictor systems has not been previously reported. At first glance, this observation may appear to be difficult to explain because
1538
GOTTLIEB ET AL. ATRIAL NATRIURETIC PEPTIDE AND
JACC Vol. 13, No. 7 June 1989:1534-9
SURVIVAL
a high level of atria1 natriuretic peptide might be expected to suppress the release of renin from the kidney and the release of norepinephrine from nerve terminals (19-21). Consequently, there is an inverse relation between the plasma level of atria1 peptide and plasma renin activity in normal individuals (7); this inverse relation is preserved during the early stages of congestive heart failure (30). In contrast, in our patients with long-standing heart failure, levels of atria1 natriuretic peptide and renin activity were increased in parallel. This direct relation may in part be related to abnormalities of atria1 baroreflex function that are known to occur in chronic heart failure. In normal subjects, atria1 stretch enhances the release of atria1 natriuretic peptide and suppresses adrenergic activity and the release of renin (31). In contrast, in chronic heart failure, despite a similar stimulatory effect on peptide release, atria1 stretch fails to suppress the sympathetic nervous or renin-angiotensin systems (32). Patients with high atria1 pressures may also have coexistent hemodynamic abnormalities that directly trigger the activation of vasoconstrictor systems. Because both plasma norepinephrine and renin activity are powerful prognostic factors in chronic heart failure (5,6,33), parallel changes in the level of atria1 natriuretic peptide in these patients should also predict long-term survival. Relationof atria1natriureticpeptideto ventriculararrhythmias. Our patients with very high levels of atria1 natriuretic peptide had more frequent and complex ventricular arrhythmias than did patients with normal or mildly elevated peptide concentrations. This association may be related to the magnitude of ventricular dilatation because patients with the highest ventricular filling pressures and volumes have both high levels of atria1 peptide and abnormal electrophysiologic responses (34). In addition, because vasoconstrictor and vasodilator hormones were increased in parallel in our study, patients with elevated levels of atria1 peptides also had increased levels of potentially arrhythmogenic neurohormones (for example, norepinephrine and angiotensin) (35,36). Both ventricular arrhythmias and vasoconstrictor neurohormones are powerful predictors of survival in chronic heart failure, although both factors are thought to predict total mortality rather than the occurrence of sudden death (5,37). The lack of an association of these variables with sudden death must be viewed cautiously, however, because it is difficult to determine with confidence the mode of death of patients with chronic heart failure (38). Because of such difficulties, we made no effort to classify the mode of death in our study. Nevertheless, our observations suggest that ventricular arrhythmias may contribute to the high mortality rate of patients with heart failure who have high levels of atria1 natriuretic peptide. Conclusions. The measurement of atria1 natriuretic pepetide is a powerful prognostic factor in patients with chronic heart failure, probably because it reflects the severity of established risk factors for death in patients with advanced
left ventricular dysfunction. Further work is needed to determine whether peptide measurements can be used in the clinical setting to select patients for therapeutic interventions designed to prolong life.
References 1. Levine TB, Francis GS, Goldsmith SR, Simon AB, Cohn JN. Activity of the sympathetic nervous system and the renin-angiotensin system assessed by plasma hormone levels and their relationship to hemodynamic abnormalities in congestive heart failure. Am J Cardiol 1982;49:1659-66. 2.
Thomas JA, Marks BH. Plasma norepinephrine in congestive heart failure. Am J Cardiol 1978;41:23343.
3. Bajusz E, Baker J, Nixon C, Homburger F. Spontaneous hereditary myocardial degeneration and congestive heart failure in a strain of Syrian hamsters. Ann NY Acad Sci 1969;156:105-29. 4. Packer M, Lee WH, Kessler PD, Gottlieb SS, Bernstein JL, Kukin ML. Role of neurohormonal mechanisms in determining survival in patients with severe chronic heart failure. Circulation 1987;75(suppllV):IV-80-92. 5. Cohn JN, Levine TB, Olivari MT, et al. Plasma norepinephrine as a guide to prognosis in patients with chronic congestive heart failure. N Engl J Med 1984;311:819-23. 6. Lee WH, Packer M. Prognostic importance of serum sodium concentration and its modification by converting-enzyme inhibition in patients with severe chronic heart failure. Circulation 1986;73:257-67. 7. Shenker Y, Sider RS, Ostafin EA, Grekin RJ. Plasma levels of immunoreactive atrial natriuretic factor in healthy subjects and in patients with edema. J Clin Invest 1985;76:1684-7. 8. Raine AEG, Erne P, Burgisser E, et al. Atrial natriuretic peptide and atrial pressure in patients with congestive heart failure. N Engl J Med 1986;315:533-7. 9. Dzau VJ, Packer M, Lilly LS, Swartz SL, Hollenberg NK, WilliamsGH. Prostaglandins in severe heart failure: relation to activation of the renin-angiotensin system and hyponatremia. N Engl J Med 1984;310:34752. 10. Burnett JC Jr, Kao PC, Hu DC, et al. Atria1natriuretic peptide elevation in congestive heart failure in the human. Science 1986;231:1145-7. II. Smith S, Anderson S, Ballermann BJ, Brenner BM. Role of atrial natriuretic peptide in adaptation of sodium excretion with reduced renal mass. J Clin Invest 1986;77:1395-8. 12. Packer M, Medina N, Yushak M. Hemodynamic changes mimicking a vasodilator drug response in the absence of drug therapy after right heart catherization in patients with chronic heart failure. Circulation 1985;71: 963-71. 13. Kaplan EL, Meier PL. Nonparametric estimation from incomplete observations. J Am Stat Assoc 1958;53:457-81. 14. Mantel N. Evaluation of survival data and two new rank order statistics arising in its consideration. Cancer Chemother Rep 1966;50:163-70. 15. Breslow NE. Covariance analysis of censored survival data. Biometrics 1979:30:89-99. 16. Ballermann B, Brenner BM. Role of atrial peptides in body fluid homeostasis. Circ Res 1986;58:619-30. 17. Cody RJ, Atlas SA, Laragh JH, et al. Atria1natriuretic factor in normal subjects and heart failure patients: plasma levels and renal, hormonal, and hemodynamic responses to peptide infusion. J Clin Invest 1986;78:136274. 18. Riegger GA, Kromer EP, Kochsiek K. Human atrial natriuretic peptide: plasma levels, hemodynamic, hormonal, and renal effects in patients with severe congestive heart failure. J Cardiovasc Pharmacol 1986;8:1107-12. 19. Nakamaru M, lnagami T. Atrial natriuretic factor inhibits norepinephrine release evoked by sympathetic nervous stimulation in isolated perfused rat mesenteric arteries. Eur J Pharmacol 1986;123:459-61.
JACC Vol. 13, No. 7 June 1989:15349
GOTTLIEB ET AL. ATRIAL NATRIURETIC PEPTIDE AND SURVIVAL
1539
20. Zukowska-Grojec Z. Haass M, Kopin IJ, Zamir N. Interactions of atrial natriuretic peptide with the sympathetic and endocrine systems in the pithed rat. J Pharmacol Exp Ther 1986;239:480-7.
30. Vtllarreal D. Freeman RH, Davis JO, Verburg KM. Vari RC. Atrial natriuretic factor secretion in dogs with experimental high-output heart farlure. Am J Physiol 1987:252:H692-6.
21. Maack T, Marion DN. Camargo MJ, et al. Effect of auriculin (atrial natriuretic factor) on blood pressure, renal function, and the reninaldosterone axis in dogs. Am J Med 1984;77:1069-7.5.
31. Zehr JE. Hasburgen JA. Kurz KD. Reflex suppression of renin secretion during distension of cardiopulmonary receptors in dogs. Clin Res 1976;38: 232-9.
22. Fyhrquist F. Tikkanen 1. Anndiuretic hormone and atrial natriuretic peptide in congestive heart failure. Acta Med Stand 1986:707(suppl):558. 23. Hara H, Ogihara T, Shima J, et al. Plasma atrial natriuretic pepetide level as an index for the severity of congestive heart failure. Clin Cardiol 1987:10:437-42. 24. Wilson JR, Schwartz JS, St. John Sutton M, et al. Prognosis in severe heart failure: relation to hemodynamic measurements and ventricular ectopic activity. J Am Coll Cardiol 1983;2:403-IO. 25. The CONSENSUS Trial Study Group. Effects of enalapril on mortality in severe congestive heart failure: results of the Cooperative North Scandinavian Enalapril Survival Study (CONSENSUS). N Engl J Med 1987:316: 1429-36. 26. Cohn JN, Archibald DG, Ziesche S, et al. Effect of vasodilator therapy on mortality in chronic congestive heart failure: results of a Veterans Administration cooperative study. N Engl J Med 1986;314:1547-52. 27. Richards AM, Cleland JGF, Ton010G, et al. Plasma a natriuretic peptide in cardiac impairment. Br Med J 1986;293:409-12. 28. Unverferth DV. Magorien RD, Moeschberger ML, Baker PB, Fetters JK, Leier CV. Factors influencing the one-year mortality of dilated cardiomyopathy. Am J Cardiol 1984;54:147-52. 29. Schiffrin EL, Taillefer R. Correlation of left ventricular ejection fraction and plasma atrial natriuretic peptide in congestive heart failure (letter). N Engl J Med 1986;315:765-6.
32. Zucker IH, Share L, Gilmore JP. Renal effects of left atrial distension in dogs with chronic congestive heart failure. Am J Physiol 1979:236:H55460. 33. Dargie HJ, Cleland JGF, Leckie BJ, Inglis CG, East BW, Ford I. Relation of arrhythmias and electrolyte abnormalities to survival in patients with severe chronic heart failure. Circulation 1987:75(suppllV):IV-98107. 34. Taggart P, Sutton PMI, Treasure T, et al. Monophasic action potentials at discontinuation of cardiopulmonary bypass: evidence for contractionexcitation feedback in man. Circulation 1988:77:1266-75. 35. Giotti A, Ledda F. Mannaioni PF. Effects of noradrenalin and isoprenaline in combination with alpha- and beta-receptor blocking substances, on the action potential of cardiac Purkinje fibres. J Physiol 1973;229:99-I13. 36. de Langen CDJ. van Gilst WH, de Graeff PA, Kingma JH, Wesseling H. Effects of the renin-angiotensin system on sustained ventricular tachycardie after myocardial infarction (abstr). Circulation 1986:74(supplII): 2-351. 37. Likoff MJ, Chandler SL, Kay HR. Clinical determinants of mortality in chronic congestive heart failure secondary to idiopathic dilated or to ischemic cardiomyopathy. Am J Cardiol 1987;59:63C8. 38. Packer M. Enalapril for congestive heart failure (letter). N Engl J Med 1987:317:1351.
13, No. 7 June 1989:15349
1534
Prognostic Importance of Atria1 Natriuretic Peptide in Patients With Chronic Heart Failure STEPHEN
S. GOTTLIEB,
DEBORAH
AHERN,
MD, FACC,” MARRICK
RN, MILTON
PACKER,
L. KUKIN,
MD, FACC,
MD, FACC
New York, New York
Several circulating neurohormones have been shown to have prognostic significance in patients with chronic heart failure, but the relation between plasma levels of atria1 natriuretic peptide and mortality in this disorder remains unknown. Plasma levels of immunoreactive atria1 natriuretic peptide were measured in 102 patients in whom left ventricular ejection fraction, ventricular arrhythmias on ambulatory electrocardiographic recording and plasma levels of norepinephrine, renin activity, aldosterone and arginine vasopressin were also measured. Compared with patients with atrial natriuretic peptide concentrations below the median value of 125 pg/ml, patients with higher levels of the peptide had a higher plasma renin activity (8.9 -C 1.8 versus 2.6 f 0.4 ng/ml per h) and plasma norepinephrine (858 + 116 versus 538 & 45 pglml), more frequent prema-
As cardiac performance declines in patients with chronic heart failure, various neurohormonal systems are stimulated in an attempt to maintain circulatory homeostasis. This neurohormonal activation becomes progressively more marked as the disease state advances (1,2) and may accelerate the deterioration of left ventricular function that occurs over time (3,4). Either because they are a marker of disease severity or play a direct pathogenetic role, circulating levels of many neurohormones strongly predict the long-term prognosis of patients with heart failure (4). Patients with the most marked activation of the sympathetic nervous system (as reflected by elevated plasma norepinephrine levels) and of
From the Division of Cardiology, Department of Medicine, Mount Sinai School of Medicine of the City University of New York, New York, New York. This study was supported by grants T32 HL 07347and ROl-HL-25055 from the National Heart, Lung, and Blood Institute, National Institutes of Health, Bethesda, Maryland. Dr. Packer is the recipient of a Research Career Development Award (KO4-HL01229)from the National Institutes of Health. Manuscript received August 8.1988; revised manuscript received November 22, 1988,accepted January 9, 1989. *Current address and address for reorints: Stephen S. Gottlieb, MD, Division of Cardiology, University of Maryland School of Medicine, 22 South Greene Street, Baltimore, Maryland 21201. 01989
by the
American College of
Cardiology
ture ventricular depolarizations (4,485 f 715 versus 2,004 f 495/day) and more advanced hemodynamic abnormalities (all p < 0.05). During the subsequent 13 to 25 months of follow-up, patients with high levels of atria1 natriuretic peptide had a significantly lower rate of survival than did those whose initial circulating peptide concentrations were normal or mildly increased (p = 0.01). These data indicate that, in patients with chronic heart failure, plasma atrial natriuretic peptide provides important prognostic information. This may relate to the ability of the hormone to reflect the interplay of several pathophysiologic factors that contribute to mortality in this disease. (.I Am Co11Cardiol1989;13:1534-9)
the renin-angiotensin system (as reflected by hyponatremia) have the highest mortality rate (5,6). These neurohormonal measurements may provide more powerful prognostic information than do the invasive and noninvasive variables that are conventionally used to assess left ventricular function (536). Previous studies that have evaluated the relation between neurohormones and survival in chronic heart failure have focused on the activation of vasoconstrictor systems. Patients with advanced left ventricular dysfunction, however, also have elevated levels of endogenous vasodilators (for example, prostaglandins and atria1 natriuretic peptide) (710); yet, it is not known whether these dilator hormones have prognostic significance. If, like the vasoconstrictor hormones, circulating levels of these vasodilators increase in parallel to the severity of the underlying disease, the release of prostaglandins and atria1 natriuretic peptide might be expected to be greatest in patients with the highest mortality. On the other hand, if the vasodilator hormones act to limit the deleterious effects of the vasoconstrictor systems, high circulating levels of endogenous vasodilators might be associated with a favorable long-term outcome. To distinguish 0735-1097/89/$3.50
JACC Vol. 13, No. 7 June 1989: 1534-9
ATRIAL NATRIURETIC
between these two possibilities, we evaluated the relation between plasma atria1 natriuretic peptide and long-term survival in 102 patients with chronic heart failure.
Methods Study patients. Our study group consisted of 102 patients with chronic congestive heart failure recruited over a 12 month period. There were 81 men and 21 women, aged 34 to 90 years (mean 64). The cause of heart failure was dilated cardiomyopathy in 53 patients, ischemic heart disease in 43 patients and primary mitral or aortic regurgitation, or both, in 6 patients. All patients had a left ventricular ejection fraction ~40% by radionuclide ventriculography. Thirty-two patients were in New York Heart Association functional class II, 39 in class III and 31 in class IV. All patients were clinically stable on constant doses of digoxin and diuretics; any previous therapy with vasodilator drugs was withdrawn for 220 half-lives
Study Design Patients underwent the following evaluations over 2 to 5 days: 1) hormonal measurements (all 102 patients); 2) 24 h ambulatory electrocardiographic (ECG) monitoring (98 patients); and 3) right heart catheterization (34 patients). Hormonal analysis. After the patient had been recumbent in a quiet room for ~30 min, blood was collected from an indwelling catheter for the following analyses: I) plasma renin activity, aldosterone and arginine vasopressin (by radioimmunoassay): 2) plasma norepinephrine, epinephrine and dopamine (by radioenzymatic assay); and 3) plasma atria1 natriuretic peptide (by radioimmunoassay). For the last assay, blood was collected in prechilled ethylenediaminetetraacetic acid (EDTA) tubes with aprotinin, and the plasma was separated immediately and stored at -70” C until extracted and analyzed as previously described (11). Blood was obtained in duplicate (on 2 sequential days) for all assays in 88 of the 102 patients. There was a strong correlation between hormonal determinations performed on the 2 days: for example, atria1 natriuretic peptide on day 1 was closely correlated with atrial natriuretic peptide on day 2 (r = 0.91). Ambulatory electrocardiographic recordings. Twentyfour hour recordings were obtained in all patients within 24 h of hormonal determinations; 48 h recordings were obtained in 88 patients. These recordings were used to measure the frequency of premature ventricular depolarizations and episodes of ventricular tachycardia (23 premature ventricular depolarizations in a row at a rate >lOO/min). Hemodynamic measurements. Right heart catheterization and arterial cannulation were performed to measure right heart and systemic pressures, respectively. Patients were permitted to rest overnight to allow dissipation of the hemo-
GOTTLIEB ET AL. PEPTIDE AND SURVIVAL
1.535
dynamic effects related to intravascular instrumentation. On the next morning, after all cardioactive medications were withheld for 212 h, the following hemodynamic variables were measured repeatedly with the patient in the fasting state until hemodynamic stability was evident: mean arterial pressure, heart rate, left ventricular filling pressure, mean right atria1 pressure and cardiac output. The methods we used have been described in detail previously (12). Mean systemic pressures were determined by electronic filtration. Derived hemodynamic variables were calculated as follow: cardiac index (CI) = CO/body surface area (liters/ min per m’): stroke volume index (SVI) = CUHR (ml/m’); stroke work index (SWI) = SVI x (MAP - LVFP) x 0.0136 (g - m/m’); systemic vascular resistance = 80 X (MAP MRAP)/CO (dynes*s.cm-‘). where CO = cardiac output, MRAP = mean right atrial pressure, LVFP = left ventricular filling pressure, HR = heart rate and MAP = mean arterial pressure. Follow-up. Survival was assessed to the day of death or to June 1, 1988; follow-up ranged from 13 to 25 months. Cumulative survival curves were constructed by KaplanMeier survivorship methods (13), and the differences between survival curves were tested for significance by both Mantel-Cox log-rank and Wilcoxon-Breslow methods (14.15). Statistics. Qualitative and quantitative differences between subgroups of patientswere tested for significance with use of the chi-square statistic and the t test for paired data, respectively. Whenever hormonal and ECG data were obtained in duplicate, the mean value was used for all calculations. All statistical calculations were performed with Statistical Analysis System (SAS) computer programs. Group data are expressed as mean values t SEM.
Results Atria1 natriuretic peptide and survival. Plasma concentrations of atria1 nutriuretic peptide ranged from 11 to 650 pg/ml (median 125) in the 102 patients with chronic heart failure evaluated in this study. Eighty-nine patients had a concentration of the peptide >50 pg/ml (upper limit of normal for the laboratory assay). Patients with a plasma level greater than the median value of 125pgiml had a significantly higher mortality rate than did those with a lower circulating level (p = 0.01 by both Wilcoxon and Mantel-Cox) (Fig. 1). Atria1 natriuretic peptide concentration and severity of heart failure. Patients with a plasma level >125 pgiml had
clinical, neurohormonal and hemodynamic evidence for more severe heart failure than did patients with lower levels of the peptide. Patients with the higher concentrations had greater limitation of functional capacity, higher levels of plasma renin activity, aldosterone and norepinephrine and more severe hemodynamic abnormalities than did patients with lower levels (all p < 0.05) (Tables 1 and 2). Patients with
GOTTLIEB ET AL. ATRIAL NATRIURETIC
1536
JACC Vol. 13, No. 7 June 1989:1534-9
PEPTIDE AND SURVIVAL
Table 2. Hemodynamic Characteristics of 34 Patients Grouped According to the Median Value for Atria1 Natriuretic Pentide* ANP< 125 pg/ml
ANP >125 pgiml (n = 15) ANP> 125 pg/ml
01 0
1
I
I
I
I
I
I
3
6
9
12
15
18
21
I 24
MONTHS
Figure 1. Kaplan-Meier analysis of cumulative rates of survival in patients with heart failure stratified into two groups on the basis of median plasma concentration of atria1 natriuretic peptide (ANP) (125 pg/mU.
higher peptide levels also tended to have a lower left ventricular ejection fraction (p = 0.07) and more frequent
ventricular premature depolarizations (p < 0.05) and episodes of ventricular tachycardia, although the last difference was not statistically significant because of the large variance seen in patients with high peptide levels. When patients were grouped according to circulating levels of vasoconstrictor hormones, plasma levels of atria1 natriuretic peptide were most markedly elevated in patients with the highest levels of plasma renin activity and norepinephrine (Fig. 2).
Cardiac index (litersimin per m*) Stroke work index (g-m/m*) Mean arterial pressure (mm Hg) Left ventricular filling pressure (mm Hg) Mean right atrial pressure (mm Hg) Heart rate (beats/min) Systemic vascular resistance (dynesscm-‘)
2.2 ” 0.1 28 If:4 94 + 6 18 ‘- 3 6.5 2 1.3 87 2 3 1,857 2 140
pg/ml (n = 19)
Value ’
1.7 ? 17 c 82 t 26 +
co.05 co.05 co.05 co.05
0.1 2 3 2
9.4 2 2.0 79 + 3 2,198 ? 141
NS NS NS
*These patients were grouped according to the median value for atrial natriuretic peptide (ANP) for all 102patients studied. NS = not significant.
Prognostic importance of peptide levels. To determine whether the levels of atria1 natriuretic peptide seen in our patients with heart failure had independent prognostic significance or only reflected the importance of other factors, all of the variables that differed between the high and low peptide subgroups (including ejection fraction) were entered into a Cox regression model. Because only 33% of the patients had hemodynamic measurements, these were not entered into the analysis. According to the regression model, plasma atria1 natriuretic peptide provided independent prog-
Table 1. Demographic and Neurohormonal Characteristics of 102 Patients Grouped According to the Median Value for Plasma Atrial Natriuretic Peptide (ANP) ANP >I25 pglml (n = 51) Age (yr) Gender Male Female Cause of heart failure ICM PDC PVD NYHA functional class II III IV LV ejection fraction Plasma renin activity (&ml per h) Plasma aldosterone (ng/dl) Plasma norepinephrine (pg/ml) Plasma epinephrine (pglml) Plasma dopamine (pglml) Plasma arginine vasopressin (pglml) Premature ventricular depolarizations (day-‘) Ventricular tachycardia (events/day)
ANP <125
63 + 2
ANP <125 pglml (n = 51) 64+2
40 11
41 10
26 25 0
17 28 6
22 19 10 0.21 f 0.01 2.6 ? 0.4 23 ? 3 538 ? 45 1092 29 132 ? 21 6.2 ? 0.8 2,004 2 495 1.5 ‘- 0.5
10 20 21 0.18 t 0.01 8.9 + 1.8 43 + 7 858 + 116 124 2 16 169 ? 42 8.2 2 0.9 4,485 ? 715 106.6? 67
P Value NS NS
co.05
co.05
NS co.05 <0.05 co.05 NS NS NS co.05 NS
KM = ischemic cardiomyopathy; LV = left ventricular; NS = not significant; NYHA = New York Heart Association; PDC = primary dilated cardiomyopathy; PVD = primary valvular disease.
JACC Vol. 13. No. 7 June 1989: 1534-9
ATRIAL
300 PC.05
rr
<2
2-5
”
‘5
PLASMA RENIN ACTIVITY ( ng/ml/hr
1
t
c400 400-800 MOO PLASMA NOREPINEPHRINE
( pg/ml)
Figure 2. Plasma concentrations of atria1 natriuretic peptide (ANP) in patients with chronic congestive heart failure grouped according to the level of plasma renin activity and plasma norepinephrine. Patients with the highest levels of both neurohormones had significantly higher concentrations of atrial natriuretic peptide than did patients with normal levels of plasma norepinephrine and plasma renin activity.
nostic information (p = 0.004) as did the functional class (p = 0.003) and the frequency of premature ventricular depolarizations (p = 0.02). Once these variables were entered into the model, other hormonal variables failed to contribute to the prediction of survival.
Discussion Prognostic significance of atria1 natriuretic peptide. Our findings are the first to demonstrate that the measurement of atria1 natriuretic peptide is an important prognostic factor in patients with chronic heart failure. Patients with very high circulating levels of the peptide had a substantially greater mortality than did those with normal or only moderately increased levels of the hormone. Plasma concentrations of atria1 natriuretic peptide provided prognostic information at least as powerful as that provided by the measurement of vasoconstrictor neurohormones. Two factors may explain the close relation between survival and levels qf atria1 natriuretic peptide in our study.
Previous studies (5,6) that have observed a relation between vasoconstrictor hormones and mortality have suggested that neurohormonal activation may contribute directly to the progressive deterioration of left ventricular function that occurs in patients with chronic heart failure. Atria1 natriuretic peptide is not known to exert deleterious actions in these patients; however, its vasodilator and natriuretic properties might be expected to lower ventricular wall stress (16
NATRIURETIC
PEPTIDE
ET AL. GOTTLIEB AND SURVIVAL
1537
18) and its ability to diminish the release of norepinephrine (19,20) and renin (17,21) might reduce the arrhythmogenic potential of these hormonal systems. All of these effects should have a beneficial effect on survival. Consequently, it is likely that the prognostic importance of atria1 natriuretic peptide in chronic heart failure is related to its ability to accurately gauge the severity of chronic heart failure and to reflect the interplay of the multiple factors that contribute to mortality in this disease. Relation of atria1 natriuretic peptide to symptoms. Patients with very high levels of the peptide and the most unfavorable prognosis also had the most limited functional capacity. This observation is consistent with previous reports that circulating levels of the atrial peptide rise in parallel with the severity of symptoms (22.23) and that the functional status of patients with heart failure is an excellent predictor of mortality (24). The survival rate of our patients with very high peptide concentrations was similar to that previously reported in patients with class III and 1V heart failure (24,25). whereas patients with lower levels of the peptide had a mortality rate comparable with that of patients with only mild to moderate symptoms (26). Relation of atria1 natriuretic peptide to cardiac performance. The patients in our study with very high levels of
atrial natriuretic peptide also demonstrated the most abnormal cardiac performance; these patients had a lower cardiac index and higher ventricular filling pressures than did patients with normal or mildly elevated levels of the peptide. Similar findings have been reported by other investigators (8,271.Because patients with severe left ventricular dysfunction have an extremely high mortality rate (38), the unfavorable outcome of patients with markedly increased concentrations of atria1 natriuretic peptide in our study may be related to the association of high levels of the hormone with advanced hemodynamic abnormalities. We could not confirm the very close relation between plasma atrial natriuretic peptide and mean right atria1 pressure (8,27) and left ventricular ejection fraction (29) observed in other investigations. In these previous studies, however. values for right atria1 pressure and ejection fraction varied widely because patients without heart failure were included in the analysis. In contrast, in our study, we evaluated only patients with heart failure. Consequently, there was only a narrow range of values for these two variables, and this may have limited our ability to discern a relation between plasma levels of atria1 natriuretic peptide and either right atrial pressure or ejection fraction. Relation of atria1 natriuretic peptide to vasoconstrictor neurohormones. Patients with the highest concentrations of
atria1 natriuretic peptide in our study also had the highest circulating levels of norepinephrine and renin activity. Such parallel activation of both vasodilator and vasoconstrictor systems has not been previously reported. At first glance, this observation may appear to be difficult to explain because
1538
GOTTLIEB ET AL. ATRIAL NATRIURETIC PEPTIDE AND
JACC Vol. 13, No. 7 June 1989:1534-9
SURVIVAL
a high level of atria1 natriuretic peptide might be expected to suppress the release of renin from the kidney and the release of norepinephrine from nerve terminals (19-21). Consequently, there is an inverse relation between the plasma level of atria1 peptide and plasma renin activity in normal individuals (7); this inverse relation is preserved during the early stages of congestive heart failure (30). In contrast, in our patients with long-standing heart failure, levels of atria1 natriuretic peptide and renin activity were increased in parallel. This direct relation may in part be related to abnormalities of atria1 baroreflex function that are known to occur in chronic heart failure. In normal subjects, atria1 stretch enhances the release of atria1 natriuretic peptide and suppresses adrenergic activity and the release of renin (31). In contrast, in chronic heart failure, despite a similar stimulatory effect on peptide release, atria1 stretch fails to suppress the sympathetic nervous or renin-angiotensin systems (32). Patients with high atria1 pressures may also have coexistent hemodynamic abnormalities that directly trigger the activation of vasoconstrictor systems. Because both plasma norepinephrine and renin activity are powerful prognostic factors in chronic heart failure (5,6,33), parallel changes in the level of atria1 natriuretic peptide in these patients should also predict long-term survival. Relationof atria1natriureticpeptideto ventriculararrhythmias. Our patients with very high levels of atria1 natriuretic peptide had more frequent and complex ventricular arrhythmias than did patients with normal or mildly elevated peptide concentrations. This association may be related to the magnitude of ventricular dilatation because patients with the highest ventricular filling pressures and volumes have both high levels of atria1 peptide and abnormal electrophysiologic responses (34). In addition, because vasoconstrictor and vasodilator hormones were increased in parallel in our study, patients with elevated levels of atria1 peptides also had increased levels of potentially arrhythmogenic neurohormones (for example, norepinephrine and angiotensin) (35,36). Both ventricular arrhythmias and vasoconstrictor neurohormones are powerful predictors of survival in chronic heart failure, although both factors are thought to predict total mortality rather than the occurrence of sudden death (5,37). The lack of an association of these variables with sudden death must be viewed cautiously, however, because it is difficult to determine with confidence the mode of death of patients with chronic heart failure (38). Because of such difficulties, we made no effort to classify the mode of death in our study. Nevertheless, our observations suggest that ventricular arrhythmias may contribute to the high mortality rate of patients with heart failure who have high levels of atria1 natriuretic peptide. Conclusions. The measurement of atria1 natriuretic pepetide is a powerful prognostic factor in patients with chronic heart failure, probably because it reflects the severity of established risk factors for death in patients with advanced
left ventricular dysfunction. Further work is needed to determine whether peptide measurements can be used in the clinical setting to select patients for therapeutic interventions designed to prolong life.
References 1. Levine TB, Francis GS, Goldsmith SR, Simon AB, Cohn JN. Activity of the sympathetic nervous system and the renin-angiotensin system assessed by plasma hormone levels and their relationship to hemodynamic abnormalities in congestive heart failure. Am J Cardiol 1982;49:1659-66. 2.
Thomas JA, Marks BH. Plasma norepinephrine in congestive heart failure. Am J Cardiol 1978;41:23343.
3. Bajusz E, Baker J, Nixon C, Homburger F. Spontaneous hereditary myocardial degeneration and congestive heart failure in a strain of Syrian hamsters. Ann NY Acad Sci 1969;156:105-29. 4. Packer M, Lee WH, Kessler PD, Gottlieb SS, Bernstein JL, Kukin ML. Role of neurohormonal mechanisms in determining survival in patients with severe chronic heart failure. Circulation 1987;75(suppllV):IV-80-92. 5. Cohn JN, Levine TB, Olivari MT, et al. Plasma norepinephrine as a guide to prognosis in patients with chronic congestive heart failure. N Engl J Med 1984;311:819-23. 6. Lee WH, Packer M. Prognostic importance of serum sodium concentration and its modification by converting-enzyme inhibition in patients with severe chronic heart failure. Circulation 1986;73:257-67. 7. Shenker Y, Sider RS, Ostafin EA, Grekin RJ. Plasma levels of immunoreactive atrial natriuretic factor in healthy subjects and in patients with edema. J Clin Invest 1985;76:1684-7. 8. Raine AEG, Erne P, Burgisser E, et al. Atrial natriuretic peptide and atrial pressure in patients with congestive heart failure. N Engl J Med 1986;315:533-7. 9. Dzau VJ, Packer M, Lilly LS, Swartz SL, Hollenberg NK, WilliamsGH. Prostaglandins in severe heart failure: relation to activation of the renin-angiotensin system and hyponatremia. N Engl J Med 1984;310:34752. 10. Burnett JC Jr, Kao PC, Hu DC, et al. Atria1natriuretic peptide elevation in congestive heart failure in the human. Science 1986;231:1145-7. II. Smith S, Anderson S, Ballermann BJ, Brenner BM. Role of atrial natriuretic peptide in adaptation of sodium excretion with reduced renal mass. J Clin Invest 1986;77:1395-8. 12. Packer M, Medina N, Yushak M. Hemodynamic changes mimicking a vasodilator drug response in the absence of drug therapy after right heart catherization in patients with chronic heart failure. Circulation 1985;71: 963-71. 13. Kaplan EL, Meier PL. Nonparametric estimation from incomplete observations. J Am Stat Assoc 1958;53:457-81. 14. Mantel N. Evaluation of survival data and two new rank order statistics arising in its consideration. Cancer Chemother Rep 1966;50:163-70. 15. Breslow NE. Covariance analysis of censored survival data. Biometrics 1979:30:89-99. 16. Ballermann B, Brenner BM. Role of atrial peptides in body fluid homeostasis. Circ Res 1986;58:619-30. 17. Cody RJ, Atlas SA, Laragh JH, et al. Atria1natriuretic factor in normal subjects and heart failure patients: plasma levels and renal, hormonal, and hemodynamic responses to peptide infusion. J Clin Invest 1986;78:136274. 18. Riegger GA, Kromer EP, Kochsiek K. Human atrial natriuretic peptide: plasma levels, hemodynamic, hormonal, and renal effects in patients with severe congestive heart failure. J Cardiovasc Pharmacol 1986;8:1107-12. 19. Nakamaru M, lnagami T. Atrial natriuretic factor inhibits norepinephrine release evoked by sympathetic nervous stimulation in isolated perfused rat mesenteric arteries. Eur J Pharmacol 1986;123:459-61.
JACC Vol. 13, No. 7 June 1989:15349
GOTTLIEB ET AL. ATRIAL NATRIURETIC PEPTIDE AND SURVIVAL
1539
20. Zukowska-Grojec Z. Haass M, Kopin IJ, Zamir N. Interactions of atrial natriuretic peptide with the sympathetic and endocrine systems in the pithed rat. J Pharmacol Exp Ther 1986;239:480-7.
30. Vtllarreal D. Freeman RH, Davis JO, Verburg KM. Vari RC. Atrial natriuretic factor secretion in dogs with experimental high-output heart farlure. Am J Physiol 1987:252:H692-6.
21. Maack T, Marion DN. Camargo MJ, et al. Effect of auriculin (atrial natriuretic factor) on blood pressure, renal function, and the reninaldosterone axis in dogs. Am J Med 1984;77:1069-7.5.
31. Zehr JE. Hasburgen JA. Kurz KD. Reflex suppression of renin secretion during distension of cardiopulmonary receptors in dogs. Clin Res 1976;38: 232-9.
22. Fyhrquist F. Tikkanen 1. Anndiuretic hormone and atrial natriuretic peptide in congestive heart failure. Acta Med Stand 1986:707(suppl):558. 23. Hara H, Ogihara T, Shima J, et al. Plasma atrial natriuretic pepetide level as an index for the severity of congestive heart failure. Clin Cardiol 1987:10:437-42. 24. Wilson JR, Schwartz JS, St. John Sutton M, et al. Prognosis in severe heart failure: relation to hemodynamic measurements and ventricular ectopic activity. J Am Coll Cardiol 1983;2:403-IO. 25. The CONSENSUS Trial Study Group. Effects of enalapril on mortality in severe congestive heart failure: results of the Cooperative North Scandinavian Enalapril Survival Study (CONSENSUS). N Engl J Med 1987:316: 1429-36. 26. Cohn JN, Archibald DG, Ziesche S, et al. Effect of vasodilator therapy on mortality in chronic congestive heart failure: results of a Veterans Administration cooperative study. N Engl J Med 1986;314:1547-52. 27. Richards AM, Cleland JGF, Ton010G, et al. Plasma a natriuretic peptide in cardiac impairment. Br Med J 1986;293:409-12. 28. Unverferth DV. Magorien RD, Moeschberger ML, Baker PB, Fetters JK, Leier CV. Factors influencing the one-year mortality of dilated cardiomyopathy. Am J Cardiol 1984;54:147-52. 29. Schiffrin EL, Taillefer R. Correlation of left ventricular ejection fraction and plasma atrial natriuretic peptide in congestive heart failure (letter). N Engl J Med 1986;315:765-6.
32. Zucker IH, Share L, Gilmore JP. Renal effects of left atrial distension in dogs with chronic congestive heart failure. Am J Physiol 1979:236:H55460. 33. Dargie HJ, Cleland JGF, Leckie BJ, Inglis CG, East BW, Ford I. Relation of arrhythmias and electrolyte abnormalities to survival in patients with severe chronic heart failure. Circulation 1987:75(suppllV):IV-98107. 34. Taggart P, Sutton PMI, Treasure T, et al. Monophasic action potentials at discontinuation of cardiopulmonary bypass: evidence for contractionexcitation feedback in man. Circulation 1988:77:1266-75. 35. Giotti A, Ledda F. Mannaioni PF. Effects of noradrenalin and isoprenaline in combination with alpha- and beta-receptor blocking substances, on the action potential of cardiac Purkinje fibres. J Physiol 1973;229:99-I13. 36. de Langen CDJ. van Gilst WH, de Graeff PA, Kingma JH, Wesseling H. Effects of the renin-angiotensin system on sustained ventricular tachycardie after myocardial infarction (abstr). Circulation 1986:74(supplII): 2-351. 37. Likoff MJ, Chandler SL, Kay HR. Clinical determinants of mortality in chronic congestive heart failure secondary to idiopathic dilated or to ischemic cardiomyopathy. Am J Cardiol 1987;59:63C8. 38. Packer M. Enalapril for congestive heart failure (letter). N Engl J Med 1987:317:1351.