Author + information
- Received June 26, 1997
- Revision received October 3, 1997
- Accepted October 23, 1997
- Published online February 1, 1998.
- Ayman S Al-Khadra, MDA,
- Deeb N Salem, MD, FACCA,
- William M Rand, PhDA,
- James E Udelson, MD, FACCA,
- John J Smith, MD, PhD, FACCA and
- Marvin A Konstam, MD, FACCA,* ()
- ↵*Dr. Marvin A. Konstam, Box 108, Division of Cardiology, New England Medical Center Hospitals, 750 Washington Street, Boston, Massachusetts 02111.
Objectives. This study sought to evaluate the relation between antiplatelet agent (APA) use and survival and morbidity from cardiac disease in patients with left ventricular (LV) systolic dysfunction.
Background. APAs play an important role in the prevention and treatment of coronary disease. Their effects in patients with LV systolic dysfunction are unknown.
Methods. We reviewed data on APA use in 6,797 patients enrolled in the Studies of Left Ventricular Dysfunction (SOLVD) trial and analyzed the relation between their use and all-cause mortality as well as the combined end point of death or hospital admission for heart failure (HF). We used Cox regression to adjust for differences in baseline characteristics and to test for the interaction between APA use and selected patient variables in relation to outcome.
Results. APA use (46.3% of patients) was associated with significantly reduced mortality from all causes (adjusted hazard ratio [HR] 0.82, 95% confidence interval [CI] 0.73 to 0.92, p = 0.0005) and reduced risk of death or hospital admission for HF (adjusted HR 0.81, 95% CI 0.74 to 0.89, p < 0.0001) but was not influenced by trial assignment, gender, LV ejection fraction, New York Heart Association class or etiology. A strong interaction was observed among APA use, randomization group and all-cause mortality. The association between APA use and survival was not observed in the enalapril group, nor was an enalapril benefit on survival detectable in patients receiving APAs at baseline. However, randomization to enalapril therapy significantly reduced the combined end point of death or hospital admission for HF in APA users.
Conclusions. In patients with LV systolic dysfunction, use of APAs is associated with improved survival and reduced morbidity. This association is retained after adjustment for baseline characteristics. APA use is associated with retained but reduced benefit from enalapril.
Heart failure (HF) is a leading cause of death among patients with cardiovascular diseases. Despite the use of angiotensin-converting enzyme (ACE) inhibitors, the annual mortality rate continues to be as high as 36% . Aspirin has proved to be a beneficial drug for the primary and secondary prevention of coronary events and in the treatment of acute coronary syndromes [2–5]. The antiplatelet effects of this agent are thought to mediate its efficacy in preventing acute coronary thrombosis and possibly prevention of progression of atherosclerosis [6, 7]. The effect of aspirin therapy in patients with HF of various etiologies has not been adequately tested. Of the studies that evaluated aspirin use after myocardial infarction (MI), only the Aspirin in Myocardial Infarction Study (AMIS) and Persantine Aspirin Reinfarction Study II trials reported the influence of baseline characteristics on outcome, and no conclusions can be drawn regarding the use of aspirin in patients with chronic HF or nonischemic left ventricular (LV) systolic dysfunction, or both.
Although aspirin therapy may prevent further ischemic insults in patients with ischemic cardiomyopathy, its effects on prostaglandins may adversely affect hemodynamic function. Aspirin antagonizes the beneficial effects of antihypertensive therapy [10–12], and may reduce systemic vasodilatory reserve. Aspirin also decreases glomerular filtration pressure, reducing sodium and water excretion and blunting the effect of diuretic drugs .
To explore the effect of aspirin and other antiplatelet agents (APAs) on outcome in patients with HF, we analyzed data from the Studies of Left Ventricular Dysfunction (SOLVD) concerning use of APAs, particularly the relation between use of these agents and survival.
SOLVD was a multicenter, randomized, double-blind, placebo-controlled investigation that studied the effect of enalapril on morbidity and mortality in patients with LV systolic dysfunction, defined as LV ejection fraction (LVEF) ≤0.35 [14, 15]. Patients receiving treatment other than ACE inhibitors for symptoms of HF were enrolled in the treatment trial; patients without such symptoms were enrolled in the prevention trial . Patients were ineligible if they were >80 years of age or had any of the following: hemodynamically serious valvular heart disease requiring operation, unstable angina pectoris, angina thought to be severe enough to require revascularization procedures, MI during the previous month, severe pulmonary disease, serum creatinine levels >177 μmol/liter (2 mg/dl) or any other disease that might substantially shorten survival or impede participation in a long-term trial. Ventricular function was assessed by contrast angiography, radionuclide ventriculography or two-dimensional echocardiography. The baseline study visit included a comprehensive review of medical history, and patients were specifically asked whether they were taking certain medications regularly, including APAs and those shown in Table 1. After randomization to enalapril or placebo therapy, patients were regularly evaluated, and their clinical status, including the development of HF, hospital admissions, adherence to study regimen and adverse effects, was recorded at each follow-up visit. For patients who died or were admitted to the hospital, the cause of death or the primary reason for hospital admission was evaluated and classified by an investigator, who had no knowledge of the study treatment.
We reviewed the data on APA use, including aspirin and dipyridamole, at baseline. All patients in both trials with complete data were included in the analysis. The end points of all-cause mortality, death or hospital admission for any cardiac event (HF, acute coronary syndromes or arrhythmia), death or hospital admission for HF, cardiovascular death, sudden death without antecedent worsening HF, death due to HF and fatal MI were analyzed.
1.2 Statistical Analysis
The primary outcome measures were death and length of survival (time from entry into the study until death, end of study or loss to follow-up). Risk factors of primary interest were APA use randomization to enalapril therapy and trial (treatment or prevention). The chi-square test for independence was used to analyze unadjusted mortality rates and to calculate relative risk (RR) and 95% confidence interval (CI). A p value < 0.05 was considered significant. Kaplan-Meier survival analysis was used to examine duration of survival, with log rank tests to compare survival analysis curves between subsets of patients.
Cox regression was used to adjust for potential influence of confounding factors on survival time and included age, gender, etiology, trial assignment, study drug randomization, LVEF, New York Heart Association functional class, history of angina, history of atrial fibrillation, history of cerebrovascular disease and baseline drug therapy with warfarin, digitalis or diuretic drugs. Results are reported as adjusted hazard ratio (HR) with 95% CI. The computer statistical software package SPSS (version 7.5) was used for all analyses.
Of 6,797 patients enrolled (2,569 in the treatment trial, 4,228 in the prevention trial), data from 6,512 were considered in the final analysis; 285 patients (4.2%) were excluded because of missing baseline information. The mean follow-up period was 41.4 months in the treatment trial and 37.4 months in the prevention trial. The clinical characteristics of APA users and nonusers in the combined trial are shown in Table 1. Of the total population, 46.3% of patients reported using APA and 53.7% did not use APAs at the time of randomization. There was no significant difference in the mean age between the two groups. The proportion of male patients was higher in the APA users group. Patients taking APAs generally had a higher LVEF and were in a lower mean functional class, and most had ischemic heart disease. They were less likely to have atrial fibrillation or to be using antiarrhythmic agents, anticoagulant agents, digitalis or diuretic drugs. Most (>95%) patients receiving APAs were taking aspirin.
2.2 All-Cause Mortality
By the end of the study, in the combined trial, there were 548 deaths in the APA users group (18.2%) compared with 997 in the APA nonusers group (28.5%) (RR 0.64, 95% CI 0.58 to 0.70 p < 0.0001). Mortality risk reduction in APA users was present in both the treatment trial (RR 0.83, 95% CI 0.74 to 0.93, p = 0.0001) and the prevention trial (RR 0.68, 95% CI 0.59 to 0.79, p < 0.0001). Mean survival time in the combined trial was 1,590 days in APA users versus 1,470 days in nonusers (p < 0.0001) (treatment trial: 1,282 days vs. 1,208 days, p = 0.008; prevention trial: 1,673 days vs. 1,614 days, p = 0.0002).
After adjusting for confounding factors, APA use emerged as a significant predictor of favorable outcome (Fig. 1). In the combined trial, HR for all-cause mortality was 0.82 (95% CI 0.73 to 0.92, p = 0.0006) in users compared with nonusers (Table 2). This benefit was seen in the treatment and prevention trials when analyzed separately (Table 3).
2.3 Causes of Cardiac Death
Of 1,545 deaths in the combined trial population, 1,357 were due to cardiovascular causes (484 in the APA users group vs. 873 in the nonusers group; RR 0.64, 95% CI 0.58 to 0.71, p < 0.0001), of which there were 1,273 cardiac deaths (447 in the APA users group vs. 826 in the nonusers group; RR 0.63, 95% CI 0.56 to 0.70, p < 0.0001). All three modes of cardiac death (sudden death not associated with worsening HF, death associated with worsening HF, fatal MI) were reduced.
After adjusting for confounding variables, APA use was no longer associated with a reduction in death associated with worsening of HF but remained associated with a reduction in the other causes of cardiac death (Table 2). Noncardiac vascular deaths, including fatal strokes and pulmonary embolism, were not different between groups. APA use was associated with a similar reduction in the risk of cardiac death in both trials when analyzed separately.
2.4 Death or Hospital Admission for Cardiac Events
In the combined group, 42.6% of APA users experienced death or hospital admission for any cardiac event, and 26.4% experienced death or hospital admission for HF, compared with 54.5% and 40.6% in the nonusers group, respectively (p < 0.0001 for both). The mean time to death or hospital admission for any cardiac event was 1,187 days for APA users versus 1,053 days for nonusers (p < 0.0001), and the mean time to death or hospital admission for HF was 1,455 days for APA users versus 1,271 days for APA nonusers (p < 0.0001).
Tables 2 and 3show the HRs for the combined, treatment and prevention trials. After adjusting for confounding variables, APA use was associated with a statistically significant reduction in combined end points of death or hospital admission for any cardiac event and death or hospital admission for HF (Fig. 2).
2.5 Effect of APA Use in Different Subgroups
The relation between APA use and the various end points was analyzed by age, gender, etiology, functional class, LVEF and randomization to enalapril therapy. Interaction terms were introduced into the Cox regression analysis, and adjustment for confounding variables was performed. The presence of a significant interaction (p < 0.05) was considered to indicate dependence of the association between APA use and survival on the presence or variation in one of these variables. Gender, etiology and functional class had no influence on the relation between APA use and any of the end points (Table 4). No such influence was seen when both trials were analyzed separately. Despite the absence of a statistically significant interaction with etiology, a significant relation between APA use and survival was seen only in the ischemic group after adjustment for confounding variables. In the nonischemic group, there was a nonsignificant trend toward improved survival among patients receiving APA (p = 0.3). However, the number of patients was small (n = 1,385), and the frequency of APA use was lower in this group (only 19.1% reported using APAs). The nonischemic group, therefore, lacked sufficient power to detect a statistically significant difference between APA users and nonusers.
There was a significant interaction among APA use, age and all-cause mortality (p = 0.02). This interaction was present within each trial, separately and after adjusting for confounding variables. The survival benefit associated with APA use was lower in magnitude with increasing patient age.
2.6 Interaction Between APA and Enalapril Therapy
When the APA–enalapril interaction term was tested in the Cox regression model in the combined trial, it was found to be a significant predictor of all-cause mortality (p = 0.0005). This interaction was explored further by analyzing mortality rates by APA use and randomization to enalapril therapy, with adjustment for baseline differences.
Tables 5 and 6⇓⇓show the HRs (with 95% CIs) for the combined population. Table 5shows the absence of a relation between APA use and survival among patients randomized to enalapril therapy. In contrast, Table 6shows that in patients with baseline use of APA, a significant survival benefit of enalapril could not be identified, although enalapril-induced reduction in the combined end point of death or hospital admission for HF persisted. Use of enalapril in these patients was associated with an HR of 0.81 for this end point, which was statistically significant.
When each trial was analyzed separately, significant interaction among APA use, randomization group, and all-cause mortality was present within the treatment trial only. Within the prevention trial, there was no statistically significant interaction among APA use, randomization group and any of the clinical end points.
To our knowledge, the present analysis is the first to identify an association between APA use and improved survival in patients with LV systolic dysfunction. Users of APA in the SOLVD trial had lower all-cause mortality and were more likely to be free from death or hospital admission for any cardiac event or HF. The difference between users and nonusers was statistically significant, even after adjusting for differences in baseline characteristics between the two groups. Findings were consistent in both the treatment and prevention trials. Reduction in all-cause mortality was primarily derived from a reduction in sudden death not preceded by worsening HF and a reduction in fatal MI. There was no evidence of beneficial association between APA use and death associated with worsening HF or noncardiac vascular deaths in the Cox regression model. The results of our analysis apply to patients with chronic LV dysfunction, regardless of their symptom status, as shown by the lack of statistical interaction between APA use and functional class or trial assignment for all end points and by consistency of the findings when the prevention and treatment trials were analyzed separately.
3.1 Effect of Other Baseline Variables
The beneficial association between APA use and primary end points of this analysis were present when male and female patients were analyzed separately, and there was no interaction with LVEF. However, this association was lower in magnitude with advancing age for both end points (all-cause mortality and death or hospital admission for HF, p = 0.02 and 0.002, respectively). The influence of age is probably multifactorial and may be related to the presence of concomitant illnesses. Although there was no statistical interaction between etiology and the benefit associated with APA use, when analyzed separately, the benefit in patients with nonischemic etiology only reached statistical significance for the end point of death or hospital admission for HF (p = 0.04). There are two possible explanations for this difference: 1) It may represent a true finding reflecting the lack of beneficial association in this group of patients; 2) because the number of patients in this population was small, sufficient power to detect a statistically significant difference between users and nonusers of APA was lacking.
3.2 Possible Mechanisms of the Associated Benefit and Role of Prostaglandins
Platelets, platelet products and thrombosis are known to play a critical role in the pathogenesis of cardiovascular disease. APAs, which are known to be beneficial in the prevention and treatment of acute coronary syndromes [2–4], inhibit the synthesis of prostaglandin G2and, consequently, prostaglandin endoperoxides and thromboxane A2. Prothrombotic platelet responses are thus attenuated . The role of cyclooxygenase products in the pathophysiology of HF has not been fully elucidated. Inhibition of platelet function may be beneficial in patients with LV dysfunction, preventing further episodes of myocardial ischemia and loss of myocardium. APA use may therefore retard the progression of HF and prolong survival. Earlier reports have documented the presence of high levels of circulating vasodilatory prostaglandins (prostacyclin [prostaglandin I2] and prostaglandin E2) in patients with severe chronic HF, and hemodynamic indexes in these patients worsened with the administration of indomethacin . Upregulation of prostaglandin synthesis may be an important compensatory mechanism that counteracts various mediators of vasoconstriction in patients with HF. By interfering with prostaglandin production, aspirin and other cyclooxygenase inhibitors may exert harmful effects, compounded by adverse effects on renal hemodynamic function , thus decreasing responsiveness to diuretic agents.
Alternatively, more recent studies have provided evidence that although prostaglandins are upregulated, the balance is tilted toward the production of thromboxanes rather than prostacyclin. In addition to causing platelet activation and aggregation, thromboxane A2directly causes vasoconstriction and is thought to mediate, at least in part, the vasoconstricting effect of angiotensin II . In this case, selective inhibition of thromboxane production may be beneficial. Baur et al. found in a group of patients with class II and III ischemic HF and LVEF <40% that the ratio of urinary excretion of prostacyclin metabolites to urinary excretion of thromboxane A2metabolites was low. This change is consistent with a shift toward production of vasoconstricting as opposed to vasodilating prostaglandins. This ratio was increased by the administration of a single dose of 250 mg of aspirin. This relative increase in vasodilating prostaglandins may improve vascular function and response to vasodilators. Evidence for the latter was shown by a recent study in patients with class III HF , where the administration of 100 mg of aspirin significantly improved sodium nitroprusside-induced increase in radial artery diameter and forearm blood flow.
APA may have other effects that are important in the setting of myocardial injury and healing, including reduction of fibrocellular response and remodeling in the noninfarcted myocardium , and inhibition of inducible nitric oxide synthase production , which is augmented in patients with HF [24–26].
3.3 APA–enalapril Interaction
Within treatment trial patients we observed a significant interaction among APA use, randomization to enalapril and all-cause mortality. In patients receiving APA, we were unable to document an influence of enalapril on survival, although a beneficial effect on the combined end point of death or hospital admission for HF was present. There are three possible explanations for this observation: 1) It could have occurred by chance. This possibility appears unlikely given the statistical strength of the observation. 2) APAs antagonize the effect of enalapril. This antagonism was demonstrated in hemodynamic studies showing that aspirin may reverse the vasodilator effects of enalapril . However, these studies used large doses of aspirin, and other studies failed to show similar findings [28, 29]. In addition, when the levels of neurohormones were measured in the study by Hall et al. , there was no evidence of reversal of changes induced by ACE inhibitor therapy. 3) APA and enalapril may have overlapping mechanisms of action, detracting from benefit when both agents are used together. Some of the benefit of ACE inhibitors may be mediated through their effects on prostaglandin metabolism. Enalapril and other ACE inhibitors inhibit the degradation of bradykinins, which are known to enhance local production of vasodilator prostaglandins in animals and humans [30, 31]. This is similar to the effect of low doses of aspirin. The study by Baur et al. supports this mechanism of interaction. Administration of aspirin in patients receiving enalapril did not change plasma-converting enzyme activity, plasma angiotensin II or norepinephrine concentrations. The ratio of urinary prostacyclin metabolites to thromboxane metabolites improved after enalapril treatment and showed further improvement with subsequent addition of salicylates. Furthermore, the coadministration of aspirin in the last 4 weeks did not reverse the improvements induced by enalapril therapy in blood pressure, plasma neurohormones, excretion of prostaglandin metabolites, LV mass index, LVEF, maximal aerobic capacity and maximal oxygen consumption. The study by Jeserich et al. showed that administration of either aspirin or perindopril was associated with improvements in vascular function. However, no additional benefit was seen when perindopril was given to patients who were pretreated with aspirin.
Finally, a combination of partial agonism and antagonism between APAs and angiotensin-converting enzyme inhibitors can explain the complex relation between these agents and would be most consistent with our data. It is likely that APAs and ACE inhibitors interact at multiple levels with various degrees of agonism and antagonism, depending on their relative concentration and on the existing balance between vasodilating and vasoconstricting prostaglandins.
3.4 Strengths and Limitations
Our analysis has important strengths, including large sample size, prospective definition of end points and collection of data, long follow-up time, availability of data for most patients (>95%) and consistency of the results in different subgroups. However, it has limitations common to cohort studies, including its retrospective nature and lack of randomization. Our analysis also lacked data on the dosage and consistency of APA use throughout the study period.
The use of APA in patients with LV systolic dysfunction is associated with improved survival and reduced morbidity. This association is not altered by gender, etiology, functional class or LVEF. It is reduced with advancing age and by enalapril administration. APA use identifies a group of patients with retained but reduced benefit from enalapril. Further investigation is warranted to explore the basis for the observed benefit associated with APA use and for the interaction between APA and ACE inhibitors.
☆ Support for this study was provided through the General Clinical Research Center, funded by the Division of Research Resources of the National Institutes of Health, Bethesda, Maryland under Grant MO1-RR00054. The Studies of Left Ventricular Dysfunction (SOLVD) trial was funded by the National Heart, Lung, and Blood Institute, National Institutes of Health, with a contribution from Merck Sharpe and Dohme Pharmaceutical Company, West Point, Pennsylvania.
This work was presented in part at the 68th Annual Scientific Sessions of the American Heart Association, Anaheim, California, November 1995.
- angiotensin converting enzyme
- antiplatelet agent
- confidence interval
- heart failure
- hazard ratio
- left ventricular
- left ventricular ejection fraction
- myocardial infarction
- relative risk
- Studies of Left Ventricular Dysfunction
- Received June 26, 1997.
- Revision received October 3, 1997.
- Accepted October 23, 1997.
- The American College of Cardiology
- Hennekens CH,
- Buring JE,
- Sandercock P,
- Collins R,
- Peto R
- Klimt CR,
- Kratterud GL,
- Stamler J,
- Meier P
- Oates JA
- Stein B,
- Fuster V
- Lin L,
- Nasjletti A
- Baur LHB,
- Schipperheyn JJ,
- Van der Laarse A,
- et al.
- Farivar RS,
- Chobanian AV,
- Brecher P
- Haywood GA,
- Tsao PS,
- von der Leyen HE,
- et al.
- Hall D,
- Zeitler H,
- Rudolph W
- Evans MA,
- Burnett JC Jr.,
- Redfield MM
- McGiff R,
- Terragno N,
- Malik KU,
- Longro AJ