Health Outcomes Associated with Beta-Blocker and Diltiazem Treatment of Unstable Angina


Synopsis: Diltiazem demonstrated no survival benefit compared to beta-blockers in unstable angina.

Source: Smith NL, et al. J Am Coll Cardiol 1998; 32:1305-1311.

Therapy for unstable angina was explored in a retrospective outcome study from the Seattle VA hospital. This analysis sought to assess outcomes in unstable angina in men discharged with this diagnosis who were treated with a beta-blocker or diltiazem. Eligible veterans discharged from the Seattle VA Healthcare System between 1989 and 1995 were identified by chart review. Discharge medications and clinical outcomes were documented, including rehospitalization and revascularization. Complex statistical analyses were used, including a multivariate model that "adjusted for potential confounders using propensities score matching." From an initial cohort of almost 1500 individuals, 247 were identified who had accurate pharmacy records concerning drug use within 30 days of discharge, identifying those on a beta-blocker or diltiazem monotherapy. The characteristics and demographics of these individuals were typical for a VA population with unstable angina. Two-thirds were on either a beta-blocker or a calcium blocker on admission, and most were discharged on similar medications. Diltiazem use was threefold more common; those on this drug were more likely to have had prior heart failure, COPD, and higher blood pressures. In outcomes analysis, an unadjusted proportionate hazards regression model, diltiazem was associated with a 60% increase in all-cause mortality but, after adjustment for potential confounders, there was no longer any difference in risk for subsequent death between patients taking diltiazem or a beta-blocker. The risk ratio was also similar when the analysis was restricted to men without lung disease or heart failure. An analysis of veterans who were Washington residents was comparable, showing a nonstatistical increased risk ratio for diltiazem, even after adjustments. Smith and colleagues conclude that "Veterans treated with diltiazem for unstable angina … had no overall survival advantage compared with … beta blockers." The composite end point of rehospitalization or death was associated with a nonsignificantly elevated risk with diltiazem; however, confidence intervals were wide. The data related to these drugs in unstable angina were inconclusive with respect to the question of calcium blockers vs. beta-blockers. Multiple study limitations are discussed, most important, that this was not a prospective randomized trial. Diltiazem users at baseline were less healthy. Patients who were on polypharmacy for myocardial ischemia were not included in the analysis. Finally, the use of beta-blocker monotherapy was low and substantially less than diltiazem. Smith et al conclude that beta-blockers were chronically underused and that diltiazem demonstrated no clinical benefits following hospitalization in unstable angina.

Comment by Jonathan Abrams, MD

This report is of interest but is not terribly useful to clinicians. Underuse of beta-blockers is not a new subject, particularly in a cohort that has significant co-morbidity of asthma and COPD. In that there was no placebo group, it is impossible to know whether diltiazem or a beta-blocker was beneficial during the average follow-up (51 months). While Smith et al conclude that diltiazem imparted no survival benefit, another interpretation would be that diltiazem was as good as beta-blocker treatment, which is the case after the propensity score adjustment. The statistical methods used are complex and it is unclear whether the propensity score analysis solved the problem of nonequality between the cohorts in this retrospective analysis. Clearly, data from a variety of sources indicate that for individuals with preserved left ventricular function, diltiazem is safe and, at least in the post-non-Q myocardial infarction population, is beneficial when compared to placebo. These data indicate that diltiazem is safe in unstable angina. Prior experience confirms that this drug should not be used in individuals with overt heart failure when associated with a depressed LV ejection fraction (< 40%). The current interest in beta-blockers for heart failure may lower the threshold to use these drugs in patients with depressed LV function and unstable angina; overall, beta-blocker use remains inadequate for coronary disease, as seen in this database as well as recent major Medicare cohort analyses. I would conclude that either drug, if appropriately selected, can be used safely for unstable angina patients. Physicians should pay attention to the status of LV function as well as other co-morbidity in making a choice of therapy.