Nonsteroidal anti-inflammatory drugs after acute myocardial infarction

I has been shown to play an important role in the development and progression of atherosclerosis.1 Inflammatory markers like C-reactive protein are related to prognosis in acute coronary syndrome.2,3 Inflammation has also been identified as an important factor in the process of atherosclerotic plaque rupture4 that may lead to acute myocardial infarction (AMI) or unstable angina pectoris. Furthermore, epidemiologic studies have found that risk of ischemic events increases with increased leukocyte count.5 Theoretically, anti-inflammatory drugs may alter the prognosis in ischemic heart disease. The results of the Multiple Risk Factor Intervention Trial show that the relative advantage of aspirin in primary prevention of ischemic heart disease is directly related to the baseline concentration of C-reactive protein.6 The effect of aspirin in this setting has been attributed at least partially to its anti-inflammatory properties.6 In an animal model of myocardial infarction, Romson et al7 found that treatment with ibuprofen reduced infarction size and leukocyte infiltration in the infarctrelated areas. However, no clinical study has reported the effect of nonsteroidal anti-inflammatory drugs (NSAIDs) on prognosis after an AMI. We hypothesized that NSAIDs, other than aspirin, may improve the prognosis after AMI, and designed a retrospective study to evaluate this effect based on prospective data from a randomized placebo-controlled clinical trial in the pre-aspirin era. • • • The study included 1,775 patients consecutively randomized in the Danish Verapamil Infarction Trial (DAVIT) II. A detailed description of the organization, design, data acquisition and management, and follow-up in the DAVIT II study has previously been reported.8 In summary, the aim of DAVIT II was to evaluate the effect of long-term treatment with verapamil on mortality and major events (i.e., first reinfarction or death) after an AMI in a multicenter, double-blind, randomized, placebo-controlled study. Patients were included in the second week after AMI. Patients aged ,76 years were eligible for the study. AMI was verified by typical history, electrocardiographic changes compatible with myocardial infarction, and significant elevation in myocardial serum enzymes. Exclusion criteria were treatment with calcium antagonists or b-blocking agents, uncontrolled congestive heart failure, sinoatrial block within the last 3 days before randomization, secondor thirddegree atrioventricular block persisting after the third day of admission, hypotension, other severe diseases, unwillingness to participate, or residence outside the catchment area. Patients were treated with verapamil 120 mg 3 times daily or matching placebo. The mean duration of treatment was 15 months. Eighty-eight patients were receiving treatment with NSAIDs at randomization. Chi-square and Mann-Whitney tests were performed to compare baseline characteristics of these patients with those of 1,687 patients not on treatment with NSAIDs at randomization. The rate of death, reinfarction, and major events (death or first reinfarction) were estimated by the method of KaplanMeier and the differences were evaluated by means of the Tarone-Ware test. A p value ,0.05 was considered significant. Cox multivariate regression analysis was applied to investigate the combined prognostic significance of $2 variables. Patients in the NSAID group were on average 2 years older than controls (62 6 8 vs 60 6 9 years; p 5 0.043), and had a higher prevalence of hypertension (p 5 0.017). There were relatively more women in the NSAID group than in the control group (p 5 0.026). Other baseline characteristics showed no significant differences (Table I). In the NSAID group, during the average 15-month follow-up period, 7 (8.0%) died, 8 (9.1%) had a reinfarction, and 12 (13.6%) had a major event. The corresponding numbers for the controls were 207 (12.3%) deaths, 183 (10.8%) reinfarctions, and 314 (18.6%) major events. There were no significant differences in the rate of mortality (p 5 0.24), reinfarction (p 5 0.63), or major events (p 5 0.27) between the 2 groups. When corrected for age, sex, and hypertension, multivariate models showed a trend in favor of NSAIDs with regard to mortality (hazard ratio 0.59; 95% confidence interval 0.28 to 1.25, p 5 0.17), reinfarction (hazard ratio 0.76; 95% confidence interval 0.37 to 1.55, p 5 0.45), and major cardiac events (hazard ratio 0.67; 95% confidence interval 0.37 to 1.19, p 5 0.17) (Figure 1). However, the effect did not reach the significance level. • • • The present study is the first clinical investigation to indicate that NSAID treatment after an AMI is safe with regard to mortality and reinfarction. Furthermore, there was a nonsignificant positive effect of NSAIDs, implying 41% risk reduction for mortality and 33% for major events. This lack of statistical significance may be attributed to the relative small number of patients in the NSAID group (i.e., a possible type 2 error). From the Department of Cardiovascular Medicine, H:S Bispebjerg Hospital, Bispebjerg Bakke, Copenhagen NV, and UNI-C Olof Palmes Alle 38, Arhus N, Denmark. Dr. Sajadieh’s address is: Department of Cardiovascular Medicine, Bispebjerg Hospital, Bispebjerg Bakke 23, 2400 Copenhagen NV, Denmark. Manuscript received October 14, 1998; revised manuscript received and accepted December 17, 1998.