| Abstract: | While percutaneous transluminal coronary angioplasty (PTCA) as a primary modality for treating acute myocardial infarction (MI) has been shown to have important advantages over thrombolysis, a survival benefit has not been demonstrated because of the small size of the individual trials. To increase the statistical power to detect a survival benefit, we performed a meta-analysis of trials of PTCA and thrombolysis. We pooled the data for all randomized, controlled trials; randomized, controlled trials stratified according to thrombolytic agent [streptokinase vs. tissue plasminogen activator (TPA)]; and all trials. Pooling was performed by calculating the Mantel-Haenszel odds ratio with the Robins, Greenland, and Breslow estimate of variance. Calculation of the Q statistic was performed to assess heterogeneity. For all four analyses, the odds ratio indicated a significant survival advantage of PTCA over thrombolysis: all randomized controlled trials [0.57,95% confidence index (CI): 0.48,0.68)]; streptokinase trials [0.61,95% CI: 0.43,0.87); TPA trials (0.52,95% CI: 0.36,0.76); all trials (0.51,95% CI: 0.43,0.61). The Q statistic was not significant for any of the analyses. The results of our meta-analysis support the hypothesis that PTCA is associated with a significant reduction in mortality compared with thrombolysis. |
