Dipyridamole in the treatment of angina pectoris: A meta-analysis 610 A meta-analysis was performed to reevaluate the efficacy of dipyridamole for prophylaxis of angina pectoris. We found 10 articles that reported 11 randomized control trials published between 1960 and 1970. Three trials found a statistically significant benefit for the drug vs placebo, four showed a positive trend, two found no difference, and two showed a slight trend favoring placebo. When the results of all 11 trials were combined, two different statistical methods showed a statistically significant benefit from the drug. These combined results must be interpreted cautiously because of excluded patients and other methodologic variations in the studies, as well as evidence from other studies that dipytidamole may aggravate angina. Nevertheless, we conclude that there is some evidence for efficacy of the drug and believe the question should be restudied in larger and better-designed trials. (CLIN PHARMACOL THER 1988;43:610-5.) Henry S. Sacks, PhD, MD, V. A. Ancona-Berk, PhD, Jayne Berrier, MA, Raguraman Nagalingam, and Thomas C. Chalmers, MD New York, N. Y. Dipyridamole has been marketed in this country since 1959 and has been used for a variety of problems in- cluding coronary artery disease, myocardial infarction, cerebrovascular disease, prosthetic heart valve replace- ment, and others. The drug has effects on blood vessels (coronary vasodilation) and platelets (it appears to po- tentiate the antiaggregation effect of aspirin). Dipyridamole was studied extensively in the 1950s and 1960s in disease states in which its vascular effects were thought to be potentially beneficial. In recent years the drug has been widely studied as an adjunct to aspirin in the prevention of stroke or myocardial infarction. The antiplatelet activity of dipyridamole and clinical trials of this effect have been reviewed recently.' A number of poorly controlled studies and a few double-blind placebo-controlled trials of dipyridamole for the prevention of attacks of angina pectoris have been reported. Some of these trials found a significant benefit, and some did not. Apparently because of these inconsistent results many authorities concluded that the From the Clinical Trials Unit and the Departments of Medicine, Health Care Management and Biomathematical Sciences, Mount Sinai School of Medicine of The City University of New York. Supported in part by a contract from Chelsea Laboratories, West Hempstead, N.Y., and grant No. 5 ROI LM03116 from the Na- tional Library of Medicine. Received for publication July 21, 1987; accepted Dec. 2, 1987. Reprint requests: Dr. Henry S. Sacks, Clinical Trials Unit, Mount Sinai Medical Center, I Gustave Levy Place, New York, NY 10029. drug's benefit was not proved and it did not become widely accepted for this indication. We have reexam- ined the data on dipyridamole in the treatment of angina pectoris using the techniques of meta-analysis. Meta- analysis is a new research discipline that critically re- views and statistically combines the results of previous research.' It is useful for overcoming sample size de- ficiencies and sharpening effect size estimates. METHODS We conducted an exhaustive search of the literature to identify all randomized control trials of dipyridamole (Persantine) in the treatment or prophylaxis of chronic or stable angina. The following steps were performed: (1) All articles in the English language literature (or with English abstracts) indexed under the key words "dipyridamole" or "Persantine" and "angina pectoris" and "not animal" were retrieved from Medline, the Na- tional Library of Medicine's on-line bibliographic data base. The years covered were from 1966 through April 1987. (2) Our collection of reprints obtained by weekly perusal of Current Contents was searched, and two cardiology textbooks'''' were scanned. (3) The reference lists of all clinical trials and review articles found through the above methods were examined. The methods sections of all clinical trials retrieved were photocopied with identifying data and references to results omitted. These were reviewed independently by two investigators to determine which fulfilled our