Search

We replicated a randomized clinical trial (RCT) that investigated the effectiveness of a beta-blocker therapy in reducing cardiac events among 400 patients undergoing major non-cardiac surgeries to determine if observational studies can achieve comparable results to RCTs. From the electronic medical records of Cleveland Clinic using the same inclusion criteria as the RCT, 200 cases were matched to 200 controls by risk set. Our study achieved comparable results as the RCT: no significant differences in cardiac events existed between cases and controls based on t-tests and chi-square tests, although different predictors were identified by logistic regression (age, myocardial infarction, and valvular heart disease in our study vs. systolic blood pressure and hypotension in the RCT). The effectiveness of the therapy was then investigated on a larger sample (358 cases vs. 1614 controls) matched by propensity scores. The short-term postoperative survival was essentially the same between cases and controls based on a Cox proportional hazard model.