To test the hypothesis that the result of steroid injection in the carpal tunnel provides a better predictor of the outcome of later surgery. We also explored other possible factors that might predict the outcome directly or interact with the results of steroid injection to better predict the outcome.We performed a historical cohort study on 57 patients who had carpal tunnel release. Care was taken to avoid problems of statistical nonindependence caused by both hands being studied and confounding from previous surgeries.We found a large and significant difference in the success rate of surgery for patients who had obtained some relief from injection (87%) versus those who had not (54%). No other significant predictor was found. We discovered factors that may interact with the results of injection in predicting the outcome of surgery (eg, Katz and Stirrat hand diagram assessment of the probability of carpal tunnel syndrome) although not significant in our study.Some relief from steroid injection is the best predictor for success of surgery. Further study is warranted to identify factors that interact with this predictor.