Abstract
Purpose: Failure to adequately recruit patients in orthopaedic trials has often led to early stopping and publication of research findings from smaller sample sizes than originally planned. The purpose of this study is to demonstrate the effect of sample size in a large, clinical trial by using SPRINT trial data to evaluate the results that would have been reported if the trial were stopped at various enrollments.
Method: The SPRINT trial evaluated reamed vs. unreamed nailing in 1226 tibia fractures. We analyzed the re-operation rates after various increments in sample size and compared the early results that would have been reported at smaller enrollments with those seen in the final, adequately powered study.
Results: In the final analysis of 1226 patients, there was a significant reduction in the risk of re-operation with reamed nails for closed fractures (35% reduction; p=0.02) and a trend towards an increased risk of re-operation for open fractures (23% increase; p=0.26). In stark contradiction, the results for the first 50 patients enrolled in the trial revealed a substantial increased risk for reamed nails in closed fractures (risk increase: 165%). It was not until enrollment reached 800 patients that the results reflected the final findings of an advantage for reamed nails. In open fractures, the trend favoring unreamed nails was not seen until 200 patients had been enrolled.
Conclusion: Our findings suggest that stopping the SPRINT trial early would have led to misleading estimates of the treatment effect between reamed and unreamed nails.
Correspondence should be addressed to: COA, 4150 Ste. Catherine St. West Suite 360, Westmount, QC H3Z 2Y5, Canada. Email: meetings@canorth.org
