Biomechanically superior treatments do not translate into improved outcomes in randomized controlled trials. European journal of orthopaedic surgery & traumatology : orthopedie traumatologie Xiong, G. X., Kang, J. R., Sharma, J., Finlay, A., Gardner, M. J., Bishop, J. A. 2021

Abstract

PURPOSE: Significant time and resources are devoted to conducting orthopaedic biomechanics research; however, it is not known how these studies relate to their subsequent clinical studies. The purpose of the present study was to determine whether biomechanically superior treatments were associated with improved clinical outcomes as determined by analogous randomized controlled trials (RCTs).METHODS: A systematic review was conducted to find RCTs that tested a research question based on a prior biomechanical study. PubMed and SCOPUS databases were queried for orthopaedic randomized controlled trials, and full text articles were reviewed to find RCTs which cited biomechanical studies with analogous comparison groups. A random-effects multi-level logistic regression model was conducted examining the association between RCT outcome and biomechanics outcome, adjusting for multiple outcomes nested within study.RESULTS: In total, 20,261 articles were reviewed yielding 21 RCTs citing a total of 43 analogous biomechanical studies. In 7 instances (16.2%), the RCT and a cited biomechanical study showed concordant results (i.e. the superior treatment in the RCT was also the superior construct in the biomechanical study). RCT outcome was not associated with biomechanical outcome (beta=-1.50, standard error=0.78, p=.05).CONCLUSION: This study assessed 21 orthopaedic RCTs with 43 corresponding biomechanical studies and found no association between superior biomechanical properties of a given orthopaedic treatment and improved clinical outcomes. Favourable biomechanical properties alone should not be the primary reason for selecting one treatment over another. Furthermore, RCTs based on biomechanical studies should be carefully designed to maximize the chance of providing clinically relevant insights.

View details for DOI 10.1007/s00590-021-03051-8

View details for PubMedID 34176011