Author’s Reply: Prehabilitation Before Major Abdominal Surgery

Abstract

We thank Chirico and colleagues for their letter and highlighting that Trial Sequential Analysis (TSA) was not performed in the meta-analysis assessing prehabilitation prior to major abdominal surgery [1]. Following TSA analysis, Chirico and colleagues suggest that, contrary to the findings of the article, the prehabilitation effect on postoperative morbidity is inconclusive and that further studies are required to evaluate this. We agree that the use of TSA can further validate the results of meta-analyses and provide a statistical context on which to base findings, and has been used increasingly in recent years with important clinical implication [2]. The analysis for our study was performed based on the best available evidence at the time to answer a topical question. However, we accept that the use of TSA could have added to the interpretation of this result with the knowledge of a potentially inadequate sample size. However, we dispute that further studies are required to assess the effect of prehabilitation versus standard care. Prehabilitation is well practiced in a number of countries and surgical specialties [3]. To conduct a new trial comparing prehabilitation versus no prehabilitation, i.e. standard of care without advice on pre-operative exercise, smoking cessation or dietary advice would be, we consider, sub-optimal, despite the TSA findings. This concept likely explains the heterogeneity, and subsequent apparent reduced effect, of the prehabilitation groups in the included trials, detected by TSA. A number of the included trials incorporated a control group protocol comprising prehabilitation elements (e.g. non-supervised exercise recommendations, dietary supplementation and breathing exercises) [4–6], thus potentially impacting the effect size between the two groups. To deny trial participants, some of the more established practices of pre-operative optimisation to provide a cleaner control group could be considered inappropriate. In conclusion, we agree with the authors regarding the usefulness of TSA in meta-analysis assessment and the potential impact on the primary outcome of this study as a result of TSA. However, this analysis, like meta-analysis and all statistical measures, only tells a part of the story. Because a number of prehabilitation care factors have become established practice, we disagree that further trials are required to determine the benefit of prehabilitation versus a non-prehabilitation ‘‘standard care’’ pathway to patients before undergoing major surgery. We advocate that future studies evaluate the optimal prehabilitation pathway makeup, to determine best practice.