Minimal clinically important difference in postoperative recovery among patients with gastrointestinal cancer.

PURPOSE The minimal clinically important difference (MCID) based on patient-reported outcomes is the smallest outcome change sufficiently significant to influence management and is crucial to the design and interpretation of comparative effectiveness trials. The purpose of this study was to estimate the MCID for postoperative recovery metrics in gastrointestinal cancer patients. METHODS This was a three-institutional cohort study. Participants were 219 patients scheduled for gastrointestinal cancer elective surgery. Body mass index (BMI), isometric knee extension torque (IKET), 6-min walk test (6 MWT), and Short-Form 36-Item Health Survey (SF-36) version 2 were evaluated 1-2 days prior to surgery (baseline) and 4 weeks after surgery. Patients received postoperative rehabilitative care from a physical therapist during hospitalization. The MCID used anchor-based methods. The anchor was a score on the SF-36 physical functioning subscale greater or lower than the average score of the general Japanese population. RESULTS The receiver operating curve indicated a cutoff value on the 6 MWT of -7.8 m for clinically relevant decline (area under curve [AUC] = 0.67, 95% confidence interval [CI] = 0.599-0.741) or a 1.5% change. The cutoff value on the SF-36 role-physical subscale was -34.4 for clinically relevant decline (AUC = 0.691, 95% CI = 0.621-0.761) or a 36.6% decrease. No significant correlation was found between changes in BMI, IKET, and anchor. CONCLUSION Plausible MCIDs are present in patients with gastrointestinal cancer. These values can assist the interpretation of clinical trials and observation of the postoperative clinical course of gastrointestinal cancer surgery.