Predictive value of selected geriatric parameters for postoperative outcomes in older patients with rectal cancer - A multicenter cohort study.

INTRODUCTION:Older patients have a higher risk for complications after rectal cancer surgery. Although screening for geriatric impairments may improve risk prediction in this group, it has not been studied previously. METHODS:We retrospectively investigated patients ≥70 years with elective surgery for non-metastatic rectal cancer between 2014 and 2018 in nine Dutch hospitals. The predictive value of six geriatric parameters in combination with standard preoperative predictors was studied for postoperative complications, delirium, and length of stay (LOS) using logistic regression analyses. The geriatric parameters included the four VMS-questionnaire items pertaining to functional impairment, fall risk, delirium risk, and malnutrition, as well as mobility problems and polypharmacy. Standard predictors included age, sex, body mass index, American Society of Anesthesiologists (ASA)-classification, comorbidities, tumor stage, and neoadjuvant therapy. Changes in model performance were evaluated by comparing Area Under the Curve (AUC) of the regression models with and without geriatric parameters. RESULTS:We included 575 patients (median age 75 years; 32% female). None of the geriatric parameters improved risk prediction for complications or LOS. The addition of delirium risk to the standard preoperative prediction model improved model performance for predicting postoperative delirium (AUC 0.75 vs 0.65, p = 0.03). CONCLUSIONS:Geriatric parameters did not improve risk prediction for postoperative complications or LOS in older patients with rectal cancer. Delirium risk screening using the VMS-questionnaire improved risk prediction for delirium. Older patients undergoing rectal cancer surgery are a pre-selected group with few impairments. Geriatric screening may have additional value earlier in the care pathway before treatment decisions are made.