Functional Compromise Reflected by Sarcopenia, Frailty, and Nutritional Depletion Predicts Adverse Postoperative Outcome After Colorectal Cancer Surgery

Abstract

To determine the association of sarcopenia with postoperative morbidity and mortality after colorectal surgery. Functional compromise in elderly colorectal surgical patients is considered as a significant factor of impaired postoperative recovery. Therefore, the predictive value of preoperative functional compromise assessment was investigated. Sarcopenia is a hallmark of functional compromise. A total of 310 consecutive patients who underwent oncologic colorectal surgery were included in a prospective digital database. Sarcopenia was assessed using the L3 muscle index utilizing Osirix on preoperative computed tomography. Groningen Frailty Indicator and Short Nutritional Assessment Questionnaire scores were used to assess frailty and nutritional compromise. Predictors for anastomotic leakage, sepsis, and mortality were analyzed by logistic regression analysis. Age was an independent predictor of mortality [P = 0.04; odds ratio, 1.17; 95% confidence interval (CI), 1.01-1.37]. Thirty-day/in-hospital mortality rate in sarcopenic patients was 8.8% versus 0.7% in nonsarcopenic patients (P = 0.001; odds ratio, 15.5; 95% CI, 2.00-120). Sarcopenia was not predictive for anastomotic leakage or sepsis. Combination of high Short Nutritional Assessment Questionnaire score, high Groningen Frailty Indicator score, and sarcopenia strongly predicted sepsis (P = 0.001; odds ratio, 25.1; 95% CI, 5.11-123), sensitivity, 46%; specificity, 97%; positive likelihood ratio, 13 (95% CI, 4.4-38); negative likelihood ratio, 0.57 (95% CI, 0.33-0.97). Functional compromise in colorectal cancer surgery is associated with adverse postoperative outcome. Assessment of functional compromise by means of a nutritional questionnaire (Short Nutritional Assessment Questionnaire), a frailty questionnaire (Groningen Frailty Indicator), and sarcopenia measurement (L3 muscle index) can accurately predict postoperative sepsis.