Ability of malnutrition screening and assessment tools to identify computed tomography defined low muscle mass in colorectal cancer surgery

Background Malnutrition and low muscle mass are independently associated with poor outcomes in colorectal cancer (CRC). However, tools to identify low muscle mass are limited in the clinical setting. We investigated the ability of existing malnutrition screening and assessment tools to identify low muscle mass assessed by computed tomography (CT). Secondary aims were to determine the feasibility of CT analysis and handgrip strength (HGS). Methods and Analysis An exploratory study of patients who underwent curative surgery for CRC between February and September 2019. Nutrition tools used included body mass index (BMI), Malnutrition Screening Tool (MST), and Patient-Generated Subjective Global Assessment (PG-SGA). Muscle mass was determined by preoperative CT image at the third lumbar vertebral level (L3), and muscle strength was determined by HGS dynamometry. Fisher's exact and Mann-Whitney U tests were used to compare results of nutrition tools with CT muscle assessment. Results In total, 57 patients were included. MST classified 18 patients (32%) as at risk of malnutrition, and PG-SGA classified 10 patients (17%) as malnourished. Fifty-one (90%) CT scans were analysable and 21 (47%) had low muscle mass. Of those with low muscle mass, PG-SGA classified 22 patients (92%) as well nourished and MST classified 17 patients (71%) as not being at nutrition risk. No tool was able to identify CT-diagnosed low muscle mass. Inability to complete HGS was associated with malnutrition (P = .001). Conclusion In this cohort, nutrition screening and assessment tools did not identify CT-diagnosed low muscle mass. Feasible tools to identify low muscle mass in the clinical setting are required.