Abstract

Malnutrition is an important but neglected predictor for outcomes and healthcare costs in cancer patients. A simple screening tool for detecting malnutrition may have clinical utility in their preoperative assessment. This study compared three validated indices, for their predictive ability for prolonged length of stay (LOS) and 30-day postsurgical complications in malignancies in a tertiary hospital in South India. 342 cancer patients admitted for elective surgery were stratified on their preoperative day using MUST - Malnutrition Universal Screening Tool, SGA - Subjective Global Assessment and NRI - Nutritional Risk Index. The postoperative LOS and 30-day morbidity as per Clavien-Dindo classification (CDC) were compared to calculate the predictive accuracy of the various nutritional indices. In our study, 44% patients were classified as malnourished by SGA. SGA was found to have maximum correlation coefficient with LOS (σ=0.410), followed by MUST (σ=0.401) and NRI (σ=0.36). On logistic regression, MUST, NRI and SGA were all found to be significant independent predictors of LOS and CDC class. Age, acute illness and comorbidities were found to have significant confounding effects. Sensitivity of MUST and SGA to predict prolonged LOS (>10d) were. Our study showed that surgical cancer population was at high risk for malnutrition. MUST and SGA were good risk-stratification tools with independent predictive ability for prolonged LOS and postoperative complications. Patients having higher MUST and SGA scores fared poorly postoperatively with significant prolongation of stay. MUST had 88% agreement with SGA, and took shorter time to administer in the in-patient setting. Routine preoperative nutritional assessment is important in cancer patients. SGA and MUST were found to be reliable tools, with MUST being the simplest and quickest tool for preoperative nutritional assessment. Patients stratified to be at risk for malnutrition may benefit with nutritional therapy.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call