Comparison of different nutritional assessments and body-composition measurements in detecting malnutrition among gynecologic cancer patients

Background: Few studies have assessed global nutritional assessment tools and body-composition measurements in gynecologic cancer patients. Objective: We aimed to assess the convergent validity of different nutritional tools such as the scored Patient-Generated Subjective Global Assessment (PG-SGA), serum albumin, skinfold-thickness measurements, and total-body potassium (TBK) and body density measurements to identify gynecologic cancer patients at risk of malnutrition. Design: We assessed the nutritional status of 194 patients with suspected or proven gynecologic cancer according to the SGA and the scored PG-SGA, and skinfold-thickness (n = 145), TBK (n = 51), and body density measurements (n = 42) before primary treatment. Results: According to the SGA and the scored PG-SGA global rating, 24% of gynecologic cancer patients were classified as malnourished. The prevalence of malnutrition was highest in ovarian (67%) and lowest in endometrial (6%) cancer patients. The ability of the PG-SGA score (P < 0.001) and albumin (P < 0.001), triceps skinfold-thickness (P = 0.041), and TBK (P = 0.005) measurements to predict the SGA was significantly better than chance. TBK significantly correlated with measurements associated with protein depletion, including age (P < 0.001), arm muscle area (P < 0.001), fat-free mass (P < 0.001), and the PG-SGA score (P = 0.009). Multiple regression analysis showed that, together, the PG-SGA score and arm muscle area adjusted for age accounted for 66% of total TBK variance. Conclusions: The PG-SGA is significantly associated with subjective and objective parameters and is a widely recognized, clinically relevant method of evaluating nutritional status. It therefore seems most appropriate for identifying malnourishment in gynecologic cancer patients.