[Comparison of the CONUT method with SGA and NSA for the prediction of complications, hospital stay, readmissions, and mortality].

INTRODUCTION: disease-related malnutrition (DRM) is a specific type of malnutrition caused by the inflammatory response to the underlying disease. Its prevalence worldwide varies from 30 % to 50 %, being similar in Spanish hospitals. DRE is not commonly recognized but is usually misdiagnosed and generally not treated. It is associated with an increased risk of morbidity, mortality, and costs. Nutritional societies recommend that screening be performed within the first 24 to 48 hours after admission for the early detection of malnutrition. No screening tool is universally accepted.

OBJECTIVES: to evaluate the predictive validity (hospital stay, complications, readmissions and mortality) of the CONUT method as compared to SGA and NSA.

MATERIAL AND METHOD: a retrospective study included in a prospective observational study of 365 hospitalized patients from July to December 2012.

RESULTS: the most frequent admission services were Internal Medicine and Oncology (30.7 % and 29.3 %). Moderate and severe risk of malnutrition: CONUT, 42.2 % and 12.1 %, SGA 25.8 % and 10.1 %, and NSA 13.7 % and 14.5 %. Malnutrition evaluated using the CONUT method was significantly related to complications (p = 0.036), readmissions (p = 0.041) and mortality (p = 0.007). The ROC curves for mortality, for all the methods evaluated, showed that CONUT is the best tool.

CONCLUSIONS: CONUT is an automatic detection tool that can be used as a first step in the diagnosis risk of malnutrition. CONUT offers the advantage of being a prognostic factor for complications, readmission, and mortality.