Indicadores bioquímicos e terapia nutricional enteral como preditores de mortalidade em hospital universitário de Minas Gerais

Abstract

Introduction: The decline in nutritional status, long hospital stay and high hospital mortality rates are still a reality in many hospitals. Studies have shown that malnutrition, inflammatory, immune, multimorbidity and nutritional therapy have been associated with mortality. Objective: To evaluate predictors of hospital mortality in patients admitted to a university hospital. Material and Methods: A retrospective study was conducted between 2014 and 2016, including patients over the age of 18 admitted to the Clinical and Surgical Clinic wards. Two articles were developed, the first one evaluating patients submitted to nutritional evaluation and requesting biochemical tests, whose hypothesis was the high distribution of red blood cell distribution (RDW) distribution in hospital admission could be a predictor of hospital mortality. For the second article, it was hypothesized that in patients undergoing exclusive enteral nutritional therapy with caloric and protein supply lower than the recommendation could also predict in-hospital mortality. Cox regression was used to determine the predictors of in-hospital mortality. Results: Article 1: Of the 2923 patients, 46.1% were older than 60 years and 58.7% were male, 55.7% were malnourished and 4.7% died. The accuracy of the RDW in predicting hospital mortality was characterized by the area under the ROC curve adjusted by covariates of 0.577 (95% CI: 0.525, 0.629). RDW was associated with an hospital mortality risk for all causes (HR 1.11 - 95% CI: 1.01-1.21), regardless of age, type of ward, nutritional status, immunological status, and multimorbidity. Each 1% in the RDW value increased the risk of all-cause hospital death by 17% in surgical patients. Article 2: Of the 240 patients, 58.3% were older than 60 years and 60.0% were males. The frequencies of hospital mortality (19.2%) and malnutrition (78.8%) were high. The mean values of protein (0.75g/kg/day) and energy (17.60kcal/kg/day) were below the general recommendations and 37.8% did not reach the mean of 20 kcal/kg/day period of hospitalization. Hypocaloric nutrition (HR: 5.78; 95% CI: 1.59-21.04) and hypoprotein nutrition (HR: 3.69; 95% CI: 1.25-10.93) were predictors of in-hospital mortality by all causes in adjusted multivariate models. However, hypoprotein nutrition when adjusted for non-protein calories/g of nitrogen increased the risk of death (HR: 3.15; 95% CI: 1.04-9.53). Conclusion: The biochemical indicator RDW and the low supply of protein are associated with the risk of hospital mortality. Therefore, it is recommended to use the RDW in practice for the identification of surgical patients with a higher risk of death, and it is recommended to include them in the clinical evaluation, as well as the greater nutritional surveillance of the non-critical patient in exclusive enteral nutritional therapy, and dietary evolution at a minimum of 20 kcal/kg/day of calories and 0.8kg/ kg/day considering that patients have a greater chance of death when these parameters are inadequate.