Jaqueline da Silva Fink1, Aline Marcadenti2,3, Estela Iraci Rabito4, Flávia Moraes Silva5. 1. Serviço de Nutrição e Dietética, Hospital Nossa Senhora da Conceição (SND/HNSC), Porto Alegre, Brasil. 2. Departamento de Nutrição, Universidade Federal de Ciências da Saúde de Porto Alegre (UFCSPA), Porto Alegre, Brasil. 3. Programa de Pós-Graduação em Ciências da Saúde: Cardiologia, Instituto de Cardiologia/Fundação Universitária de Cardiologia do Rio Grande do Sul (IC/FUC), Porto Alegre, Brasil. 4. Departamento de Nutrição e Programa de Pós graduação em Alimentação e Nutrição, Universidade Federal do Paraná (UFPR), Curitiba, Brasil. 5. Departamento de Nutrição e Programa de Pós Graduação em Nutrição e Saúde, Universidade Federal de Minas Gerais (UFMG), Belo Horizonte, Brasil.
Abstract
BACKGROUND: Recently, the European Society for Clinical Nutrition and Metabolism (ESPEN) provided novel consensus criteria for malnutrition diagnosis. This study aimed to evaluate the applicability of this instrument in combination with different nutrition screening tools (1) to identify malnutrition and (2) to predict morbidity and mortality in hospitalized patients. MATERIALS AND METHODS: Observational prospective study in 750 adults admitted to the emergency service of a tertiary public hospital. Subjective Global Assessment (SGA-reference method) and the new ESPEN criteria were used to assess nutrition status of patients, who were initially screened for nutrition risk using 4 different tools. Outcome measures included length of hospital stay, occurrence of infection, and incidence of death during hospitalization, analyzed by logistic regression. RESULTS: There was a lack of agreement between the SGA and ESPEN definition of malnutrition, regardless of the nutrition screening tool applied previously (κ = -0.050 to 0.09). However, when Malnutrition Screening Tool and Nutritional Risk Screening-2002 (NRS-2002) were used as the screening tool, malnourished patients according to ESPEN criteria showed higher probability of infection (relative risk [RR], 1.54; 95% confidence interval [CI], 1.02-2.31 and RR, 2.06; 95% CI, 1.37-3.10, respectively), and when the NRS-2002 was used, the risk for death was 2.7 times higher (hazard ratio, 2.69; 95% CI, 1.07-6.81) in malnourished patients than in well-nourished patients. CONCLUSION: Although the new ESPEN criteria had a poor diagnostic value, it seems to be a prognostic tool among hospitalized patients, especially when used in combination with the NRS-2002.
BACKGROUND: Recently, the European Society for Clinical Nutrition and Metabolism (ESPEN) provided novel consensus criteria for malnutrition diagnosis. This study aimed to evaluate the applicability of this instrument in combination with different nutrition screening tools (1) to identify malnutrition and (2) to predict morbidity and mortality in hospitalized patients. MATERIALS AND METHODS: Observational prospective study in 750 adults admitted to the emergency service of a tertiary public hospital. Subjective Global Assessment (SGA-reference method) and the new ESPEN criteria were used to assess nutrition status of patients, who were initially screened for nutrition risk using 4 different tools. Outcome measures included length of hospital stay, occurrence of infection, and incidence of death during hospitalization, analyzed by logistic regression. RESULTS: There was a lack of agreement between the SGA and ESPEN definition of malnutrition, regardless of the nutrition screening tool applied previously (κ = -0.050 to 0.09). However, when Malnutrition Screening Tool and Nutritional Risk Screening-2002 (NRS-2002) were used as the screening tool, malnourished patients according to ESPEN criteria showed higher probability of infection (relative risk [RR], 1.54; 95% confidence interval [CI], 1.02-2.31 and RR, 2.06; 95% CI, 1.37-3.10, respectively), and when the NRS-2002 was used, the risk for death was 2.7 times higher (hazard ratio, 2.69; 95% CI, 1.07-6.81) in malnourished patients than in well-nourished patients. CONCLUSION: Although the new ESPEN criteria had a poor diagnostic value, it seems to be a prognostic tool among hospitalized patients, especially when used in combination with the NRS-2002.