Prognostic Value of the SGA and NUTRIC in the ICU

Critical Illness, Malnutrition

Subjective Global Assessment, Nutrition Risk in Critically ill Score, Critical Illness

There are two nutrition assessment tools that are commonly used in the ICU, namely the Subjective Global Assessment (SGA) and Modified Nutrition Risk in Critically Ill Score (mNUTRIC). It has been proposed that both assessments should be performed in the ICU but their combined prognostic ability has not been adequately assessed. This study aimed to: 1) determine the agreement between SGA and mNUTRIC scores, and 2) quantify their utility in discriminating and quantifying hospital mortality risk both independently and in combination.

This prospective observational study will be conducted in a 35-bed mixed ICU in Ng Teng Fong General Hospital, and all the intensivists and nurses will be blinded to the objectives of the study. All patients admitted to the ICU will be consecutively included in the study. For patients readmitted to the ICU during the same hospitalisation, only data from the first admission will be collected. As per routine care, all patients will have their nutritional status assessed by the dietitian within 48 hours of ICU admission. Information required for the nutritional assessment (SGA) will be obtained from the patients or their main care givers, and nutritional status will be dichotomized into well-nourished and malnourished. The electronic medical records automatically and prospectively collects all data required to calculate the mNUTRIC. At the end of the study, the mNUTRIC will be retrospectively calculated. Patients with values of 0-4 will be classified as low-mNUTRIC and 5-9 as high-mNUTRIC. The primary outcome will be hospital mortality and all patients will be followed until discharge or death, for up to one year after admission to the ICU. Agreement and mortality discriminative value (i.e. discrimination) of the 2-category classification of mNUTRIC (Low- and high-mNUTRIC) and SGA (SGA-A and SGA-B/C) will be assessed by Kappa statistics and C-statistics respectively. A multivariate logistic regression will be used to generate the adjusted odds ratios that quantify the association between high-mNUTRIC, malnutrition, and their combination (mNUTRIC ≥ 5 and SGA-B/C) with hospital mortality.

No related papers found yet.

None listed