To evaluate the nutritional status of patients in intensive care unit (ICU) by using nutritional risk screening 2002 scale (NRS2002), subjective general assessment (SGA) and critical illness nutritional risk score (NUTRIC), and to compare the characteristics and applicability of three scoring tools.

A cross-sectional survey was conducted. 315 patients admitted to the comprehensive ICU of Affiliated Qingdao Municipal Hospital of Qingdao University from April 2018 to July 2019 were enrolled. Basic information of patients was collected, and patients were divided into two groups with 65 years old as the standard to compare the nutritional status of patients among different genders and ages. The nutritional status of patients were assessed by NRS2002, SGA, and NUTRIC. Height, weight, body mass index (BMI), triceps skinfold thickness (TSF), upper arm circumference (AC), leg circumference (LC), and other related parameters of human nutrition were measured. Total protein (TP), albumin (Alb), prealbumin (PA), serum creatinine (SCr), blood urea nitrogen (BUN), total cholesterol (TC), triglyceride (TG), total number of lymphocytes (LYM), hemoglobin (Hb), C-reactive protein (CRP) and other blood biochemical indicators were performed. Spearman rank correlation analysis was used to analyze the correlation between the three nutrition evaluation scales and other objective nutrition parameters. Binary multivariate Logistic regression analysis was used to evaluate the influencing factors of nutritional status with three scales of patients in ICU.

Among 315 patients in ICU, 183 were male and 132 were female. There were 143 patients < 65 years old and 172 ≥ 65 years old. In male patients, the acute physiology and chronic health evaluation Ⅱ (APACHEⅡ) score, age and BUN of patients aged ≥ 65 years old were significantly increased, and the height, weight, BMI, TSF, AC, LC, Alb and PA were significantly lowered as compared with those aged < 65 years old, while the difference in other indicators was not statistically significant. In the female patients, the APACHEⅡ score, age, SCr and BUN of the patients aged ≥ 65 years old were significantly increased, the height, Alb, PA and Hb were significantly decreased as compared with those aged < 65 years old, and the difference in other indicators was not statistically significant. The proportion of patients with nutritional risk evaluated by NRS2002 (NRS2002 score ≥ 3) was 87.62% (276/315). SGA showed that the proportion of malnourished patients (SGA was grade B or C) was 62.86% (198/315). NUTRIC showed 66.03% of patients (208/315) in high nutritional risk (NUTRIC score ≥ 5). Spearman rank correlation analysis showed that there were significant correlations among NRS2002, SGA and NUTRIC of patients in ICU ( r_{NRS2002 with SGA} = 0.522, r_{NRS2002 with NUTRIC} = 0.392, r_{SGA with NUTRIC} = 0.442, all P < 0.01). Among the three assessment tools, SGA had the best correlation with blood biochemical indicators and body measurements to assess nutritional status, followed by NRS2002, and NUTRIC had the worst correlation. Binary multivariate Logistic regression showed that APACHEⅡ score, BMI, AC, BUN and TG were factors influencing NRS2002 assessment of nutritional status in ICU patients [odds ratio ( OR) were 2.535, 0.404, 1.438, 0.858, and 2.391, respectively, all P < 0.05]; APACHEⅡ score, age, weight, TP, BUN, LYM and CRP were influence factors of SGA for evaluating the malnutrition of patients in ICU ( OR values were 1.074, 1.038, 0.921, 0.947, 1.077, 1.625 and 0.991, respectively, all P < 0.05); APACHEⅡ score, age, LYM and CRP were the influence factors of NUTRIC assessment for malnutrition of patients in ICU ( OR values were 1.159, 1.049, 0.715 and 0.995, respectively, all P < 0.05).

The nutrition status of ICU patients evaluated by NRS2002, SGA and NUTRIC was simple and easy to operate, and the positive screening rate of NRS2002 was the highest, which was suitable for patients with mild conditions in ICU. SGA is the most valuable tool to evaluate the nutritional status of ICU patients. NUTRIC has a poor correlation with objective indicators reflecting nutritional status, while its indicators are objective and easy to obtain, which is suitable for ICU patients with critical condition and unclear consciousness. Nutritional assessment tools should be integrated with the patient's gender, age, anthropometric and biochemical indicators.

Nutritional risk screening 2002; Subjective general assessment; Critical illness nutritional risk score; Intensive care unit; Nutrition assessment