At the same time，nutritional risk screening（NRS）2002 and PG-SGA combined with body composition analysis were used to evaluate the applicability of preoperative nutritional risk screening and assessment in patients with gastric cancer and gastrectomy，so as to find more accurate predictors of longer postoperative hospital stay and increased incidence of complications，so as to provide empirical evidence for clinical practice.

NRS 2002 and PG-SGA were used to conduct preoperative nutritional risk screening and nutritional assessment in 366 patients with gastric cancer after gastrectomy. Body composition was observed and differences of various indexes were analyzed to analyze the effects of different nutritional risk grades on postoperative complications and length of hospital stay. SPSS 23.0 was used for data analysis. Analysis of variance was used for multi-group comparison of measurement data conforming to normal distribution；Mann-Whitney U test was used for comparison of count data conforming to normal distribution；Chi-square test was used for comparison of count data and Kappa test was used for consistency. P<0.05 was considered to be statistically significant.

The screening rates for NRS 2002 and PG-SGA were 50.3% and 71.3%，including 48.1% for mild or suspicious malnutrition and 23.2% for severe malnutrition. The consistency test results of the two screening methods were medium，Kappa=0.439. Preoperative albumin，body mass index（BMI），total body water content，muscle mass，fat-free weight，body fat，skeletal muscle，skeletal muscle index，basal metabolic rate，upper arm circumference and upper arm muscle circumference were all lower in patients with nutritional risk than in patients without nutritional risk（P<0.05）. ROC curve analysis showed that the best cut-off point for NRS 2002 to predict the length of hospital stay was 3.5 points，the best cut-off point for preoperative albumin to predict the length of hospital stay was 38.6g/L，and the best cut-off point for PG-SAG to predict complications was 8.5 points.

NRS 2002 and PG-SGA had moderate consistency in nutritional risk screening，and had different values in predicting complications and length of hospitalization，which could not be substituted for each other. Patients with NRS 2002 score ≥4，preoperative albumin less than 38.6 g/L had longer hospital stay，and patients with PG-SGA score ≥9 had a higher incidence of complications.