Development and validation of a prediction model for GLIM-defined malnutrition at the 6-month postoperative landmark after gastrectomy for gastric cancer: a multicenter study

This study aimed to develop and validate a model for predicting Global Leadership Initiative on Malnutrition (GLIM)-defined malnutrition at the 6-month postoperative landmark after curative gastrectomy for gastric cancer. The endpoint represented nutritional status at this landmark rather than strictly new-onset postoperative malnutrition. This retrospective multicenter study included 604 patients who underwent curative gastrectomy at Gansu Provincial Hospital between January 2017 and June 2023 or at Gansu Wuwei Tumor Hospital between February 2019 and December 2021. The outcome assessment was defined as the eligible nutritional evaluation closest to postoperative day 180 within days 170–190. Patients with a Nutritional Risk Screening 2002 score ≥ 3 underwent further assessment according to the GLIM criteria. Of 501 patients from Gansu Provincial Hospital, 351 were assigned to the training cohort and 150 to the internal validation cohort; 103 patients from Gansu Wuwei Tumor Hospital comprised the external validation cohort. Least absolute shrinkage and selection operator regression and multivariable logistic regression were used to develop a nomogram. Analyses were restricted to patients who survived to and completed the eligible assessment. GLIM-defined malnutrition was identified in 179 of 604 patients (29.6%). The final model included TNM stage, postoperative adjuvant chemotherapy, age, body mass index, and preoperative serum albumin. The nomogram achieved an area under the receiver operating characteristic curve of 0.897 (95% CI, 0.842–0.952) in the internal validation cohort and 0.863 (95% CI, 0.784–0.942) in the external validation cohort. Decision curve analysis suggested potential net benefit across threshold probabilities of approximately 10%–80%. This nomogram may support individualized estimation of the probability of GLIM-defined malnutrition at the 6-month postoperative landmark and help identify patients who warrant enhanced nutritional assessment, surveillance, and follow-up. The predicted outcome reflects nutritional status at this landmark and may include both persistent and newly identified malnutrition.