A study and intervention method for factors affecting weight change after curative gastrectomy for the patients of the gastric cancer

Background: Weight changes after gastrectomy affect not only quality of life but also prognosis and survival. However, it remains challenging to predict the weight changes of individual patients. Using clinicopathological variables, we built a user-friendly tool to predict weight change after curative gastrectomy for gastric cancer. Materials and Methods: The clinical data of 984 patients who underwent curative gastrectomy between 2009 and 2013 were retrospectively reviewed and analyzed. Multivariate logistic regression was performed to identify variables predictive of postoperative weight change. A nomogram was developed and verified via bootstrap resampling. Results: Age, sex, performance status, body mass index, extent of resection, pathological stage, and postoperative weight change significantly influenced postoperative weight recovery. Postoperative levels of hemoglobin, albumin, and ferritin, and total iron-binding capacity were significant covariates. The nomogram performed well (concordance index = 0.637); calibration curves indicated appropriate levels of agreement. We developed an online version of the weight prediction tool based on the nomogram (http://gc-weightchange.com/en/front/). Conclusion: The novel, Web-based predictive nomogram allows surgeons to explore patient weight patterns quickly. The model identifies patients at high risk for weight loss after gastrectomy; such patients require multidisciplinary medical support.