Establishment of a predictive model for postoperative bleeding risk in elderly patients receiving long-term anticoagulants after radical gastrectomy

Objective To establish a risk predictive model for postoperative bleeding risks in elderly gastric cancer patients receiving long-term anticoagulants after gastrectomy, and to provide reference for nursing care work. Methods A total of 30 gastric cancer patients with bleeding after gastrectomy (an observation group) and 200 elderly gastric cancer patients with little or without bleeding after gastrectomy (a control group) were analyzed, and their clinical data were retrospectively analyzed. The risk factors of postoperative bleeding were screened out by Logistic regression analysis to establish a predictive model for bleeding risks. A receiver operating characteristic (ROC) curve was used to evaluate the predictive performance of the risk predictive model. Both groups were compared for their postoperative clinical manifestation. Results Multivariate analysis showed that platelets<100×10⁹/L, TNM stage Ⅲ and abdominal infection were independent risk factors for postoperative bleeding in elderly patients receiving long-term anticoagulants after gastrectomy. The AUC of the established predictive model was 0.736 (95% CI: 0.674—0.791), with a sensitivity of 60.00% and a specificity of 79.00%. The incidence of hematemesis, black stool and abnormal drainage fluid in the observation group was significantly higher than that in the control group (P<0.05). Conclusions Low platelet level, advanced patients and postoperative abdominal infection were the independent risk factors of postoperative bleeding in elderly patients receiving long-term anticoagulants after gastrectomy. It is necessary for medical nurses to be alert for the occurrence of postoperative bleeding according to clinical feature.