Abstract: Objective To To establish a risk prediction model for postoperative gastrointestinal dysfunction in patients with gynecological malignant tumor, and to verify the prediction effect of the model. Methods A total of 281 patients with malignant tumor who underwent radical surgery in Shenzhen Maternity and Child Healthcare Hospital from January 1, 2015 to December 31, 2020 were selected as the object and they were divided into two groups: patients with postoperative gastrointestinal dysfunction (n=109) and patients without postoperative gastrointestinal dysfunction (n=172). Variables were screened by the Lasso model, and the variables screened out by the Lasso model were incorporated into the logistics regression model to construct the risk prediction model of gastrointestinal dysfunction after gynecological malignant tumor surgery. The area under the working curve (ROC) of the model was calculated to test the differentiation of the model by using the ten-fold cross-validation method. The goodness of fit and prediction effect of the models were tested respectively. Results The incidence of postoperative gastrointestinal dysfunction was 38.8%. The influencing factors were BMI (OR=0.594,95%CI：0.381~0.907), serum sodium level (OR=0.915,95%CI：0.822~1.015), omentum greater surgical resection (OR=8.388,95%CI：1.741~6.569), pelvic lymph node dissection (OR=4.148,95%CI：1.368~1.665),anesthesia time (OR=1.528,95%CI：0.910~2.612) and uric acid alkalinity(OR=0.773,95%CI：0.579~1.015). The AUC under the receiver operating characteristic curve was 0.70 (95%CI:0.63~0.75,P<0.001); the sensitivity 69%, and the specificity 60%. Conclusion The model established in the study is relatively reliable in predicting the risk of postoperative gastrointestinal dysfunction in patients with gynecological malignant tumor, and it is helpful for medical personnel to screen high-risk patients and develop targeted nursing interventions.

Key words: gynecology, malignant neoplasm, risk factor, gastrointestinal dysfunction, prediction model