微创医学

目的探究腹腔镜手术治疗结直肠癌老年患者术后并发肠梗阻的危险因素，构建列线图模型并验证。方法回顾性分析接受腹腔镜手术治疗的330例结直肠癌老年患者的临床资料，以术后是否发生肠梗阻为分组依据，将发生肠梗阻的26例患者纳入肠梗阻组，未发生肠梗阻的304例患者纳入非肠梗阻组。采用多因素Logistic和LASSO两种回归模型分析患者发生肠梗阻的危险因素，构建列线图模型并验证。结果330例患者中，肠梗阻的发生率为7.88%（26/330），多因素Logistic回归分析结果显示，肠道肿瘤切除手术史、术中造瘘、术前肠梗阻、术前贫血、术后腹腔感染是结直肠癌老年患者腹腔镜术后并发肠梗阻的独立危险因素（均P＜0.05）。列线图模型进行内部验证后，校正曲线趋势落在对角线附近区域，平均绝对误差为0.03。列线图模型预测老年患者腹腔镜术后并发肠梗阻风险的曲线下面积（AUC）为0.761（95% CI: 0.6635～0.8592），特异度为0.701，灵敏度为0.692。Hosmer-Lemeshow检验的 χ2=7.621，P=0.471，模型的预测数据与实际数据之间差异不显著，提示模型具有较好校准能力。结论根据腹腔镜手术治疗结直肠癌老年患者术后并发肠梗阻的危险因素，构建的列线图模型具有较好的校准能力和预测效果，能够有效预测患者术后并发肠梗阻的风险，具有显著的临床应用价值。

ObjectiveTo explore the risk factors of postoperative intestinal obstruction in elderly patients with colorectal cancer treated by laparoscopy, and to construct and verify a nomogram model. MethodsThe clinical data of 330 elderly patients with colorectal cancer who underwent laparoscopic treatment were retrospectively analyzed. According to whether intestinal obstruction occurred after surgery, 26 patients with intestinal obstruction were included in the intestinal obstruction group, and 304 patients without intestinal obstruction were included in the non-intestinal obstruction group. Multivariate Logistic and LASSO regression models were used to analyze the risk factors of intestinal obstruction in patients, and a nomogram model was constructed and verified. ResultsAmong the 330 patients, the incidence of intestinal obstruction was 7.88% (26/330). The result of multivariate Logistic regression analysis showed that the history of intestinal tumor resection, intraoperative fistula, preoperative intestinal obstruction, preoperative anemia, and postoperative abdominal infection were independent risk factors for postoperative intestinal obstruction in elderly patients with colorectal cancer after treated by laparoscopic surgery (all P＜0.05). After the internal validation of the nomogram model, the correction curve trend fell in the area near the diagonal, and the mean absolute error was 0.03. The area under the curve (AUC) of the nomogram model for predicting the risk of postoperative intestinal obstruction in elderly patients with colorectal cancer after treated by laparoscopic surgery was 0.761 (95% CI: 0.6635-0.8592), with a specificity of 0.701 and a sensitivity of 0.692. χ2=7.621, P=0.471 of the Hosmer-Lemeshow test showed that there was no statistically significant difference between the predictive data and actual data, which indicated that the model had good calibration ability. ConclusionThe nomogram model constructed on the risk factors of postoperative intestinal obstruction in elderly patients with colorectal cancer treated by laparoscopy has good calibration ability and prediction effect, which can effectively predict the risk of postoperative intestinal obstruction in patients and has significant clinical application value.