急性阑尾炎患者腹腔镜阑尾切除术后发生粘连性肠梗阻预测模型的建立及验证

Establishment and verification of prediction model for adhesive ileus after laparoscopic appendectomy in patients with acute appendicitis

摘要:

目的 探讨急性阑尾炎患者腹腔镜阑尾切除术（LA）后发生粘连性肠梗阻的影响因素，建立预测模型，并进行验证。方法 回顾性分析2021年1月－2024年1月该院收治的100例接受LA治疗的急性阑尾炎患者的临床资料，按7∶3的比例随机将患者分为训练集（70例）和验证集（30例），再根据LA术后随访6个月是否发生粘连性肠梗阻，将训练集患者分为发生组（12例）和未发生组（58例），比较两组患者的临床资料，采用多因素Logistic回归模型，分析急性阑尾炎患者LA术后发生粘连性肠梗阻的独立危险因素；并构建列线图预测模型，利用Bootstrap法、校准曲线、受试者操作特征曲线（ROC curve）和决策曲线，对训练集和验证集进行预测模型性能评价。结果 多因素Logistic回归模型结果显示，病程 > 24 h、术前白蛋白 ≤ 34 g/L、术者LA手术台数 < 60台、手术时间 > 1 h和首次下床活动时间 > 24 h是影响急性阑尾炎患者LA术后发生粘连性肠梗阻的独立危险因素（P < 0.05）；在训练集与验证集中，Bootstrap法和ROC curve结果显示，该预测模型具有较好的区分度，C指数分别为0.723（95%CI：0.642～0.875）和0.706（95%CI：0.628～0.768）；曲线下面积（AUC）分别为0.874（95%CI：0.625～0.924）和0.867（95%CI：0.612～0.895）；拟合优度检验表明：预测模型拟合效果良好（P > 0.05）；训练集与验证集的校准曲线接近于理想曲线，该预测模型准确度良好；训练集与验证集的决策曲线显示，当高风险阈值概率分别在0.10～0.80和0.25～0.48，净获益率 > 0，该列线图模型能产生较好的临床效益。结论 急性阑尾炎患者LA术后粘连性肠梗阻的发生与病程、白蛋白水平、术者经验、手术时间和术后首次下床活动时间有关，该列线图模型，具有良好的预测效能。

Abstract:

Objective To explore the influencing factors of adhesive ileus after laparoscopic appendectomy (LA) in patients with acute appendicitis, then establish and verify its prediction model.Methods 100 cases of acute appendicitis treated by LA from January 2021 to January 2024 were retrospectively analyzed. According to the ratio of 7:3, the patients were randomly divided into the training group (70 cases) and the verification group (30 cases), and then the patients in the training group were divided into the occurrence group (12 cases) and the non-occurrence group (58 cases) according to whether adhesive ileus occurred during the follow-up for 6 months after LA. The difference of clinical data between the two groups was compared, and multivariate Logistic regression method was used to analyze the independent risk factors of adhesive ileus after LA. The nomogram prediction model was constructed, and the performance of the prediction model was evaluated by Bootstrap method, calibration curve, receiver operator characteristic curve (ROC curve) and decision curve.Results The results of multivariate Logistic regression model showed that the risk factors of postoperative adhesive ileus in patients with acute appendicitis after LA were the course of disease > 24 h, preoperative albumin ≤ 34 g/L, the number of LA operating tables for LA operators < 60, the operation time > 1 h, and the time of getting out of bed for the first time after operation > 24 h were independent risk factors for adhesive ileus after LA operation (P < 0.05). In the training set and verification set, the results of Bootstrap method and ROC curve showed that the prediction model had good discrimination, and the C indexes were 0.723 (95%CI: 0.642 ~ 0.875) and 0.706 (95%CI: 0.628 ~ 0.768) respectively. AUCs were 0.874 (95%CI: 0.625 ~ 0.924) and 0.867 (95%CI: 0.612 ~ 0.895) respectively. Goodness-of-fit test showed that the prediction model had good fitting effect (P > 0.05). The calibration curves of training set and verification set were close to the ideal curve, and the prediction model had good accuracy. The decision curves of training set and verification set showed that the nomogram model could produce better clinical benefits when the threshold probability of high risk was 0.10 ~ 0.80 and 0.25 ~ 0.48 respectively, and the net benefit rate was greater than 0.Conclusion The occurrence of adhesive ileus after LA in patients with acute appendicitis is related to the course of disease, albumin level, operator experience, operation time and the first time to get out of bed after surgery. The nomogram model has good predictive performance.