早期胃癌患者内镜黏膜下剥离术后疲劳预测模型的构建

Analyse of influencing factors and predictive model of fatigue after endoscopic submucosal dissection in patients with early gastric cancer

目的探究早期胃癌患者胃内镜黏膜下剥离术后疲劳(POF)的危险因素,建立并验证早期胃癌POF风险预测模型。方法采用便利抽样法选取2020年10月-2022年12月于我院治疗的232例早期胃癌患者为研究对象,使用一般资料表、中文版多维度疲乏症状量表、阿姆斯特丹术前焦虑与信息需要量表、匹兹堡睡眠质量量表、艾森克人格问卷简式量表收集相关数据,通过单因素分析和二分类logistic回归分析发生术后疲劳的危险因素,并建立列线图预测模型,选取2023年1-6月在我院治疗的72例患者对模型进行验证。结果经检验,年龄、性别、睡眠障碍、病灶切除面积、气质类型、术后疼痛和胃肠功能紊乱是早期胃癌患者POF发生的独立危险因素,并以此建立早期胃癌患者POF的列线图模型,经验证该模型具有较好的准确度(H-L检验:χ^(2)=2.822,P=0.945)和区分度(AUC=0.0.909,95%CI:0.937~0.980,P<0.001),最大约登指数为0.648,灵敏度为0.832,特异度为0.813,截断值0.319,校准曲线为斜率近似于1的直线。结论该模型预测效果良好,可为医护人员对早期胃癌POF高风险人群筛查提供参考,及时采取预防性措施。

Objective To explore the risk factors of postoperative fatigue(POF)after endoscopic submucosal dissection(ESD)in patients with early gastric cancer,and to establish and verify the risk prediction model of POF of early gastric cancer.Methods A total of 232 patients with early gastric cancer in our hospital from January 2020 to December 2022 were selected by convenience sampling method.The general information questionnaire,Chinese version of Multidimensional Fatigue Inventory,Amsterdam Preoperative Anxiety and Information Need Scale,Pittsburgh Sleep Quality Scale,and Eysenck Personality Questionnaire Short Scale were used to collect relevant data.The influencing factors of postoperative fatigue were analyzed by single factor analysis and binary logistic regression,and the nomogram prediction model was established.The calibration curve was used to verify the prediction effect of the model.Results Age,gender,sleep disorder,area of lesion resection,temperament type,postoperative pain,and gastrointestinal dysfunction were tested as independent risk factors of POF in patients with early gastric cancer,and the nomograph model for POF in patients with early gastric cancer was established based on this model,which was verified to have good accuracy(H-L test:χ^(2)=2.822,P=0.945)and discrimination(AUC=0.0909,95%CI was 0.937-0.980,P<0.001),the maximum Youden index was 0.648,the sensitivity was 0.832,the specificity was 0.813,the cut-off value was 0.319,and the calibration curve was a straight line with slope approximately 1.Conclusion The model prediction effect is good,which can be used as a reference for medical staff to screen high-risk groups of postoperative fatigue in early gastric cancer and take preventive measures in time.