直肠癌辅助化疗相关性便秘发生风险预测模型的构建与验证

Construction and validation of the risk prediction model for the occurrence of adjuvant chemotherapy-associated constipation in rectal cancer

目的探讨影响直肠癌辅助化疗相关性便秘发生的危险因素,并以此构建风险预警模型并验证。方法回顾性选取2022年4月—2023年5月于江西省肿瘤医院收治的188例进行手术后并接受辅助性化疗的直肠癌患者作为模型组,根据有无便秘发生模型组分为便秘组46例和非便秘组142例。选取2023年6—12月于江西省肿瘤医院收治的79例进行手术后并接受辅助性化疗的直肠癌患者作为验证组进行外部验证。采用单因素及多因素Logistic回归分析直肠癌辅助化疗相关性便秘发生的独立危险因素,应用R语言软件绘制相应的列线图,以此构建对应的风险预警模型。同时在模型组数据(内部验证)和验证组数据(外部验证)中采用受试者工作特征(ROC)曲线、校准曲线及Hosmer-Lemeshow(H-L)拟合优度检验来评估模型的预测性能;采用临床决策曲线来评估模型的临床适用性。结果单因素结果表明便秘组的年龄、活动量、高蛋白质、低纤维食物摄入、是否使用止吐药和焦虑紧张与非便秘组相比差异均有统计学意义(P<0.05);多因素分析结果提示年龄、活动量、食物中蛋白质、纤维素的摄入、是否使用止吐药、焦虑紧张是影响直肠癌化疗后发生便秘的因素。年龄≥65岁、活动量少、高蛋白质、低纤维食物摄入、使用止吐药和焦虑紧张会增加直肠癌辅助化疗后发生便秘的风险(P<0.05)。基于以上6个独立影响因素构建相应的风险预测模型。模型组ROC曲线下AUC为0.815(95%CI=0.743~0.886,P=0.000),敏感度为59.50%、特异度为86.10%,H-L拟合优度检验结果显示χ^(2)=5.383,P=0.716,P>0.05,模型拟合情况良好;校准曲线结果显示,模型预测概率与实际发生情况之间具有良好的一致性;临床决策曲线结果显示,模型具有较高的临床适用价值。结论直肠癌辅助化疗相关性便秘发生受多种因素影响,构建的预测模型具有较高临床应用价值。医护人员可针对其中的危险因素采取相应措施以降低便秘发生风险。

Objective To explore the risk factors affecting the occurrence of adjuvant chemotherapy-associated constipation in rectal cancer,to construct and validate a risk warning model according to those risk factors.Methods There were 188 rectal cancer patients admitted to Jiangxi Provincial Cancer Hospital from April 2022 to May 2023 who underwent surgery and received adjuvant chemotherapy were retrospectively selected as the model group,they were divided into 46 cases in the constipation group and 142 cases in the non-constipation group according to the occurrence of constipation.Another 79 rectal cancer patients who underwent surgery and received adjuvant chemotherapy in Jiangxi Provincial Cancer Hospital from June to December 2023 were selected as the validation group for external validation.The independent risk factors of constipation associated with adjuvant chemotherapy for rectal cancer were analyzed by univariate factor and multifactor Logistic regression,and the corresponding line graphs were drawn by R language software to construct the corresponding risk warning model.The predictive performance of the model was evaluated using the Receiver Operating Characteristic(ROC)curve,calibration curve and Hosmer-Lemeshow(H-L)goodness-of-fit test in the model group(internal validation)and validation group(external validation),and the clinical applicability of the model was evaluated using the clinical decision curve.Results The univariate results showed that the differences in age,activities level,high protein,low fiber food intake,whether to take antiemetic drugs and anxiety and tension in the constipated group were statistically significant compared with those in the non-constipated group(P<0.05),The results of the multifactorial analysis suggested that age,activities level,protein and fiber intake of food,whether to take antiemetic drugs,and anxiety and tension were the factors affecting the occurrence of constipation after chemotherapy for rectal cancer.Age≥65 years,low activities level,high protein,low fiber food intake,use of antiemetic drugs,and anxiety and tension increased the risk of developing constipation after adjuvant chemotherapy for rectal cancer(P<0.05).A risk prediction model was constructed based on the above six independent factors.The AUC under the ROC curve of the model group was 0.815.(95%CI=0.743~0.886,P=0.000),sensitivity level was 59.50%,specificity was 86.10%,and the results of H-L goodness-of-fit test showed thatχ^(2)=5.383,P=0.716,P>0.05,the model fit was good;the calibration curve results showed that there was a good agreement between the model predicted probability and actual occurrence.There was good agreement between the model and the actual occurrence,the results of the clinical decision curve showed that the model had a high clinical applicability value.Conclusion The occurrence of constipation associated with adjuvant chemotherapy for rectal cancer was affected by various factors,the constructed prediction model had high clinical application value.The predictive model has high clinical application value,and healthcare professionals can take corresponding measures to reduce the risk of constipation according to the risk factors.