肠癌患者术后代谢性酸中毒的预测模型构建及应用价值评价  

Prediction model construction and application value evaluation of postoperative metabolic acidosis in patients with intestinal cancer

作　　者：任船 葛荣忠 邱正才 朱莉[1] Ren Chuan;Ge Rongzhong;Qiu Zhengcai;Zhu Li(Department of General Surgery,Shuyang Hospital Affiliated to Nanjing University of Traditional Chinese Medicine,Shuyang 223600,China;Department of General Surgery,The Affiliated Shuyang Hospital of Xuzhou Medical University,Shuyang 223600,China)

机构地区：[1]南京中医药大学沭阳附属医院普外科,江苏省223600 [2]徐州医科大学附属沭阳医院普外科,江苏省223600

出　　处：《中华消化病与影像杂志(电子版)》2025年第1期34-39,共6页Chinese Journal of Digestion and Medical Imageology(Electronic Edition)

摘　　要：目的构建肠癌患者术后代谢性酸中毒的预测模型,并评价其应用价值。方法纳入2020年1月至2022年12月南京中医药大学沭阳附属医院收治的271例结直肠癌患者(构建组),按照其术后代谢性酸中毒发生情况,将患者分别纳入发生组、未发生组;对比2组患者临床资料,使用Logistic多因素回归模型归纳影响肠癌患者术后发生代谢性酸中毒的相关因素。基于危险因素建立Nomogram列线图预测模型,并将其应用于2023年1月至2024年3月收治的115例患者(评价组)的术后代谢性酸中毒预测,使用受试者工作特征曲线评估预测模型的应用效能。结果构建组271例患者中,130例(47.97%)术后发生代谢性酸中毒;评价组115例患者中,52例(45.22%)术后发生代谢性酸中毒。Logistic多因素回归分析示,合并低蛋白血症、营养不良/风险、手术时间≥180 min、术前尿素氮≥7 mmol/L是影响结直肠癌患者术后代谢性酸中毒的独立危险因素,体重指数≥18.5 kg/m^(2)、补液速度≥11 mL·kg^(-1)·h^(-1)为保护性因素(P<0.05)。预测结直肠癌患者术后代谢性酸中毒的Nomogram列线图模型C指数为0.862(95%CI 0.811~0.925),列线图模型预测术后代谢性酸中毒的一致性曲线与45°对角线贴合度较高。以总分≥185分为截断值,列线图模型预测评价组115例患者术后代谢性酸中毒的AUC为0.851,其灵敏度、特异性、准确率分别为88.49%、90.48%、89.57%。结论结直肠癌患者术后代谢性酸中毒发生率较高,且与营养状态、补液速度、手术时间和肾小球滤过功能相关,基于上述因素建立的列线图能够为患者术后代谢性酸中毒预测提供较为准确的参考。Objective To construct a predictive model of postoperative metabolic acidosis in patients with colorectal cancer and evaluate its application value.Methods A total of 271 patients with colorectal cancer admitted to Shuyang Hospital Affiliated to Nanjing University of Traditional Chinese Medicine from January 2020 to December 2022 were included.According to the occurrence of postoperative metabolic acidosis,the patients were divided into occurrence group and non-occurrence group.The clinical data of the two groups were compared,and the related factors affecting postoperative metabolic acidosis in patients with intestinal cancer were summarized by Logistic multivariate regression model.Based on the risk factors,the Nomogram prediction model was established and applied to the prediction of postoperative metabolic acidosis in 115 patients(evaluation group)admitted from January 2023 to March 2024.Receiver operating characteristic curve(ROC)was used to evaluate the application efficiency of the prediction model.Results Among 271 patients in the construction group,130 patients developed metabolic acidosis after operation,the incidence rate was 47.97%.Among 115 patients in the evaluation group,52 patients developed metabolic acidosis after operation,the incidence rate was 45.22%.Logistic multivariate regression analysis showed that hypoproteinemia,malnutrition/risk,operation time≥180 min and preoperative urea nitrogen≥7 mmol/L were independent risk factors for postoperative metabolic acidosis in patients with colorectal cancer,while body mass index≥18.5 kg/m^(2)and rehydration speed≥11 mL·kg^(-1)·h^(-1)were protective factors(P<0.05).The C-index of Nomogram model for predicting postoperative metabolic acidosis in patients with colorectal cancer was 0.862(95%CI:0.811-0.925),and the consistency curve of Nomogram model for predicting postoperative metabolic acidosis had a high degree of fit with the 45°diagonal line.Taking the total score≥185 as the cutoff value,the AUC predicted by Nomogram model for postopera

关 键 词：结直肠癌 代谢性酸中毒 预测模型 效能 

分 类 号：R735.34[医药卫生—肿瘤]