| 杨涵,李明晋,赵年,等.肝癌病人经导管动脉化疗栓塞术后并发肝脓肿的风险因素及预测研究[J].安徽医药,2026,30(6):1209-1214. |
| 肝癌病人经导管动脉化疗栓塞术后并发肝脓肿的风险因素及预测研究 |
| Study on risk factors and prediction of liver abscess in liver cancer patients after transcatheter arterial chemoembolization |
| |
| DOI：10.3969/j.issn.1009-6469.2026.06.029 |
| 中文关键词: 肝肿瘤 经导管动脉化疗栓塞 肝脓肿 风险因素 列线图模型 |
| 英文关键词: Liver neoplasms Transcatheter hepatic arterial chemoembolization Liver abscess Risk factors Nomogram model |
| 基金项目: |
|
| 摘要点击次数: 48 |
| 全文下载次数: 21 |
| 中文摘要: |
| 目的探讨肝癌病人经导管动脉化疗栓塞( TACE)并发肝脓肿的风险因素，并构建预测列线图模型。方法回顾性分析 2014年 1月至 2023年 4月湖北医药学院附属国药东风总医院收治的 1 011例肝癌病人的临床资料。以随机数字表法按 2∶1比例将病人分为建模组( 674例)和验证组( 337例)。术后随访 1个月，根据病人 TACE术后有无发生肝脓肿将建模组分为肝脓肿组和非肝脓肿组。采用 logistic回归分析肝癌病人 TACE并发肝脓肿的影响因素，并建立风险预测列线图模型。采用受试者操作特征曲线(ROC曲线)、校准曲线和决策曲线分析( DCA)评估列线图模型的预测效能、校准度和临床净收益，并使用验证组进行外部验证。结果肝癌病人 TACE术后肝脓肿的发生率为 3.86%(39/1 011)；肝脓肿组糖尿病史占比［ 46.15%(12/26)比 27.47%(178/648)］、肿瘤长径［(7.15±2.56)cm比( 5.22±1.72)cm］、肝胆切除手术史占比［ 7.69%(2/26)比 0.93%(6/648)］、肝癌破裂出血史占比［34.62%(9/26)比 11.57%(75/648)］、TACE术中栓塞彻底占比［ 80.77%(21/26)比 57.41%(372/648)］、术前白蛋白 ≤ 30 g/L占比［65.38%(17/26)比 32.25%(209/648)］均高于非肝脓肿组( P<0.05)； logistic回归分析显示，以上除糖尿病史外均是肝癌病人 TACE术后并发肝脓肿的危险因素( P<0.05)； ROC曲线显示，建模组和验证组列线图模型预测的曲线下面积( AUC)及其 95%CI分别为 0.85(0.83，0.87)、 0.84(0.77，0.89)；建模组、验证组列线图模型的一致性指数分别为 0.798、0.792，两组校准曲线均与标准曲线贴合良好； DCA显示，建模组阈值概率为 0~0.91时、验证组阈值概率为 0~0.82时模型具有良好的净效益。结论基于肿瘤长径、肝胆切除手术史、肝癌破裂出血史、 TACE术中栓塞彻底、术前白蛋白 ≤30 g/L构建肝癌病人 TACE术后并发肝脓肿列线图模型有助于早期识别肝脓肿高风险病人，指导临床制定干预方案。 |
| 英文摘要: |
| Objective To explore the risk factors of liver abscess in liver cancer patients undergoing transcatheter arterial chemoem.bolization (TACE), and to construct a predictive nomogram model.Methods The clinical data of 1 011 liver cancer patients admittedto Dongfeng General Hospital Affiliated to Hubei Medical College from January 2014 to April 2023 were retrospectively analyzed. Pa.tients were randomly divided (using simple random sampling) into a modeling group (674 cases) and a validation group (337 cases) at aratio of 2:1. The patients were followed up for one month after the operation, then the modeling group was divided into the liver abscessgroup and the non-liver abscess group based on whether the patient developed a liver abscess after TACE. Multivariate logistic regres.sion was used to analyze the risk factors of TACE complicated with liver abscess in liver cancer patients, and the risk prediction nomo.gram model was established. Receiver operating characteristic curve (ROC curve), calibration curves, and decision curve analysis(DCA) were drawed to evaluate the predictive efficacy, accuracy, and clinical benefits of the Nomogram model, and the validation groupwas used to evaluated validation.Results The incidence of liver abscess after TACE in liver cancer patients is 3.86% (39/1 011); Inthe group with liver abscess, the history of diabetes [46.15% (12/26) vs. 27.47% (178/648)], tumor length [(7.15±2.56) cm vs. (5.22± 1.72) cm], history of hepatobiliary resection [7.69% (2/26) vs. 0.93% (6/648)], history of rupture and bleeding of liver cancer [34.62% (9/26) vs. 11.57% (75/648)], complete embolization during TACE [80.77% (21/26) vs. 57.41% (372/648)], and proportion of preopera. tive albumin ≤30 g/L [65.38% (17/26) vs. 32.25% (209/648)] were all higher than those in the non-liver abscess group (P<0.05); Logis.tic regression analysis revealed that all factors except for a history of diabetes were risk factors for the of liver abscess inpatients with liver cancer after TACE surgery (P<0.05); The ROC curve analysis results showed that the area under the curve (AUC)predicted by the nomogram model of the modeling group and the validation group were 0.85 [95% confidence interval (95%CI): (0.83, 0.87)], 0.84 [95%CI: (0.77, 0.89)], respectively. The consistency indices of the calibration curves for the modeling and validation groupswere 0.798 and 0.792, respectively, and the calibration curves of both groups fitted well with the standard curve. The DCA showed thatthe model had good net benefits when the threshold probability of the modeling group was 0-0.91 and the threshold probability of the validation group was 0-0.82.Conclusion Based on the tumor's length and diameter, history of hepatobiliary resection, history of rup.ture and bleeding of liver cancer, complete embolization during TACE, and preoperative albumin ≤30 g/L, the establishment of a nomo.gram model of liver abscess in patients with liver cancer after TACE is helpful to early identify high-risk patients with liver abscess, and guide clinical intervention. |
|
查看全文
查看/发表评论 下载PDF阅读器 |
| 关闭 |
|
|
|
