控制性低中心静脉压肝癌切除术患者术后低血压的预测模型

吴俊雄; 杜小强; 陈坤; 刘建东

文章摘要

控制性低中心静脉压肝癌切除术患者术后低血压的预测模型

Predictive model of postoperative hypotension in patients undergoing hepatocellular carcinoma resection with controlled low central venous pressure

DOI：10.12089/jca.2024.08.005

中文关键词: 控制性低中心静脉压肝癌手术时间肿瘤直径

英文关键词: Controlled low central venous pressure Liver cancer Operation time Tumor diameter

基金项目:

作者	单位	E-mail
吴俊雄	363000,福建省漳州市,第九〇九医院,厦门大学附属东南医院麻醉科
杜小强	363000,福建省漳州市,第九〇九医院,厦门大学附属东南医院麻醉科
陈坤	363000,福建省漳州市,第九〇九医院,厦门大学附属东南医院麻醉科
刘建东	363000,福建省漳州市,第九〇九医院,厦门大学附属东南医院麻醉科	liujd909@126.com

摘要点击次数: 847

全文下载次数: 286

中文摘要:

目的：探讨构建控制性低中心静脉压（CLCVP）肝癌切除患者术后低血压预测模型。
方法：回顾性分析2020年1月至2023年6月收治肝癌切除术患者144例，男81例，女63例，年龄45～64岁，BMI 22～26 kg/m²，ASA Ⅰ或Ⅱ级。将144例患者按照2∶1随机分为试验集（n=96）和验证集（n=48），根据术后是否发生低血压将试验集和验证集分为低血压组和非低血压组。收集患者术前白蛋白、术前血红蛋白、手术方式、手术时间、肿瘤部位、肿瘤大小、肝门阻断时间、肝门阻断次数、肝门阻断间隔时间、出血量、术中CVP平均值、术中补液量、术中尿量等指标。采用单因素和多因素Logistic分析试验集中低血压发生危险因素，并建立风险预测模型，将风险预测模型在验证集中进行验证。
结果：试验集中术后低血压29例（30.2%），验证集中术后低血压15例（31.3%）。与非低血压组比较，试验集中低血压组术前白蛋白明显降低（P<0.05），手术时间明显延长、肿瘤直径≥5 cm比例、出血量明显升高（P<0.05）。多因素分析显示，术前白蛋白升高（OR=0.216，95%CI 0.164～0.665，P<0.05）是术后低血压的独立保护因素，手术时间延长（OR=2.649，95%CI 1.802～7.553，P<0.05）、肿瘤直径≥5 cm（OR=3.789，95%CI 2.011～12.458，P<0.05）、出血量增加（OR=8.873，95%CI 2.750～17.553，P<0.05）是术后低血压发生的独立危险因素。根据多因素分析结果建立控制性低中心静脉压肝癌切除术患者术后低血压危险因素：F=-408.64-（1.534×术前白蛋白）+（0.974×手术时间）+（1.332×肿瘤直径≥5 cm）+（2.183×出血量）。风险模型在验证集中进行验证，受试者工作特征曲线（ROC）下面积（AUC）为0.821（0.695～0.943），敏感性71.7%，特异性86.5%。Hosmer-Lemeshow拟合优度检验结果显示， χ²=10.654，P=0.222。
结论：手术时间延长、肿瘤直径≥5 cm、出血量增加是采用控制性低中心静脉压技术肝癌切除术后低血压发生的危险因素，术前更高的白蛋白是保护因素。通过多因素分析建立风险预测模型具有较好的预测价值。

英文摘要:

Objective: To investigate the prediction model of hypotension in patients with hepatocellular carcinoma resection using controlled low central venous pressure technique (CLCVP).
Methods: A total of 144 patients with liver cancer admitted from January 2020 to June 2023 were retrospectively analyzed, including 81 males and 63 females, aged 45-64 years, BMI 22-26 kg/m², ASA physical status Ⅰ or Ⅱ. 144 patients were randomly divided into trial set (n = 96) and verification set (n = 48) according to 2∶1. The trial set and verification set were divided into hypotensive group and non-hypotensive group according to whether hypotension occurred after operation. Preoperative albumin, preoperative hemoglobin, operation method, operation time, tumor site, tumor size, hilar block time, number of hilar block times, hilar block interval time, blood loss, mean intraoperative CVP, intraoperative fluid volume, and intraoperative urine volume were collected. Univariate and multivariate Logistic analysis were used to analyze the risk factors of hypotension in the experimental group, and a risk prediction model was established. The risk prediction model was verified in the validation group.
Results: There were 29 patients (30.2%) of postoperative hypotension in the test group and 15 patients (31.3%) of postoperative hypotension in the validation group. Compared with the non-hypotensive group, the preoperative albumin in the hypotensive group was significantly decreased (P < 0.05), the operation time was significantly prolonged, the ratio of tumor diameter ≥ 5 cm, and the amount of blood loss were significantly increased (P < 0.05). Multivariate analysis showed that preoperative albumin elevation (OR = 0.216, 95% CI 0.164-0.665, P < 0.05) was an independent protective factor for postoperative hypotension. Prolonged operative time (OR = 2.649, 95% CI 1.802-7.553, P < 0.05), tumor diameter ≥ 5 cm (OR = 3.789, 95% CI 2.011-12.458, P < 0.05), increased blood loss (OR = 8.873, 95% CI 2.750-17.553, P < 0.05) was an independent risk factor for postoperative hypotension. According to the results of multi-factor analysis, the risk factors of postoperative hypotension in patients with controlled low central venous pressure hepatocellular carcinoma resection were established: F = -408.64 - (1.534 × preoperative albumin)+ (0.974 × operation time) + (1.332 × tumor diameter ≥ 5 cm)+(2.183 × blood loss). The risk model was validated in the validation set, and the area under the ROC curve (AUC) was 0.821 (0.695-0.943), the sensitivity was 71.7%, and the specificity was 86.5%. The Hosmer-Lemeshow goodness of fit test showed that χ² = 10.654, P = 0.222.
Conclusion: Prolonged operative time, tumor diameter ≥ 5 cm and increased blood loss are risk factors for hypotension after hepatocellular carcinoma resection using CLCVP technique, and higher preoperative albumin is protective factor. The establishment of risk prediction model through multi-factor analysis has good forecasting value.

查看全文查看/发表评论下载PDF阅读器

关闭