欢迎光临《中华消化外科杂志》官方网站
欢迎您, 游客! 登录 | 注册
首页 » 过刊浏览 » 杂志期刊
《中华消化外科杂志》2018年5月第17卷第5期论著

术前肝功能Child-Pugh评分与白蛋白-胆红素评分对肝癌患者肝切除术后肝衰竭和预后的预测价值

阅读全文:
收藏:【点击:115】【下载:44】
引用本文:
张长坤,张龙辉,王东,等.术前肝功能Child-Pugh评分与白蛋白-胆红素评分对肝癌患者肝切除术后肝衰竭和预后的预测价值[J].中华消化外科杂志,2018,17(5):474-482.DOI:10.3760/cma.j.issn.1673-9752.2018.05.011.
【摘要】

目的:探讨术前肝功能Child-Pugh评分与白蛋白-胆红素(ALBI)评分对肝癌患者肝切除术后肝衰竭和预后的预测价值。
方法:采用回顾性队列研究方法。收集2010年1月至2014年10月北京大学人民医院收治的的226例肝癌患者的临床资料。患者术前行相关检查,根据患者具体情况多学科团队讨论判断肝切除术的可行性及肝切除范围,行相应手术治疗。观察指标:(1)手术情况。(2)影响肝癌患者肝切除术后肝衰竭的因素分析。(3)肝功能Child-Pugh评分和ALBI评分预测肝切除术后肝衰竭的受试者工作特征(ROC)曲线分析。(4)随访和生存情况。(5)影响肝癌患者肝切除术后预后的因素分析。采用门诊和电话方式进行随访,了解患者术后生存情况。随访时间截至2016年8月。正态分布的计量资料采用±s表示,偏态分布的计量资料采用M(P25,P75)表示。采用Kaplan-Meier法绘制生存曲线,Log-rank检验进行生存分析。采用logistic回归模型进行肝切除术后肝衰竭影响因素分析;采用COX回归模型进行肝癌患者肝切除术后预后因素分析。采用ROC曲线分析肝功能Child-Pugh评分和ALBI评分的预测价值。
结果:(1) 手术情况:226例患者均成功完成手术,171例行小范围手术切除(<3个Couinaud肝段切除),55例行大范围手术切除(≥3个Couinaud肝段切除)。226例患者术中出血量为1~22 550 mL,中位术中出血量为800 mL。226例患者中,89例出现术后肝衰竭,其他并发症包括肺部感染9例、胆汁漏4例、消化道出血4例、切口感染3例、感染性休克1例,经过积极生命支持治疗及对症治疗,患者均顺利恢复出院。226例患者术后住院时间为2~49 d,中位术后住院时间为12 d。(2)影响肝癌患者肝切除术后肝衰竭的因素分析。单因素分析结果显示:性别、总胆红素(TBil)、白蛋白(Alb)、凝血酶原时间(PT)、国际标准化比值、血小板(PLT)、腹腔积液、术中出血量、肝功能Child-Pugh评分、肝功能ALBI评分是影响肝癌患者肝切除术后肝衰竭的相关因素[优势比(OR)=0.490,1.077,0.763,1.613,26.342,0.990,2.458,5.052,2.875,34.570,95%可信区间(CI):0.248~0.971,1.030~1.127,0.699~0.833,1.248~2.087,2.722~254.936,0.985~0.995,1.386~4.361,2.467~10.347,1.807~4.576,11.674~102.376,P<0.05]。多因素分析中为了避免共线性,肝功能Child-Pugh评分和ALBI评分采用不同logistic回归模型进行分析。肝功能Child-Pugh评分logistic回归模型中剔除TBil、Alb、PT和腹腔积液,结果显示:PLT、术中出血量、肝功能Child-Pugh评分是影响肝癌患者肝切除术后肝衰竭的独立因素(OR=0.991,4.462,2.412,95%CI:0.986~0.996,2.090~9.527,1.479~3.934,P<0.05)。肝功能ALBI评分logistic回归模型中剔除Alb和TBil,结果显示:PLT、术中出血量、肝功能ALBI评分是影响肝癌患者肝切除术后肝衰竭的独立因素(OR=0.990,4.867,42.947,95%CI:0.984~0.995,2.088~11.346,12.409~148.637,P<0.05)。(3)肝功能Child-Pugh评分和ALBI评分预测肝切除术后肝衰竭的ROC曲线分析: 肝功能Child-Pugh评分和ALBI评分对应ROC曲线下面积分别为0.652(95%CI:0.577~0.727)和0.801(95%CI:0.741~0.861),两者比较,差异有统计学意义(Z=3.590,P<0.05)。肝功能ALBI评分预测肝切除术后肝衰竭的最佳临界值-2.58,灵敏度为68.5%,特异度为86.9%。进一步分析结果显示:排除大范围手术切除对肝切除术后肝衰竭的影响,肝功能Child-Pugh评分和ALBI评分对应ROC曲线下面积分别为0.642(95%CI:0.554~0.731)和0.813(95%CI:0.744~0.882),两者比较,差异有统计学意义(Z=3.407,P<0.05)。(4)随访和生存情况:226例患者中,217例获得术后随访,随访时间为1.3~79.5个月,中位随访时间为29.5个月。随访期间,134例患者生存,92例死亡。(5)影响肝癌患者肝切除术后预后的因素分析。单因素分析结果显示:Alb、PLT、甲胎蛋白(AFP)、腹腔积液、肿瘤直径、手术切除范围、合并门静脉癌栓、合并脉管癌栓、肝功能Child-Pugh评分是影响肝癌患者预后的相关因素[风险比(HR)=0.954,1.003,2.958,1.698,1.155,1.785,2.326,3.001,1.324,95%CI:0.911~0.999,1.000~1.005,1.955~4.476,1.115~2.585,1.103~1.209,1.138~2.802,1.310~4.130,1.983~4.546,1.037~1.690,P<0.05]。多因素分析为了避免共线性,剔除Alb、腹腔积液,结果显示:AFP、肿瘤直径、合并脉管癌栓、肝功能Child-Pugh评分是影响肝癌患者预后的独立因素(HR=2.237,1.080,2.122,1.309,95%CI:1.439~3.476,1.028~1.134,1.362~3.305,1.010~1.697,P<0.05)。Kaplan-Meier曲线进一步分析结果显示:肝功能Child-Pugh A级和B级肝癌患者中位生存时间分别为30.6个月、25.2个月,两者生存情况比较,差异有统计学意义(x2=4.491,P<0.05)。肝功能ALBI 1级和2级肝癌患者中位生存时间分别为29.6个月、31.0个月,两者生存情况比较,差异无统计学意义(x2=0.539,P>0.05)。
结论:术前肝功能ALBI评分预测肝癌患者肝切除术后肝衰竭的价值优于肝功能Child-Pugh评分,但肝功能ALBI评分不是影响肝癌患者肝切除术后预后的独立因素。

【Abstract】

Objective:To investigate the value of the preoperative Child-Pugh score and albumin-bilirubin (ALBI) score predicting posthepatectomy liver failure (PHLF) and prognosis of patients with hepatocellular carcinoma (HCC).
Methods:The retrospective cohort study was conducted. The clinical data of 226 HCC patients who were admitted into the People′s Hospital of Peking University between January 2010 and October 2014 were collected. After preoperative related examinations, feasibility and extent of liver resection were discussed according to patients′ conditions by multidisciplinary team, and then surgery was performed. Observation indicators: (1) surgical situations; (2) factors analysis affecting PHLF of HCC patients; (3) receiver operating characteristic (ROC) curve analysis of Child-Pugh and ALBI scores predicting PHLF; (4) follow-up and survival situations; (5) prognosis analysis of HCC patients after hepatectomy. Follow-up using outpatient examination and telephone interview was performed to detect postoperative survival up to August 2016. Measurement data with normal distribution were represented as ±s. Measurement data with skewed distribution were described as M (P25, P75). The survival curve was drawn by the Kaplan-Meier method, and the Log-rank test was used for survival analysis. The influencing factors of PHLF were analyzed using the logistic regression model. The prognostic factors were analyzed by the COX regression model. ROC analysis was used for predictive value of Child-Pugh and ALBI scores.
Results:(1) Surgical situations: 226 patients underwent successful surgery, including 171 receiving localized liver resection (< 3 Couinaud hepatic segments resection) and 55 receiving extensive liver resection (≥3 Couinaud hepatic segments resection). Volume of intraoperative blood loss of 226 patients was 1-22 550 mL, with a median of 800 mL. Of 226 patients, 89, 9, 4, 4, 3 and 1 were complicated with liver failure, pulmonary infection, bile leakage, gastrointestinal hemorrhage, incision infection and infectious shock, respectively, they were cured and discharged from hospital by life-sustaining treatment and symptomatic treatment. Duration of hospital stay was 2-49 days, with a median of 12 days. (2) Factors analysis affecting PHLF of HCC patients: results of univariate analysis showed that gender, total bilirubin (TBil), albumin (Alb), prothrombin time (PT), international normalized ratio (INR), platelet (PLT), peritoneal effusion, volume of intraoperative blood loss, Child-Pugh score and ALBI score were related factors affecting PHLF of HCC patients [Odds ratio (OR)=0.490, 1.077, 0.763, 1.613, 26.342, 0.990, 2.458, 5.052, 2.875, 34.570, 95% confidence interval (CI): 0.248-0.971, 1.030-1.127, 0.699-0.833, 1.248-2.087, 2.722-254.936, 0.985-0.995, 1.386-4.361, 2.467-10.347, 1.807-4.576, 11.674-102.376, P<0.05]. Child-Pugh score and ALBI score in the multivariate analysis were respectively analyzed using the logistic regression model for avoiding multicollinearity. Excluding TBil, Alb, PT and peritoneal effusion, logistic regression model of Child-Pugh score showed that PLT, volume of intraoperative blood loss and Child-Pugh score were independent factors affecting PHLF of HCC patients (OR=0.991, 4.462, 2.412, 95%CI: 0.986-0.996, 2.090-9.527, 1.479-3.934, P<0.05). Excluding TBil and Alb, Logistic regression model of ALBI score showed that PLT, volume of intraoperative blood loss and ALBI score were independent factors affecting PHLF of HCC patients (OR=0.990, 4.867, 42.947, 95%CI: 0.984-0.995, 2.088-11.346, 12.409-148.637, P<0.05). (3) ROC analysis of Child-Pugh and ALBI scores predicting PHLF: area under the ROC was respectively 0.652 (95%CI: 0.577-0.727) in the Child-Pugh score and 0.801 (95%CI: 0.741-0.861) in the ALBI score, with a statistically significant difference (Z=3.590, P<0.05). The best critical value, sensitivity and specificity of PHLF that were predicted by ALBI score were -2.58, 68.5% and 86.9%, respectively. Further analysis showed that area under the ROC was respectively 0.642 (95%CI: 0.554-0.731) in Child-Pugh score and 0.813 (95%CI: 0.744-0.882) in ALBI score, excluding factors of extensive liver resection affecting PHLF, with a statistically significant difference (Z=3.407, P<0.05). (4) Follow-up and survival situations: of 226 patients, 217 were followed up for 1.3-79.5 months, with a median time of 29.5 months. During the follow-up, 134 patients had survival and 92 died. (5) Prognosis analysis of HCC patients after hepatectomy: results of univariate analysis showed that Alb, PLT, alpha-fetoprotein (AFP), peritoneal effusion, tumor diameter, extent of resection, combined portal vein tumor thrombus (PVTT), combined vascular tumor thrombus and Child-Pugh score were related factors affecting prognosis of HCC patients [Hazard Ratio (HR)=0.954, 1.003, 2.958, 1.698, 1.155, 1.785, 2.326, 3.001, 1.324, 95%CI: 0.911-0.999, 1.000-1.005, 1.955-4.476, 1.115-2.585, 1.103-1.209, 1.138-2.802, 1.310-4.130, 1.983-4.546, 1.037-1.690, P<0.05]. Excluding Alb and peritoneal effusion for avoiding multicollinearity, results of multivariate analysis showed that AFP, tumor diameter, combined vascular tumor thrombus and Child-Pugh score were independent factors affecting prognosis of HCC patients (HR=2.237, 1.080, 2.122, 1.309, 95%CI: 1.439-3.476, 1.028-1.134, 1.362-3.305, 1.010-1.697, P<0.05). Further analysis of Kaplan-Meier curve showed that median survival time in patients with grade A and B of Child-Pugh score were respectively 30.6 months and 25.2 months, with a statistically significant difference in survival (x2=4.491, P<0.05). The median survival time in patients with grade 1 and 2 of ALBI score were respectively 29.6 months and 31.0 months, with no statistically significant difference in survival (x2=0.539, P>0.05).
Conclusions:Preoperative ALBI score in predicting PHLF is superior to that of Child-Pugh score, but ALBI score is not independent factor affecting prognosis of HCC patients.

DOI:10.3760/cma.j.issn.1673-9752.2018.05.011
过刊浏览 优先数字出版

提示:期刊搜索栏可按特定关键词、作者、单位、栏目或日期进行搜索相关期刊。

你也可以直接点击期刊目录进行浏览。

期刊目录
2018 年
2017 年
2016 年
2015 年
2014 年
2013 年
2012 年
2011 年
2010 年
2009 年
2008 年
2007 年