肝硬化患者肝内LR-3类结节恶变风险评估及预测模型的建立*

目的 探讨存在LR-3类肝内结节的肝硬化患者恶性转变的危险因素,并在此基础上建立恶性转变预测模型。方法 2014年3月～2016年3月我院诊治的124例存在肝内LR-3类结节的肝硬化患者,规律随访进行磁共振成像(MRI)检查,对可疑病灶进行穿刺活检。应用多因素Logistic回归分析影响癌变的危险因素,并建立预测模型。结果 经随访60～78个月(平均74个月),120例患者完成随访,发现癌变24例(20.0%),另96例无癌变;癌变组有饮酒、吸烟、合并糖尿病、乙型肝炎病毒感染和酒精性肝炎者分别为91.7%、87.5%、75.0%、100.0%和41.7%,显著高于未癌变组(分别为45.8%、52.1%、41.7%、58.3%和20.8%,P<0.05);Cox风险回归分析显示,饮酒(HR=2.264,95%CI=1.597～3.210,P<0.001)、糖尿病(HR=1.294,95%CI=1.107～1.513,P=0.001)、乙型肝炎(HR=1.795,95%CI=1.329～2.561,P<0.001)和酒精性肝炎(HR=0.658,95%CI=0.552～0.784,P<0.001)是存在LR-3类病变结节的肝硬化患者癌变的独立危险因素;根据Cox多因素分析结果建立癌变预测模型,即Y=0.817X₁+0.258X₂+0.585X₃-0.419X₄(X₁=饮酒,X₂=糖尿病,X₃=乙型肝炎,X₄=酒精性肝炎)。经ROC分析显示,该模型判断LR-3类病变结节患者癌变风险的AUC为0.812(SE=0.064,95%CI=0.687～0.936,P<0.001),其敏感度为0.833,特异度为0.673。结论 了解危险因素可能对判断肝硬化患者肝内结节的性质有帮助,加强随访,必要时穿刺才是明确诊断的正确方法。

Qualitative evaluation of intrahepatic nodules in patients with liver cirrhosis

Objective The purpose of this study was to explore the qualitative evaluation of intrahepatic nodules in patients with liver cirrhosis. Methods A retrospective analysis was performed on 124 cirrhotics with liver imaging reporting and data system (LI-RADS) 3 intrahepatic nodules. All, but fourpatients were followed-up for 60-78 (mean 74) months, and the MRI was conducted regularly. The fine needle aspiration liver biopsies werecarried out when necessary. The multivariate Logistic analysis was performed to reveal the risk factors. Results At the end of follow-up, 4 patients lost, and the cancerous foci were found in 24 cases (20.0%); the percentages of alcohol-taking, cigarettes smoking, concomitant diabetes, hepatitis B viral infections and alcoholic hepatitis in patients with intrahepatic cancerous lesions were 91.7%, 87.5%, 75.0%, 100.0% and 41.7%, allsignificantly higher than 45.8%, 52.1%, 41.7%, 58.3% and 20.8% (P<0.05) in patients without cancerous lesions; the Cox analysis showed that the alcohol-taking(HR=2.264, 95%CI=1.597-3.210, P<0.001), diabetes (HR=1.294, 95%CI=1.107-1.513, P=0.001), hepatitis B (HR=1.795, 95%CI=1.329- 2.561, P<0.001) and alcoholic hepatitis (HR=0.658, 95%CI=0.552-0.784, P<0.001) were the independent risk factors for patients with LR-3 intrahepatic foci; the formula was established based on the Cox analysis results, e.g.,Y=0.817X₁+0.258X₂+0.585X₃-0.419X₄(X₁=alcohol-taking, X₂=diabetes, X₃=hepatitis B, X₄=alcoholic hepatitis), and the ROC analysis demonstrated that the AUC was 0.812(SE=0.064, 95%CI=0.687-0.936, P<0.001), with the sensitivity of 0.833 and the specificityof 0.673, for liver cirrhotics with intrahepatic LR-3 nodules. Conclusion It might be important to follow-up liver cirrhosis patients with intrahepatic LR-3 nodules, especially in those with risk factors for malignant transformation, which could find cancerous lesions as soon as possible.