A high value of serum des-γ-carboxy prothrombin before hepatocellular carcinoma treatment can be associated with long-term liver dysfunction after treatment
A high value of serum des-γ-carboxy prothrombin before hepatocellular carcinoma treatment can be associated with long-term liver dysfunction after treatment
复制标题
DOI：
10.1007/s00535-012-0577-0
复制
发表时间：
2012-10-01
影响因子：
6.3
通讯作者：
Azuma, Takeshi
中科院分区：
文献类型：
--
作者：
Saito, Masaya;Seo, Yasushi;Azuma, Takeshi
肝动脉化疗栓塞术(TACE)是治疗肝细胞癌的有效方法，但有时会使肝功能恶化。TACE前对TACE后肝功能障碍的预测将有助于避免长期的肝功能障碍。我们对在神户大学医院接受TACE的100名肝癌患者进行了一项病例对照研究。术前行尿/血生化检查。作为肝功能的指标，在TACE前和术后3个月对肝功能进行Child‘s评分。比较儿童评分增加2分及以上的病例，并对独立危险因素进行统计学检验。通过Logistic回归分析，建立了介入治疗前儿童评分增加2分以上的预测模型，单因素分析显示，介入治疗前儿童评分恶化组的DES-γ-羧基原凝血酶和乳酸脱氢酶显著高于未恶化组(p=0.036和0.003)。所有可能的多元回归分析显示，腰椎侧凸(p=0.003)和腰椎脱位(p=0.002)是影响Child‘s分级恶化的独立因素。建立了预测模型：EXP(0.014 x Ldh+0.572 x ln(Dcp)-8.655)/(1+EXP(0.014 x Ldh+0.572 x ln(Dcp)-8.655))。该模型具有较好的区分性，AUC值为0.837(95%可信区间[CI]0.662-1.000)。其最佳临界点为0.073，其敏感度和特异度分别为90.9%和69.7%，术前血清乳酸脱氢酶和乳酸脱氢酶升高与肝功能长期恶化有关。我们的治疗前预测模型可能有助于识别高危病例。
Transcatheter arterial chemoembolization (TACE) is an effective treatment for hepatocellular carcinoma (HCC), but it sometimes makes liver function worse. The pre-TACE prediction of liver dysfunction after TACE would be helpful to avoid long-term liver dysfunction.We performed a case-control study in 100 HCC patients who underwent TACE at Kobe University Hospital. Urinary/blood biochemical examinations were performed before TACE. As an indicator of liver function, Child's score was also evaluated before and 3 months after TACE. Cases with and without an increase of 2 points or more in the Child's score were compared, and independent risk factors were statistically examined. A pre-TACE predictive model of an increase of 2 points or more in the Child's score after TACE was developed using logistic regression.Univariate analyses showed that des-gamma-carboxy prothrombin (DCP) and lactate dehydrogenase (LDH) before TACE were significantly higher in the Child's score-deteriorated group than in the group with no deterioration (p = 0.036 and 0.003, respectively). All possible multivariate regressions showed that DCP (p = 0.003) and LDH (p = 0.002) were independent factors determining the deterioration of Child's class. A predictive model was developed, as follows: exp(0.014 x LDH + 0.572 x ln(DCP) - 8.655)/(1 + exp(0.014 x LDH + 0.572 x ln(DCP) - 8.655)). The model discriminated well, with AUC being 0.837 (95 % confidence interval [CI] 0.662-1.000). The optimal cut-off point was 0.073, and the sensitivity and specificity were 90.9 and 69.7 %, respectively.High values of DCP and LDH before TACE were associated with the long-term deterioration of liver function. Our pre-therapeutic prediction model could be useful to identify high-risk cases.