College of Medicine, Chang Gung University, Taoyuan 333, Taiwan.
World J Gastroenterol. 2013 Apr 28;19(16):2466-72. doi: 10.3748/wjg.v19.i16.2466.
To investigate outcomes of hepatocellular carcinomas (HCCs) in patients with chronic kidney disease (CKD).
Four hundred and forty patients referred between 2000 and 2002 for management of HCCs were categorized according to their CKD stage, i.e., estimated glomerular filtration rate (eGFR) > 90 (stage 1), 60-90 (stage 2), 30-60 (stage 3), 15-30 (stage 4), and < 15 (stage 5) mL/min per 1.73 m², respectively. Demographic, clinical and laboratory data were collected and mortality rates and cause of mortality were analyzed. The mortality data were examined with Kaplan-meier method and the significance was tested using a log-rank test. An initial univariate Cox regression analysis was performed to compare the frequency of possible risk factors associated with mortality. To control for possible confounding factors, a multivariate Cox regression analysis (stepwise backward approach) was performed to analyze those factors that were significant in univariate models (P < 0.05) and met the assumptions of a proportional hazard model.
Most HCC patients with CKD were elderly, with mean age of diagnosis of 60.6 ± 11.9 years, and mostly male (74.8%). Hepatitis B, C and B and C co-infection virus were positive in 61.6%, 45.7% and 14.1% of the patients, respectively. It was found that patients with stages 4 and 5 CKD were not only older (P = 0.001), but also had higher hepatitis C virus carrier rate (P = 0.001), lower serum albumin level (P = 0.001), lower platelet count (P = 0.037), longer prothrombin time (P = 0.001) as well as higher proportions of advanced cirrhosis (P = 0.002) and HCCs (P = 0.001) than patients with stages 1 and 2 CKD. At the end of analysis, 162 (36.9%) patients had died. Kaplan-Meier analysis revealed that patients with stages 4 and 5 CKD suffered lower cumulative survival than stages 1 and 2 CKD (log-rank test, χ² = 11.764, P = 0.003). In a multivariate Cox-regression model, it was confirmed that CKD stage [odds ratio (OR) = 1.988, 95%CI: 1.012-3.906, P = 0.046)], liver cirrhosis stage (OR = 3.571, 95%CI: 1.590-8.000, P = 0.002) and serum albumin level (OR = 0.657, 95%CI: 0.491-0.878, P = 0.005) were significant predictors for mortality in this population.
HCC patients with stages 4 and 5 CKD had inferior survival than stages 1 and 2 CKD. This warrants further studies.
探讨慢性肾脏病（CKD）患者肝细胞癌（HCC）的结局。
将 2000 年至 2002 年间因 HCC 管理而就诊的 440 例患者根据 CKD 分期进行分类，即估算肾小球滤过率（eGFR）>90（1 期）、60-90（2 期）、30-60（3 期）、15-30（4 期）和<15（5 期）mL/min/1.73m²，分别。收集人口统计学、临床和实验室数据，并分析死亡率和死亡原因。使用 Kaplan-Meier 法检查死亡率数据，并使用对数秩检验检验其显著性。进行初始单变量 Cox 回归分析，以比较与死亡率相关的可能危险因素的频率。为了控制可能的混杂因素，使用多变量 Cox 回归分析（逐步后退法）分析单变量模型中显著的因素（P<0.05），并满足比例风险模型的假设。
大多数患有 CKD 的 HCC 患者年龄较大，诊断时的平均年龄为 60.6±11.9 岁，且大多数为男性（74.8%）。乙型肝炎、丙型肝炎和乙型肝炎合并丙型肝炎病毒阳性率分别为 61.6%、45.7%和 14.1%。结果发现，4 期和 5 期 CKD 患者不仅年龄较大（P=0.001），而且丙型肝炎病毒携带者比例较高（P=0.001），血清白蛋白水平较低（P=0.001），血小板计数较低（P=0.037），凝血酶原时间较长（P=0.001），晚期肝硬化（P=0.002）和 HCC 比例较高（P=0.001）。在分析结束时，162 名（36.9%）患者死亡。Kaplan-Meier 分析显示，4 期和 5 期 CKD 患者的累积生存率低于 1 期和 2 期 CKD（对数秩检验，χ²=11.764，P=0.003）。在多变量 Cox 回归模型中，证实 CKD 分期[优势比（OR）=1.988，95%CI：1.012-3.906，P=0.046]、肝硬化分期（OR=3.571，95%CI：1.590-8.000，P=0.002）和血清白蛋白水平（OR=0.657，95%CI：0.491-0.878，P=0.005）是该人群死亡的显著预测因素。
与 1 期和 2 期 CKD 相比，4 期和 5 期 CKD 的 HCC 患者生存情况较差。这需要进一步研究。
