探讨肝细胞癌（肝癌）破裂亚临床出血患者择期肝切除术后的预后。

回顾性分析1997年1月至2010年12月在青岛大学医学院附属医院肝胆外科行择期肝切除术的614例肝癌患者临床资料。所有患者均签署知情同意书，符合医学伦理学规定。根据有否出现肝癌破裂亚临床出血，将患者分为破裂组和非破裂组。其中破裂组30例，男25例，女5例，年龄29～73岁，中位年龄55岁；非破裂组584例，男488例，女96例，年龄14～82岁，中位年龄55岁。所有患者均行肝切除术。观察破裂组与非破裂组患者的临床特征。患者术后均接受随访，观察患者存活情况、死亡原因。根据随访结果绘制Kaplan-Meier生存曲线，比较两组患者的存活情况。分析临床参数与破裂组患者预后的关系。两组临床参数比较采用χ²检验或Fisher确切概率法。组间多因素分析采用Logistic回归分析。生存分析和比较采用Kaplan-Meier法和Log-rank检验，预后危险因素分析采用Cox比例风险回归模型分析。

破裂组中非完全切除、肝切除范围>1段、手术切缘<0.5 cm、术中输血、术中出血量≥1 000 ml、肿瘤直径>5 cm和肿瘤组织学低分化者明显多于非破裂组（χ²=37.609，6.021，6.497，20.982，19.498，22.944，8.222；P<0.05）。进一步Logistic回归分析显示，非完全切除和肿瘤直径>5 cm是肝癌破裂亚临床出血的独立危险因素（OR=3.772，5.045；P<0.05）。随访期间破裂组死亡26例，非破裂组死亡316例。Kaplan-Meier分析显示，破裂组的中位生存期为9个月，非破裂组为56个月；破裂组患者肝切除术后的1、3、5年累积生存率分别为50.0%、22.2%、5.9%，非破裂组为86.0%、62.6%、48.9%，差异有统计学意义（χ²=38.879，P<0.05）。Cox比例风险回归模型分析显示，肿瘤组织学低分化是破裂组患者肝切除术后预后的独立危险因素（RR=3.736，P<0.05）。

非完全切除和肿瘤直径>5 cm是肝癌破裂亚临床出血的独立危险因素。肝癌破裂亚临床出血患者择期肝切除术后预后仍较差，肿瘤组织学低分化是影响其预后的独立危险因素。

To discuss the prognosis of ruptured hepatocellular carcinoma (HCC) patients with subclinical bleeding after elective hepatectomy.

Clinical data of 614 patients with HCC who underwent elective hepatectomy in Department of Hepatobiliary Surgery, the Affiliated Hospital of Medical College, Qingdao University from January 1997 to December 2010 were analyzed retrospectively. The informed consents of all patients were obtained and the ethical committee approval was received. According to whether the patients had subclinical bleeding following ruptured HCC or not, they were devided into ruptured group (n=30; 25 males, 5 females; 29 to 73 of age, 55 of median age) and non-ruptured group (n=584; 488 males, 96 females; 14 to 82 of age, 55 of median age). All the patients underwent hepatectomy. Clinical characters of patients in ruptured group and non-ruptured group were observed. All the patients received follow-up after operations, the survival and causes of death were observed. The Kaplan-Meier survival curves were drawn according to the results of follow-up, and the survival of 2 groups was compared. The relation between the clinical parameters and the prognosis of patients in ruptured group was analyzed. The clinical parameters in 2 groups were compared using Chi-square test or Fisher’s exact probability method. Multiple factors analysis between groups was conducted using Logistic regression analysis. Survival analysis and comparison were conducted using Kaplan-Meier method and Log-rank test. Prognostic risk factor analysis was conducted using Cox proportional hazards regression model analysis.

The patients with non-radical resection, liver resection range > 1 segment, surgical margin < 0.5 cm, intraoperative transfusion, intraoperative blood loss ≥ 1 000 ml, tumor diameter > 5 cm and poorly differentiated tumor in ruptured group were significantly more than those in non-ruptured group (χ²=37.609, 6.021, 6.497, 20.982, 19.498, 22.944, 8.222; P<0.05). Further Logistic regression analysis showed that non-radical resection and tumor diameter > 5 cm were the independent risk factors for ruptured HCC with subclinical bleeding (OR=3.772, 5.045; P<0.05). There were 26 deaths in ruptured group and 316 deaths in non-ruptured group during the follow-up. Kaplan-Meier analysis showed that the median survival time was 9 months in ruptured group and 56 months in non-ruptured group. The 1-, 3-, 5-year accumulative survival rate were 50.0%, 22.2%, 5.9% in ruptured group and 86.0%, 62.6%, 48.9% in non-ruptured group respectively, where significant difference was observed (χ²=38.879, P<0.05). Cox proportional hazards regression model analysis showed that poorly differentiated tumor was an independent risk factor for patients in ruptured group after hepatectomy (RR=3.736, P<0.05).

Non-radical resection and tumor diameter > 5 cm are the independent risk factors for ruptured HCC with subclinical bleeding. The prognosis of ruptured HCC patients with subclinical bleeding after elective hepatectomy is still poor, and poorly differentiated tumor is an independent risk factor for it.