Clinical Research Papers:
Liver resection versus transplantation for multiple hepatocellular carcinoma: a propensity score analysis
Metrics: PDF 779 views | HTML 1213 views | ?
Chuan Li1, Jia-Ye Liu1, Wei Peng1, Tian-Fu Wen1, Lu-Nan Yan1, Jia-Yin Yang1, Bo Li1, Wen-Tao Wang1 and Ming-Qing Xu1
1 Department of Liver Surgery & Liver Transplantation, West China Hospital of Sichuan University, Chengdu, China
Tian- Fu Wen, email:
Keywords: liver transplantation, liver resection, hepatocellular carcinoma, outcomes, recurrence
Received: November 14, 2016 Accepted: August 06, 2017 Published: September 02, 2017
The aim of this study was to compare the outcomes of patients with multiple hepatocellular carcinoma (HCCs) after liver resection (LR) versus liver transplantation (LT). Patients who had multiple HCCs without macrovascular invasion and who underwent LT or LR between 2007 and 2013 were reviewed. A propensity score matching model was used to adjust baseline differences between the two groups. A total of 204 patients were selected for the current study, including 137 LR patients and 67 LT patients. During follow-up, 100 patients experienced recurrence, and 78 patients died. The 5-year recurrence-free survival rate was 71.1% for the LT group and 31.1% for the LR group (P<0.001). The 5-year overall survival rate was 73.4% for the LT group and 39.8% for the LR group (P<0.001). Moreover, the LT group had better recurrence-free survival and overall survival rates than the LR group regardless of whether the patients met or exceeded the Milan criteria. The multivariate analysis showed that microvascular invasion and LR were independent risk factors for postoperative recurrence, whereas only LR was associated with an increased incidence of mortality. After applying one-to-one propensity score matching, similar results were observed in the propensity score matching model. Our study suggested that LT provided a better prognosis for patients with multiple HCCs than LR regardless of whether the patients met the Milan criteria.
All site content, except where otherwise noted, is licensed under a Creative Commons Attribution 3.0 License.