|
|
Back to 2018 Program and Abstracts
LAPAROSCOPIC VERSUS OPEN RESECTION FOR HEPATOCELLULAR CARCINOMA IN PATIENTS WITH ADVANCED CIRRHOSIS: A PROPENSITY SCORE MATCHING ANALYSIS OF 1799 PATIENTS.
Onur Kutlu2, Eduardo A. Vega*1, Masuyuki Okuno1, Katharina Joechle1, Nestor de La Cruz2, Kanwal Raghav3, Ahmed Kaseb3, Yun Shin Chun1, Ching-Wei Tzeng1, Jean-Nicolas Vauthey1, Claudius Conrad1
1Surgical Oncology, MD Anderson Cancer Center, Houston, TX; 2Surgery, Miller School of Medicine, Miami, FL; 3Gastrointestinal Medical Oncology, MD Anderson Cancer Center, Houston, TX
Background: Mortality and survival after laparoscopic (LLR) vs. open (OLR) liver resection remain unclear in the cohort of patients with advanced cirrhosis. This study aims at a comparative analysis of long- and short term outcome of LLR vs. OLR for HCC in significantly cirrhotic patients with special consideration for selection bias.
Methods: Patients who underwent resection for HCC, years 2010-2014 were queried in the National Cancer Data Base (NCDB). Included were patients with a single lesion, M0 disease, known grade, margin status, tumor size, hospital setting, with severe fibrosis (IASL score >4), hospital stay, 30-day and 90-day mortality, 30-day readmission rates, surgical approach, and complete follow-up information. A 1:1 matched propensity score analysis between LLR and OLR groups was performed. Eight preoperative variables [age, sex, facility type, grade, margin status, tumor size, nodal status, major hepatectomy (minor <=2 segments, major>=3 segments)] were matched. Comparison of categorical variables, 30-day and 90-day mortality, 30-day unplanned readmission were performed by chi square test. Hospital stay was performed by Mann-Whitney U test. The prognostic effect of LLR was assessed using a multivariable Cox proportional hazards model.
Results: 1799 patients met inclusion criteria. Of these, 491 (27.3%) underwent minor while 1308 (72.7%) underwent major hepatectomy. There were 193 (10.7%) patients underwent LLR (minor n=130, 26.5%; major n=63, 3.5%), 190 patients were eligible for 1:1 matching with no difference seen for patient characteristic, resection margin, 30-day mortality (p=0.141), 90-day mortality (p=0.063), 30-day readmission (p=0.784) between LLR and OLR groups. Median hospital stay for the LLR group was significantly shorter than in the OLR group (6 vs 8 days, p=0.001). There was no difference in overall survival between LLR and OLR group (median, 35.3 vs 34.9 months, p=0.215). Age (HR 1.012, p=0.034), comorbidities (HR 1.290, p=0.012), grade 4 (HR 1.810, p=0.025), N1 disease (HR 1.041, p=0.048), R1 margins (HR 1.339, p=0.002) were found to be risk factor on overall survival. After adjusting for cofounders, LLR vs OLR was not significant risk factor on survival (hazard ratio (HR) 0.871, 95% CI 0.674-1.223, p=0.373).
Conclusion: LLR vs OLR in cirrhotic patients with HCC leads to the same oncologic outcome and mortality. Further, the presented data confirms that selection for resection according to age, tumor characteristics and comorbidities rather than the approach is the main determinant of outcome.
Back to 2018 Program and Abstracts
|
