Long-term and perioperative outcomes of laparoscopic versus open liver resection for colorectal liver metastases with propensity score matching: A multi-institutional Japanese study

Background The aim of the present study was to clarify the surgical outcome and long-term prognosis of laparoscopic liver resection (LLR) compared with conventional open liver resection (OLR) in patients with colorectal liver metastases (CRLM). Methods A one-to-two propensity score matching (PSM) analysis was applied. Covariates (P < 0.2) used for PSM estimation included preoperative levels of CEA and CA19-9; primary tumor differentiation; primary pathological lymph node metastasis; number, size, location, and distribution of CRLM; existence of extrahepatic metastasis; extent of hepatic resection; total bilirubin and prothrombin activity levels; and preoperative chemotherapy. Perioperative data and long-term survival were compared. Results From 2005 to 2010, 1,331 patients with hepatic resection for CRLM were enrolled. By PSM, 171 LLR and 342 OLR patients showed similar preoperative clinical characteristics. Median estimated blood loss (163 g vs 415 g, P < 0.001) and median postoperative hospital stay (12 days vs 14 days; P < 0.001) were significantly reduced in the LLR group. Morbidity and mortality were similar. Five-year rates of recurrence-free, overall, and disease-specific survival did not differ significantly. The R0 resection rate was similar. Conclusions In selected CRLM patients, LLR is strongly associated with lower blood loss and shorter hospital stay and has equivalent long-term survival comparable with OLR. Laparoscopic resection of colorectal liver metastases is rapidly gaining worldwide acceptance. In the largest and most rigorous propensity score matched study to date comparing laparoscopic and open liver resection, Beppu and colleagues clearly demonstrated the excellent perioperative benefits of the laparoscopic approach, without compromising oncologic outcomes or long-term survival. Table 2 Perioperative outcome of colorectal liver metastases (CRLM) patients who underwent laparoscopic liver resection (LLR) and open liver resection (OLR): the overall cohort and propensity score matching (PSM) cohort Overall cohort (n = 1,331) PSM cohort (n = 513) LLR (n = 210) OLR (n = 1,121) P-value LLR (n = 171) OLR (n = 342) P-value Operation time (min) median (range) 281 (60-1120) 312 (39-3350) 0.020 282 (60-1120) 277 (40-1343) 0.130 Blood loss (g), median (range) 160 (0-3355) 500 (0-11240) <0.001 163 (0-3355) 405 (0-11240) <0.001 Blood loss ≥ 1,000 (g) (yes, no) 15: 192 211: 813 <0.001 11: 159 55: 278 0.004 pRBC administration (%) 8.7 16.6 0.004 8.4 12.8 0.148 R0 190 929 154 315 R1 9 60 9 8 0.899 0.120 R2 1 7 1 1 R0+ablation 4 26 4 6 Pathological surgical margin (mm), median (range) 5(0-40) 5(0-50) 0.203 5(0-40) 5(0-45) 0.963 Morbidity (%) 13.0 13.4 >0.999 14.1 12.7 0.631 Mortality (%) within 1 month 0.0 0.1 >0.999 0.0 0.0 N.A. within 3 months 0.0 0.5 >0.999 0.0 0.6 N.A. Postoperative hospitalization (days), median (range) 12 (1-192) 16 (2-745) <0.001 12 (3-192) 14 (4-174) <0.001 N.A. not applicable, pRBC packed red blood cell Clinical parameters were compared with; Overall cohort: Wilcoxon rank-sum (Mann-Whitney) test for ordinal, and Fisher's exact test for categorical data. PSM cohort: 5-stratified conditional logistic regression for ordinal, and 5-stratified conditional logistic regression for categorical.