Oncological safety of ultrasound-guided laparoscopic liver resection for colorectal metastases: a case-control study.

Laparoscopic liver surgery has gained widespread acceptance and nowadays it is suggested even for malignant disease. Although the benefits on short-term outcomes have been proven, data on oncological safety are still lacking. The aim of this study is to assess oncologic results after ultrasound-guided laparoscopic liver resection (LLR) or open liver resection (OLR) for colorectal metastases. 37 consecutive patients undergoing LLR between 01/2004 and 03/2014 were matched at a ratio of 1:1 with 37 OLR. Matching criteria were male sex, number and diameter of liver metastases, segment location, synchronous presentation, site and stage of primary tumor, positive lymph nodes of the primary, and concomitant extrahepatic disease. Demographic characteristics were similar among groups. Parenchymal transection time was longer in the LLR group (68 ± 38.2 SD vs 40 ± 33.7 SD, p = 0.01). Mortality was nil in LLR and OLR. Overall morbidity was significantly lower in LLR (13.5 vs 37.8%, p = 0.02), although severe complications were similar among the two groups. Patients undergoing LLR were discharged earlier (5 ± 2.3 SD vs 8 ± 6.6 SD days, p < 0.001). The median margin width was 5 (0-40) mm in LLR vs 8 (0-25) mm in OLR, p = 0.897. R1 resection was recorded in four LLR and three OLR (p = 1). Overall recurrences were similar among groups. Eight patients with hepatic or extrahepatic recurrence among LLR underwent surgery vs four of OLR (p = 0.03). After a median follow-up of 35.7 months in LLR and 47.9 months in OLR, 3-year overall survival was 91.8% LLR and 74.8% OLR (p = 0.14). 3-year disease-free survival was 69.1% LLR and 65.9% OLR (p = 0.53). Multivariate analysis showed that postoperative complications [HR 3.42 (95% CI 1.32-8.89)] and multiple metastases [HR 3.84 (95% CI 1.34-10.83)] were independent predictors of worse survival (p = 0.01). Ultrasound-LLR for colorectal hepatic metastases is safe, ensuring oncologic outcomes comparable to OLR.