P-166 Re-hepatectomy is a Predictor of Long-Term Survival in Patients That Recur after Liver Resection of Colorectal Liver Metastases

Introduction: Liver recurrence is a frequent event in the evolution of patients with resected colorectal liver metastases and re-hepatectomy (RH) has been increasingly used worldwide. In this series, we analyze early and late postoperative outcomes of patients submitted to RH in a referral Brazilian oncological center. Methods: This is a retrospective analysis of a prospective database from 1998 to 2012. Inclusion criteria were liver resection in patients who already had received a hepatectomy. Data was compared with patients submitted to initial hepatectomy (IH). Second stage procedures were not included. Survival from the date of RH was compared with survival in the IH group. R2 resections and postoperative deaths were excluded of the survival analyses. Results: In the study period 346 liver resections were performed in 244 patients, 53 RH and 277 IH. Two patients received 2 RH and 1 patient received 3, totalizing 49 patients. One patient did not achieve a macroscopic complete resection (R2). Ninety-day morbidity and mortality in this group was, respectively, 45.2% and 1.8%. Four patients were excluded from survival analyses due to lack of information regarding the previous liver resection. RH group mostly received minor resections than IH group (69.1% x 45.1%, p = 0.001) and had shorter hospital stay (7d x 10d, p = 0.008). There was no difference regarding surgical duration (312min x 323min, p = 0.521), blood transfusion (17.3% x 20.7%, p = 0.559), morbidity (58.8% x 45.2%, p = 0.068) and mortality (3.6% x 1.8%, p = 1.000) in both groups. Calculating overall survival (OS) from the first liver resection, with a median follow up period of 37 months, 3-year OS in the RH group was 85.7% with a median survival of 76 months from the RH. This survival is inferior to the one observed in patients without recurrence and better than observed in patients that had recurrence and were not amenable to liver resection (85.7% x 93.4% x 60.9%, p < 0.001). Three-year OS from RH and IH did not differ (82.8% x 72.6%, p = 0.643), however, DFS was inferior in the RH group (16.3% x 35.8%, p = 0.025). In the relapsed patients group, performance of RH was an independent predictor of OS (HR = 2.7, p = 0.002, 95% CI 1.4 – 5.0). None factor influenced overall survival from RH. Patients in this group who received preoperative chemotherapy had inferior DFS (42.4% x 16.7%, p = 0.009) as well as did patients who had a R1 resection (39.5% x 0, p = 0.005). Both factors were independent on multivariate analyses. Conclusion: RH provides similar OS to the one observed in patients submitted to IH and without higher morbidity and mortality. Even with an inferior DFS, it seems to be beneficial in selected patients who experience liver recurrence.