Positive association of gastric cancer surgery outcome with surgeon specialization in a Shanghai high-volume general hospital.

4092 Background: Numerous studies suggest positive relationship between hospital volume and cancer treatment outcomes, the surgeon’s experience and specialty training may also be important. This was examined in a high volume hospital in Shanghai among patients who underwent gastric cancer (GC) surgery. Methods: Data on consecutive patients (pts) undergoing R0 or R1 GC resection in Zhongshan hospital between January 2003 and June 2010 were collected and analyzed. Follow-up on pts who were non-Shanghai residents were less complete therefore excluded. Post-operative mortality, pathologic results and survival outcome for pts treated by surgical training, i.e., sub-specialized vs., non-specialized, were obtained. Survival was calculated by the Kaplan-Meier method and Log-rank test was used to determine statistical significance. To determine whether sub-specialty surgical training was an independent factor for overall survival (OS), univariate and multivariate analyses were performed using Cox proportional hazards regression. Results: Total 5,046 pts underwent R0 or R1 GC resection were identified.1594 pts had complete covariate data, survival information and were included in the study. Of them, the sub-specialized group included 217 cases treated by 3 surgeons, while the non-specialized group included 1377 cases treated by 52 surgeons. 5-year cumulative OS was higher in the sub-specialized group (62.9% vs. 54.6%, p=0.032). Multivariate analysis showed that tumor stage(p<0.001), location of tumor (p=0.003), vascular invasion (p<0.001) and surgeon (HR=1.54, p=0.001) were all associated with OS. The incidence of positive margin was higher in non-specialized group (2.0% vs. 2.7%, p<0.001) and the probability of retrieved lymph nodes less than 15 was more in non-specialized group (25.9% vs. 7.3%, p<0.001). Postoperative mortality was also higher in non-specialized group than in specialized group(1.5% vs. 0.9%, p<0.001). Conclusions: In high volume general hospital, sub-specialty training is desirable in gastric cancer surgery, the quality of gastric cancer surgery can be further improved by sub-specialty training leading to better treatment outcome.