Abstract: Background and purpose: Pancreatic cancer is a devastating disease. Even after radical resection of pancreatic cancer, the prognosis still remains poor. To better understand the effects of perioperative anesthesia strategy on improving long-term survival, this retrospective analysis investigated associations between perioperative factors and survival in patients undergoing pancreatic cancer surgery. Methods: Survival data ［overall survival (OS)］ and perioperative clinicopathological parameters for 190 patients who underwent radical resection of pancreatic cancer between December 2011 and January 2016 were obtained in Fudan University Shanghai Cancer Center. The last follow-up time was January 2019. The authors performed univariate and multivariate analyses using the COX proportional hazards regression model. Results: Univariate analyses revealed that preoperative CA19-9 level and neutrophil-lymphocyte ratio (NLR), TNM stage, differentiation, maximum size of tumor, lymph node metastasis, intraoperative blood loss and transfusion were significantly associated with OS in patients with radical resection of pancreatic cancer. Dexamethasone also showed a little association with OS (P=0.052, HR=0.73). Differentiation (P=0.001, HR=0.59), maximum size of tumor (P=0.039, HR=1.51), lymph node metastasis (P=0.003, HR=1.61) and intraoperative transfusion (P=0.046, HR=1.39) were identified as significant associations with OS by multivariate analyses. The median survival time (15.7 months vs 23.3 months) of transfused patients was shorter than that of transfusion-free patients. Conclusion: Perioperative anesthesia strategy is very important to protect immune functions and reduce risks of recurrence and metastasis in pancreatic cancer patients. It is necessary for us to choose reasonable anesthetic techniques and agents, optimize anesthesia management and establish an individual transfusion scheme in order to improve long-term survival.

Key words: Pancreatic cancer, Prognosis, Perioperative period, Anesthesia