Epidural block associated with improved long-term survival after surgery for colorectal cancer: A retrospective cohort study with propensity score matching / 北京大学学报(医学版)

ABSTRACT

OBJECTIVE@#To investigate the effect of epidural anesthesia on the long-term prognosis of patients after selective colorectal cancer resection surgery.@*METHODS@#This was a retrospective cohort study and approved by local institution review board. Patients who underwent selective colorectal cancer resection surgery from August 2011 to December 2012 in Peking University First Hospital were enrolled. The patients were divided into general anesthesia (GA) group and combined epidural-general anesthesia (EGA) group according to anesthesia type. Primary outcome was patient's long-term survival status. Secondary outcome included the overall incidence of in-hospital complications and length of postoperative in-hospital stay. Propensity score was used to match cases between the two groups based on the probability of receiving EGA. Survival was analyzed by Kaplan-Meier analysis and compared by Log-rank test between the two groups. Multivariate Cox regression analysis was used to investigate the relationship between epidural anesthesia and other variables with long-term survival status.@*RESULTS@#A total of 264 patients were entered into final analysis, including 166 cases in GA group and 98 cases in EGA group. Mean age of the patients was (63.3±12.1) years and mean survival time was 47.2 (95%CI 45.7-48.7) months. Before the propensity score match, the mortality in EGA group was 16.9% (28/166) and 9.2% (9/98) in GA group. But comparison between the two groups had no statistical significance (P=0.091). After the propensity score match, 87 paired cases were matched and analyzed. The risk of long-term mortality in EGA group was lower than that of GA group by Kaplan-Meier analysis (5.7% vs.16.1%, HR=0.344, 95%CI 0.124-0.955, P=0.041). Mean survival time of EGA group was longer than that of GA group (50.3 months vs. 42.9 months, P=0.032). Multivariate Cox regression ana-lysis showed that EGA, in comparison with GA, was related with lower risk of long-term mortality (HR=0.326, 95%CI 0.117-0.909, P=0.032). Age (HR=1.042, 95%CI 1.001-1.085, P=0.046) and preoperative lymph node metastasis (HR=2.924, 95%CI 1.162-7.356, P=0.023) were also related with increased risk of long-term mortality.@*CONCLUSION@#Present study found that perioperative use of epidural anesthesia and analgesia was associated with improvement of the patient's long-term survival. Well-designed studies are needed to verify this hypothesis.