Aim Epidural analgesia reduces the surgical stress response. However, its effect on pro- and anti-inflammatory cytokines in the genesis of inflammation following major abdominal surgery remains unclear. Our main objective was to elucidate whether perioperative epidural analgesia prevents the inflammatory response following colorectal cancer surgery. Methods Ninety-six patients scheduled for open or laparoscopic surgery were randomized to epidural analgesia (group E) or patient-controlled intravenous analgesia (group P). Surgery and anaesthesia were standardized in both groups. Plasma cortisol, insulin and serum cytokines [interleukin 1 beta (IL-1 beta), IL-4, IL-5, IL-6, IL-8, IL-10, IL-12p70, IL-13, tumour necrosis factor , interferon , granulocyte-macrophage colony-stimulating factor, prostaglandin E-2 and vascular endothelial growth factor] were measured preoperatively (T0), 1-6h postoperatively (T1) and 3-5days postoperatively (T2). Mixed model analysis was used, after logarithmic transformation when appropriate, for analyses of cytokines and stress markers. Results There were no significant differences in any serum cytokine concentration between groups P and E at any time point except for IL-10 which was 87% higher in group P [median and range 4.1 (2.3-9.2) pg/ml] compared to group E [2.6 (1.3-4.7) pg/ml] (P = 0.002) at T1. There was no difference in plasma cortisol and insulin between the groups at any time point after surgery. A significant difference in median serum cytokine concentration was found between open and laparoscopic surgery with higher levels of IL-6, IL-8 and IL-10 at T1 in patients undergoing open surgery compared to laparoscopic surgery. No difference in serum cytokine concentration was detected between the groups or between the surgical technique at T2. Conclusions Open surgery, compared to laparoscopic surgery, has greater impact on these inflammatory mediators than epidural analgesia vs intravenous analgesia.