Our hypothesis was that, due to its sympatholytic action, epidural anesthesia (EA) administered as part of anesthesia in abdominal surgery would generate a marked venous leg flow enhancement, thus aiding in the prevention of peroperative venous stasis. We studied, and comprehensively quantified the venous haemodynamic changes in the lower limb during and immediately after abdominal surgery performed under EA and general (GA) anesthesia combined, in comparison to GA alone. This is a prospective, randomized, controlled study, stratified for hypertension and smoking, comprising ASA 1-2 patients undergoing elective total abdominal hysterectomy. Those with peripheral vascular or chronic venous disease, prior DVT or BMI>35 were excluded. Eligible recruits received either GA (Group GA) (n = 10; age 36-65, median 50) alone or epidural anesthesia (EA) and GA combined (Group EA/GA) (n = 9; age 32-58, median 46). EA (L(1-2)) was administered using lignocaine 2%. Both groups had GA induced with fentanyl and propofol, maintained with N(2)O and isoflurane; larygoscopy was facilitated with vecuronium; analgesia was provided either with morphine (Group GA) or epidurally with 2% lignocaine boli (Group EA/GA). Hemodynamics were determined at the popliteal vein in the horizontal supine position at baseline (resting prior to anesthesia), post epidural (20 min after delivery of EA), post induction (15 min after laryngeal intubation), surgery (upon uterus removal) and recovery (30 min after extubation). There was no difference in the mean velocity[V(mean)] between the 2 groups at baseline (p = 0.35([Mann-Whitney])), and post induction (p = 0.5([Mann-Whitney])). However V(mean) was significantly higher in Group EA/GA than Group GA, both at surgery (point estimate[PE]: 1.8 cm/s; 95% CI: 0.01, 6.3 cm/s; p <0.05([Mann-Whitney])) and recovery (PE: 2.6 cm/s; 95% CI: 0.4, 5.1 cm/s; p = 0.02([Mann-Whitney])). Volume flow[V(Q)] was similar in the 2 groups at baseline and post induction (both, p >0.1([Mann-Whitney])), but was significantly higher in Group EA/GA at surgery (PE: 54 ml/min; 95% CI: 18, 159 ml/min; p = 0.045([Mann-Whitney])) and recovery (PE: 49 ml/min; 95% CI: 16, 129 ml/min; p=0.0037([Mann-Whitney])). Peak velocity, V(mean) and V(Q) increased significantly post epidural in Group EA/GA. Contrary to the venous leg flow attenuation in elective abdominal surgery under GA and upon its recovery, EA administered as part of GA is associated with a significant enhancement of both V(mean) and V(Q). This beneficial hemodynamic effect of EA at the vulnerable stage of recovery may be critically essential in light of enhanced blood viscosity, fibrinolytic shut-down, endothelial/platelet activation and immobility, acting in synergy with putative cardiorespiratory protection. The results of this study lend support to the preferential selection of combined EA/GA in subjects at high risk for venous thromboembolism, particularly when optimal DVT prophylaxis is practically unattainable due to limitations pertaining to the nature of surgery.