Background: Providing sufficient and convenient analgesia is crucial during the postoperative period after total-knee replacement (TKR) to enhance patient mobility and reduce stress response to surgery. The scope of this study is to compare the effects of levobupivacaine and levobupivacaine plus fentanyl on stress response and analgesic efficiency after TKR. Method: In this study, 40 ASA I - II patients scheduled to undergo TKR were subjected to combined spinal epidural anesthesia (CSEA) injecting of 15 mg levobupivacaine and randomly assigned to receive either levobupivacaine 0.125% (Group L) or levobupivacaine 0.125% plus fentanyl 4 mu g ml(-1) (Group F) during postoperative period via the epidural route. Patient controlled epidural analgesia (PCEA) was offered for 24 hours. Venous blood samples were assayed for ach-enocorticotropic hormone (ACTH), cortisol and prolactin levels before surgery and after analgesia administration. Analgesia was assessed using a visual analogue scale (VAS) at rest (VASR) and during movement (VASM). Results: There was no statistically significant difference between the groups in terms of total doses, bolus requests, bolus delivered and side effects (p > 0.05). The ACTH, cortisol and prolactin levels increased following the surgery and decreased during PCA infusion in both groups where the decline in Group F was significant (p < 0.05) at 24 hours after the analgesic treatment and 48 hours after the surgery. Conclusion: We have demonstrated that infusion of levobupivacaine (0.125%) in combination with fentanyl (4 jig ml-1) using PCEA suppressed stress response to surgery significantly and provided better pain relief than levobupivacaine (0.125%) alone after TKR.