Objectives: In this study, we aimed to compare the efficacy of bupivacaine and levobupivacaine in the patient-controlled epidural analgesia (PCEA) in patients who underwent total knee arthroplasty (TKA), particularly with cardiac diseases.Patients and methods: Forty-four patients (16 males. 28 females; mean age 70.2±6.3 years; range 18 to 80 years) who were scheduled for TKA surgery under combined spinal and epidural anesthesia were included. Patients were randomly divided into two groups. Bupivacaine 15 mg 0.5% for group I (n=23) and levobupivacaine 15 mg 0.5% for group (n=23) were administered via spinal route. The amount of local anesthetics, PCEA starting times and bolus/demand ratio of local anesthetics were recorded. The level of pain, need for additional analgesic, total amount of analgesic, degree of motor block, complications were asked for patient satisfaction at postoperative 6, 12. 24 and 48 hours.Results: No statistical differences were determined between group 1 and group 2 in terms of the operation times, PCEA starting times, bolus number/demand ratios, visual analog scale (VAS) scores and numerical rating scores (NRS) at the postoperative 6, 12, 24 and 48 hours, need for additional analgesic, total amount of analgesic, degree of motor block and patient satisfaction. Although the incidence of side effects was lower in the group I compared to group 2, it was not statistically significant.Conclusion: We concluded that it would be proper using levobupivacaine rather than bupivakaine for patients with cardiac diseases, as the cardiotoxic and neurotoxic effects of levobupivacaine is lower then bupivacaine with a similar analgesic efficacy.