To investigate the impact of 2 postoperative analgesic regimens on heart rate variability in patients who underwent thoracotomy.A prospective, randomized trial.A single-institutional study in a university hospital.Fifty patients who underwent thoracotomy under combined general anesthesia and thoracic epidural analgesia divided by a number generator into 2 equal groups (A and B).In group A, postoperative analgesia consisted of thoracic epidural analgesia with levobupivacaine for 6 postoperative days. In group B, on the 3rd postoperative day this regimen was changed to patient-controlled intravenous morphine. Heart rate variability recordings were performed on the day before surgery, after the epidural, after operation, and on every postoperative day. Statistical analysis used chi-square and Student t tests (Bonferroni correction).In both groups, the low-frequency component of the analyzed recordings declined after epidural and after surgery. In group A, the low-frequency component was significantly lower compared with baseline from the 2nd postoperative day onward, whereas in group B it was significantly higher compared with A on the 4th and 6th postoperative days. In both groups, the changes in high frequency were statistically insignificant. Intergroup comparisons of the low-/high-frequency ratio showed statistical difference on the last day of observation. There was no difference between the groups in hemodynamic variables and visual analog scale/10 scores.Postoperatively decreased cardiac sympathetic outflow continues with epidural analgesia, whereas it is abolished by the change to intravenous patient-controlled morphine.