We investigated whether blocking afferent nociceptive inputs by continuous intra- and postoperative thoracic epidural analgesia (TEA) would decrease plasma concentrations of brain natriuretic peptide (BNP) in patients who were at risk for, or had, coronary artery disease. Twenty-eight patients undergoing major abdominal surgery received either general anesthesia supplemented with a continuous thoracic epidural infusion of 1.25 mg/mL bupivacaine and 1 microg/mL sufentanil (n = 14; TEA) or general anesthesia followed by IV patient-controlled analgesia (n = 14; IV PCA). Visual analog scale pain scores, hemodynamics, plasma catecholamines, cardiac troponin T, atrial natriuretic peptide (ANP), and BNP were serially measured preoperatively, 90 min after skin incision, at arrival in the intensive care unit, and in the morning of the first, second, and third postoperative day. Dynamic visual analog scale scores were significantly less in the TEA group. TEA reduced the postoperative heart rate without affecting other hemodynamic variables. Plasma epinephrine increased perioperatively in both groups but was significantly lower in the TEA group. Baseline ANP and BNP concentrations were similar between groups (TEA 3.4 +/- 1.8 and 27.0 +/- 12.3 pg/mL; IV PCA 3.1 +/- 2.0 and 25.9 +/- 13.0 pg/mL, respectively). ANP and BNP increased perioperatively in both groups, with significantly lower postoperative BNP levels in TEA patients (TEA 92.1 +/- 31.9 pg/mL; IV PCA 161.2 +/- 44.7 pg/mL). No such difference was observed in plasma ANP concentrations. Plasma cardiac troponin T concentrations were within normal limits in both groups at all times. We conclude that continuous perioperative TEA using local anesthetics and opioids attenuated the release of BNP in patients undergoing major abdominal surgery who were at risk for, or had, coronary artery disease.