Epidural Anesthesia in Elderly Patients Undergoing Coronary Artery Bypass Graft Surgery

Objectives: The purpose of this study was to evaluate the effects of thoracic epidural anesthesia on postoperative N-terminal pro B-natriuretic peptide (NT-proBNP) release in elderly patients undergoing elective coronary artery bypass graft (CABG) surgery. Design: A case-matched, nonrandomized study. Setting: A university hospital, single institution. Participants: 46 consecutive and 46 control patients. Interventions: Ninety-two elderly patients (>65 years old) undergoing elective CABG surgery were recruited. Forty-six patients receiving general and epidural anesthesia were case matched (preoperative medications, ejection fraction, and comorbidities) with 46 control subjects receiving general anesthesia. The primary outcome measure was postoperative NT-proBNP release. The preoperative or intraoperative variables significantly associated with an intensive care unit stay longer than 4 days were determined by logistic regression. Measurements and Main Results: The median (interquartile range) plasma concentrations of NT-proBNP before surgery were 402 (115-887 pg/mL) in the epidural group versus 508 (228-1,285 pg/mL) in the general anesthesia group (p = 0.9), whereas 24 hours after surgery it increased to 1846 (1,135-3,687 pg/mL) versus 5,005 (2,220-11,377 pg/mL) (p = 0.001), respectively. There were more patients (p = 0.043) in the control group (9/46 = 19.5%) than in the thoracic epidural anesthesia group (4/46 = 8.8%) with an intensive care unit stay longer than 4 days. The absence of preoperative β-blocker therapy (odds ratio = 3.94; 95% confidence interval, 1.123-13.833; p =0.03) and of an epidural catheter (odds ratio = 3.91; 95% confidence interval, 1.068-14.619; p = 0.04) were the only preoperative and intraoperative variables independently associated with a prolonged intensive care unit stay. Conclusions: Epidural anesthesia added to general anesthesia for CABG surgery significantly attenuates NT-proBNP release in elderly patients and reduces the incidence of prolonged intensive care unit stay. © 2009 Elsevier Inc. All rights reserved.