Background and objective: High-dose opioid anaesthesia contributes to decreasing metabolic and hormonal stress responses in patients undergoing cardiac surgery. However, the increase in context-sensitive half-life of opioids given as a high-dose regimen can affect postoperative respiratory recovery. In contrast, remifentanil can be given in high doses without prolonging context-sensitive half-life due to its rapid metabolism. Therefore, we performed a prospective, randomized trial to compare anaesthesia consisting of propofol/remifentanil or propofol/sufentanil with regard to postoperative respiratory function and outcome. Methods: Patients undergoing coronary artery bypass grafting were randomized to a propofol/remifentanil (0.5–1.0 μg kg−1 min−1) or propofol/sufentanil (30–40 ng kg−1 min−1) based anaesthetic. Carbon dioxide response, forced expiratory volume in one second, vital capacity, and functional residual capacity were measured 1 day prior to the operation, 1 h before extubation, 1, 24 and 72 h after extubation. In addition, the incidence of atelectasis, pulmonary infiltrates, intensive care unit and postoperative length of stay were compared. Patients and physicians were blinded to the treatment group. Results: Twenty-five patients in each treatment group completed the study. There was no difference between patients of the treatment groups regarding demographics, risk- or pain scores. In all patients, carbon dioxide response, forced expiratory volume in one second, vital capacity and functional residual capacity were decreased postoperatively compared to baseline. Patients randomized to remifentanil had less depression of carbon dioxide response, less atelectasis and shorter postoperative length of stay (12 d vs. 10 d) than after sufentanil (P < 0.05). Conclusions: Intraoperative use of high-dose remifentanil for coronary artery bypass grafting may be associated with improved recovery of pulmonary function and shorter postoperative hospital length of stay than sufentanil.