BACKGROUND The incidence of coronary artery bypass surgery has been increasing annually with increasing pressure on the health care system. Fast track has been proposed as a means to increase efficiency and volume, without an increase in hospital resources. To date this approach has not been critically assessed in Canada. METHODS We examined 617 consecutive patients undergoing isolated CABG surgery. The patients were divided into (1) fast track (FT) recovery (n = 219), without admission to an ICU, and (2) non-fast track (NFT) recovery (n = 398) with direct admission to the ICU. There were no differences in age, gender, timing of surgery, left main stenosis, preoperative myocardial infarction, renal failure, diabetes, peripheral vascular disease, or in the incidence of chronic obstructive pulmonary disease between the two groups. The NFT group had a higher proportion of patients with NYHA Class III/IV symptoms preoperatively (65.7% vs. 57.3%, p = 0.048), in patients with an ejection fraction < 40% (42.5% vs. 30.6%, p = 0.004), or in the number of individuals with an IABP inserted before surgery (13 vs. 1, p < 0.001). RESULTS In the FT group the average period of aortic occlusion (40.7 +/- 15.2 min vs. 71.8 +/- 26.5 min, p < 0.001) and perfusion time (67.8 +/- 24.5 min vs. 117.5 +/- 40.2 min, p < 0.001) were significantly less than in the NFT group. The number of grafts per patient was 3.3 +/- 1.0 vs. 3.2 +/- 1.0, respectively (p = 0.38). Operative mortality was 0.9% in the FT group and 1.3% in the NFT group (p = 1.0). Significant differences were seen in the proportion of patients that suffered from postoperative ventilatory failure (3.2% in FT vs. 12.1% in NFT, p < 0.001), and the proportion of patients that suffered any postoperative complication was significantly higher in the NFT group (21.4%) than in the FT group (9.1%, p < 0.001). The differences in postoperative complications resulted in a shorter length of stay (LOS) in FT patients (5.6 +/- 4.1 days vs. 9.7 +/- 9.4 days NFT, p < 0.001). Only 4.1% of patients that entered the FT group failed and required admission to the ICU. Multivariate stepwise logistic regression analysis identified non-fast track recovery as an independent predictor of morbidity in CABG surgery patients. CONCLUSIONS The data indicate it is possible to perform isolated CABG surgery, in a large proportion of the population, without the need for admission to an ICU for postoperative care.