Aims Recent cost-effectiveness analyses of percutaneous coronary intervention (PCI) versus coronary artery bypass grafting (CABG) have been limited by a short time horizon or were restricted to the US healthcare perspective. We, therefore, used individual patient-level data from the SYNTAX trial to evaluate the cost-effectiveness of PCI versus CABG from a European (Dutch) perspective. Methods and results Between 2005 and 2007, 1800 patients with three-vessel or left main coronary artery disease were randomised to either CABG (n=897) or PCI with drug-eluting stents (DES; n=903). Costs were estimated for all patients based on observed healthcare resource usage over 5 years of follow-up. Health state utilities were evaluated with the EuroQOL questionnaire. A patient-level microsimulation model based on Dutch life-tables was used to extrapolate the 5-year in-trial data to a lifetime horizon. Although initial procedural costs were lower for CABG, total initial hospitalisation costs per patient were higher (.17 506 vs .14 037, p >0.001). PCI was more costly during the next 5 years of follow-up, due to more frequent hospitalisations, repeat revascularisation procedures and higher medication costs. Nevertheless, total 5-year costs remained .2465/patient higher with CABG. When the in-trial results were extrapolated to a lifetime horizon, CABG was projected to be economically attractive relative to DES-PCI, with gains in both life expectancy and quality-adjusted life expectancy. The incremental cost-effectiveness ratio (ICER) (.5390/quality-adjusted life year (QALY) gained) was favourable and remained >.80 000/QALY in <90% of the bootstrap replicates. Outcomes were similar when incorporating the prognostic impact of non-fatal myocardial infarction and stroke, as well as across a broad range of assumptions regarding the effect of CABG on post-trial survival and costs. However, DES-PCI was economically dominant compared with CABG in patients with a SYNTAX Score .22 or in those with left main disease. In patients ...