Lack of evidence-based practice leads to major geographical variations in elective coronary revascularization by stents or surgery in England

Since their introduction, coronary artery bypass grafting (CABG) and percutaneous coronary intervention (PCI) have become the standards of care for the large majority of patients with coronary artery disease (CAD) and, by the way, two of the most performed procedures in medicine, worldwide. Both therapies are ubiquitously available in the western world but, interestingly, there are large differences in the provision of PCI and CABG: first between different countries and, as demonstrated by Baig et al., also within the same country with a maximal degree of variation of the ratio PCI/CABG of more than 13-fold. The authors have to be congratulated for bringing these important pieces of informations to the readership of EJCTS and thereby for stimulating thought on how to interpret these numbers? In the last 20 years, numerous randomized trials comparing PCI to CABG have shown comparable outcomes for selected patient populations [1, 2] and provided more evidence for wider use of PCI despite the fact that several developments (off-pump surgery, minimized extracorporeal technology, improved myocardial protection, endoscopic vein harvesting) have made CABG less invasive and highly competitive to PCI. Unfortunately, and because CABG is sometimes described by cardiologists as a procedure where ‘the chest is cracked open’, patients often prefer PCI to CABG because they believe a stitch in the groin is always better than a scar on the chest. In addition, a shorter stay in hospital and the hope for a shorter rehabilitation are attractive prospects. The paper by the group of David Taggart present some expected but also some unexpected items of information concerning the provision of PCI and CABG in the UK [3]. It is very interesting for several aspects: it shows that the provision of PCI and CABG (expressed at least partially by the ratio of PCI to CABG) may be extremely different, even in the same country. In addition, the authors have shown that these differences were not patient related and cannot be explained by the volume of interventions performed. The large differences in the treatment of CAD in the UK call for additional hypotheses: