A sterile debateBMJ 2007; 335 doi: https://doi.org/10.1136/bmj.39273.655694.BE (Published 19 July 2007) Cite this as: BMJ 2007;335:111
- Stephen Westaby, professor, department of cardiac surgery,
- Keith Channon, professor of cardiovascular medicine, department of cardiology,
- Adrian Banning, consultant, department of cardiology
Did the recent BMJ articles improve the evidence for the superiority of coronary artery bypass grafting (CABG) over percutaneous coronary intervention (PCI) as claimed?1 2 3 In 2006 the featured minimally invasive direct coronary artery bypass graft (MIDCAB) operation for isolated left anterior descending disease accounted for less than 0.5% of 24 000 CABG procedures in the United Kingdom.1 2 Equally the economic arguments apply to practice and hospital costs 10 years ago,3 when PCI strategies were limited, first generation stents were more expensive, and 2-3 days in hospital were considered necessary for safe practice. However, in a current contest between the two procedures 3-5 PCIs more reasonably equate to one CABG.
Currently CABG achieves lower reintervention rates and marginally better survival in multivessel disease with a left main stem lesion. Diabetic patients with diffuse three vessel disease fare better with CABG. Some who are unsuitable for PCI are also poor CABG candidates because of calcified vessels.
Acute coronary syndromes now account for 50% of PCI practice. Many patients have extensive comorbidity and multivessel disease. They are unlikely to be offered urgent CABG as raised troponin concentration is a relative contraindication.
Bridgewater et al suggest counterintuitively that media reporting of CABG mortality statistics (since 2001) has not caused risk averse behaviour in surgeons.4 However, data reporting practices changed at this time. CABG mortality fell, as did the number of cases with left ventricular ejection fraction <30% (only 5.5%). Without a “surgical breakthrough” this implies modification of patient selection.
The relative merits of PCI and CABG in complex multivessel disease have been addressed in a trial which recently completed recruitment of 1800 patients.5 This initiative will provide clear guidance to override the use of selected data in support of one approach over the other. Even so, many less sanguine patients will still choose one or more PCIs first, knowing that CABG is possible if symptoms return.
In summary, PCI and CABG are complementary, not competitive. PCI is preferred for multifocal discrete disease and CABG for diffuse disease with chronic occlusions. Patient choice must now be included in the evidence base.
Competing interests: The authors, a cardiac surgeon and two cardiologists, are SYNTAX investigators and benefit from private practice in myocardial revascularisation.