The most cost-effective initial treatment for diabetic patients with stable coronary heart disease (CHD) depends on the seriousness of their CHD: intensive medical management appears more cost-effective than percutaneous coronary intervention (PCI) in less seriously affected patients, however those with more severe disease should probably proceed straight to CABG (coronary artery bypass grafting).

These conclusions come from pre-specified secondary outcome and economic analyses of the BARI-2D trial (Bypass Angioplasty Revascularization Investigation 2 Diabetes), published early online in Circulation. The primary results from BARI-2D were published earlier this year (see link to NEJM study) and showed no significant difference in 5-year mortality between initial medical management and prompt revascularisation strategies, although a pre-specified secondary analysis suggested prompt revascularisation was advantageous in those whose CHD was considered sufficiently severe as to warrant CABG rather than PCI. The latest papers report other major pre-specified clinical outcomes (cardiac death and myocardial infarction, MI), and an economic analysis of the results.

Participants were patients with confirmed type 2 diabetes and stable CAD who were candidates for elective revascularisation, recruited between January 2001 and March 2005 from sites in North America, South America, and Europe. Clinical indication for immediate revascularisation was an exclusion criterion. Eligible patients were randomised in a 2x2 factorial design to either prompt revascularisation or medical therapy, and to either insulin sensitisation or insulin provision therapy. Revascularisation was by PCI or CABG as considered clinically appropriate, and patients were stratified according to which was used; medical therapy was according to current guidelines and patients received revascularisation only if specifically indicated. Insulin sensitising therapy used metformin or a glitazone, and insulin provision therapy a sulphonylurea or insulin, both to achieve a target glycated haemoglobin (HbA1c) level below 7%. There were 2,368 patients randomised to the study, and most (2,194, 92.7%) completed the study as designed; average follow-up duration was 5.3 years, and by the end of this period, 42.1% of the initial medical therapy group had undergone revascularisation.

The pre-specified secondary analysis of cardiac death and MI, analysed by intention to treat, found no significant differences for most of the groups. Over the follow-up period, there were 316 deaths of which 43% were attributed to cardiac causes, and 279 first MI events. Five-year cardiac mortality did not differ between revascularisation plus intensive medical therapy (5.9%) and intensive medical therapy alone groups (5.7%; p = 0.38) or between insulin sensitisation (5.7%) and insulin provision therapy (6%; p = 0.76).

In those assigned to CABG (n=763), MI events were significantly less frequent in revascularisation plus intensive medical therapy versus intensive medical therapy alone groups (10.0% versus 17.6%; P=0.003), and the composite end point of cardiac death or MI (P=0.03) was also less frequent. Reduction in MI (P=0.001) and cardiac death/MI (P=0.002) was significant only in the insulin sensitisation group.

The authors conclude that in many patients with type 2 diabetes and stable CHD, intensive medical therapy should be the first line of treatment. In those with more severe CHD, early CABG may be preferred provided there are no contra-indications.

In the economic analysis, costs were unsurprisingly higher for revascularisation: in those assigned to CABG, the four-year costs were $80,900 for revascularisation vs. $60,600 for medical therapy (P<0.0001). In those allocated to PCI, costs were $73,400 vs. $67,800 respectively (p <0.02). According to lifetime projections of cost-effectiveness, medical therapy was cost-effective compared with revascularisation in the PCI group ($600 per life-year added) with high confidence; however in the more severely affected (CABG) group, revascularisation may be cost-effective ($47 000 per life-year added) but with lower confidence.

The authors conclude that in patients with type 2 diabetes and stable CHD, prompt revascularisation increases costs. In patients considered suitable for PCI, initial intensive medical therapy with subsequent revascularisation as needed is more cost effective than prompt intervention. For those with more severe disease who would be considered suitable for CABG, prompt intervention is probably more cost-effective.

[Editors note: while the study results can probably be generalised more widely, the health economic evaluations, as always, will only be directly applicable in the healthcare system from which the original costs and patients were derived. Due to differences in practice and costs, they can only be indicative in other healthcare systems].