Author + information
- Erland Erdmann, MD, FESC, FACC⁎ (, )
- John A. Dormandy, FRCS, DSc,
- Bernard Charbonnel, MD,
- Massimo Massi-Benedetti, MD,
- Ian K. Moules, BSc (Hons),
- Allan M. Skene, PhD,
- PROactive Investigators
- ↵⁎Medizinische Klinik III, der Universität zu Köln, Kerpener Str. 62, Köln 50937, Germany
We thank Drs. Riche and Dale for their letter commenting on the PROactive 05 subanalysis in patients with myocardial infarction (MI) at baseline. In that prespecified analysis, we reported that pioglitazone reduced the occurrence of adverse cardiac outcomes, including recurrent MI, in the 2,445 patients with previous MI (1).
Drs. Riche and Dale suggest that it would be useful to look specifically at any reductions in cardiac revascularization in the subgroup of patients who had had a previous cardiac intervention (coronary artery bypass graft [CABG] or percutaneous intervention [PCI]). We found that, in the full PROactive analysis set, pioglitazone was equally effective in reducing the occurrence of first revascularization relative to placebo in patients with a prior history of cardiac intervention compared with those without such a history (hazard ratio 0.88 in both patient groups; p for interaction between subgroup and treatment = 0.9960). Thus, in the total PROactive population, there did not appear to be any extra benefit of pioglitazone in the group with a prior history of coronary intervention.
Looking at the same end point in those with both a history of MI and cardiac intervention at baseline, there was again no evidence of a treatment subgroup interaction (hazard ratios 0.86 in those with and 0.75 in those without prior PCI/CABG; p for interaction between subgroup and treatment = 0.6013). Only 140 patients had a prior CABG at baseline, but an analysis of those with PCI (n = 214) yielded a similar outcome.
One major limitation to these analyses is the high correlation between the end points of MI/acute coronary syndrome (ACS) and PCI/CABG. However, any direct effect on the risk of PCI/CABG should be reflected in elective PCI/CABG rates. Using the definition of elective PCI/CABG as occurring without MI or ACS or occurring >30 days after MI/ACS, there were 123 events in the pioglitazone group and 132 in the placebo group (hazard ratio 0.94; p = 0.6135).
Taking all of the above information together, we conclude that, in the PROactive study, pioglitazone reduced MI relative to placebo and this was reflected in PCI/CABG rates, but there was no conclusive evidence that pioglitazone influenced revascularization rates differently.
- American College of Cardiology Foundation
- Erdmann E.,
- Dormandy J.A.,
- Charbonnel B.,
- Massi-Benedetti M.,
- Moules I.K.,
- Skene A.M.