Four of the studies comparing CABG with plain balloon angioplasty for multivessel disease reported subgroup analyses for diabetics. The largest of these subgroup analyses was derived from BARI, which included 353 patients with diabetes. During the 5-year follow-up, the mortality rate of diabetics randomized to plain balloon angioplasty was 34.5%, and after seven years it was 44.3%, whereas after CABG the respective mortality rates were 19.4% (P = .03) and 23.6% (P = .01).57 The difference in mortality in BARI could be attributed to a difference in cardiac mortality (20.6% vs. 5.8% during 5-year follow-up). MIs showed a similar incidence in both treatment groups but were less often lethal in surgically treated patients. This may be attributed, at least in part, to a less complete revascularization in the PCI arm. The findings in BARI led to a clinical alert of the National Heart, Lung and Blood Institute, abandoning plain balloon angioplasty as a treatment option for multi-vessel coronary artery disease.
There has been considerable debate about the results of BARI, because the meta-analysis of the 5-year follow-up of the three smaller studies reporting subgroup analyses for diabetics, RITA, CABRI, and EAST, did not reveal a significant disadvantage of multivessel PCI versus CABG in diabetics.58 Nevertheless, the 8-year follow-up results of EAST demonstrated a significant survival benefit of CABG over PCI.59 In addition, seven registries were reported comparing plain balloon angioplasty with bypass surgery. In each of these registries, adjusted mortality after catheter intervention was higher than after bypass surgery. However, statistical significance was reached only in the largest registry, the Northern New England Cardiovascular Disease Study Group, and in the subgroup of the Emory registry with insulin treatment.60,61 Except for the Emory registry, adjustment did not include completeness of revascu-larization, probably because incomplete revascular-ization was considered an inherent problem of the interventional approach. In the Emory registry, the adjustment for completeness of revascularization abrogated the differences between catheter treatment and bypass surgery. These findings again point to the importance of achieving complete revascularization with PCI.
The studies during the era of plain balloon angio-plasty are not transferable to current practice. As first demonstrated by the studies on abciximab, the increased risk of thrombotic complications during early and longer term follow-up can be abrogated by intense antiplatelet therapy.62,63 More recently, the ISAR-SWEET study suggested that a similar effect can be achieved by effective pretreatment with clopido-grel.64 In addition, it was shown by various studies that stents compared with plain balloon angioplasty reduce the subsequent incidence of restenosis, although this incidence continues to be higher than in nondiabetics.54,65,66 Given the major impact of restenosis on survival, it is plausible that stents, compared with plain balloon angioplasty, may improve the long-term outcome of PCI substantially. Finally, the recently improved means of achieving tight metabolic control can further improve outcome after catheter intervention. Independent studies demonstrated that outcome after PCI in diabetic patients with tight metabolic control is similar to that in nondiabetic patients.67,68
Was this article helpful?