Current issues of ACP Journal Club are published in Annals of Internal Medicine


Coronary restenosis was greater in diabetic patients than in nondiabetic patients after atherectomy but not after angioplasty

ACP J Club. 1997 Sep-Oct;127:33. doi:10.7326/ACPJC-1997-127-2-033

Source Citation

Levine GN, Jacobs AK, Keeler GP, et al., for the CAVEAT-I Investigators. Impact of diabetes mellitus on percutaneous revascularization (CAVEAT-I). Am J Cardiol. 1997 Mar 15;79:748-55.



To determine the effects of diabetes on the outcomes of patients having percutaneous revascularization.


6-month randomized controlled trial (Coronary Angioplasty Versus Excisional Atherectomy Trial [CAVEAT-I]).


32 centers in the United States and 3 in Europe.


1012 patients who had symptomatic myocardial ischemia with ≥ 60% stenosis of a native coronary artery and without previous intervention for the target lesion. 90% of patients had angiographic follow-up at 6 months. 191 patients (19%) were defined by their physician as having diabetes before or at the time of revascularization.


Patients were allocated to directional atherectomy (n = 512 [95 patients with diabetes]) or percutaneous transluminal coronary angioplasty (n = 500 [96 patients with diabetes]).

Main outcome measures

Angiographic restenosis (> 50% luminal diameter stenosis at 6 mo after a successful initial procedure). Secondary angiographic end points were initial success rate (reduction in stenosis to ≤ 50%) and the actual percentage of stenosis before and after revascularization and at 6 months. Clinical events (death, myocardial infarction [MI], or repeat revascularization) were analyzed at 1 year.

Main results

Analysis was by intention to treat. At 6 months, the restenosis rate after atherectomy was 59.7% in diabetic patients and 47.4% in nondiabetic patients (P = 0.068); the median actual percentages of stenosis were 60.7% and 50.4% (P < 0.001), respectively. For angioplasty, the restenosis rates in diabetic and nondiabetic patients were 56.1% and 56.8% and the median actual percentages of stenosis were 53.1% and 55%, respectively. Initial success rates did not differ between diabetic and nondiabetic patients for atherectomy (87.8% vs 89.2%) or angioplasty (75% vs 80.7%); acute procedure complication rates also did not differ (5.2% vs 8.9%, P = 0.097).


A higher degree of restenosis occurred in diabetic patients than in nondiabetic patients 6 months after atherectomy. Restenosis rates and degree of restenosis after angioplasty were similar.

Sources of funding: Devices for Vascular Intervention and Eli Lilly.

For article reprint: Dr. A.K. Jacobs, Section of Cardiology, Boston Medical Center, 88 East Newton Street, Boston, MA 02118, USA. FAX 617-638-8712.


The study by Levine and colleagues adds a little further misery to results of percutaneous revascularization procedures in diabetic patients by extending to atherectomy the poor results previously documented with angioplasty. An earlier study of angioplasty in persons with diabetes (n = 1133) and without (n = 9300) showed that within the first year, patients with diabetes had increased rates of restenosis and greater need for another revascularization procedure (1). Further, 5 years after angioplasty, survival rates were slightly lower in patients with diabetes (89% vs 93%), rates of MI were slightly higher (19% vs 11%), and rates of revas-cularization or MI were slightly higher (64% vs 47%) (1).

A study in patients with stent placement showed that diabetic patients had a greater restenosis rate than those without diabetes (55% vs 32%) at 6 months, suggesting intimal hyperplasia rather than acute postprocedural recoil (absent in this study, which used stents) as the mechanism of increased restenosis after angioplasty (2). A recent comparison of angioplasty and coronary artery bypass graft (CABG) showed no difference in survival or MI in patients without diabetes at 5 years. But in the 20% of the patients with treated diabetes (n = 353), 5-year survival rates were substantially lower for those having angioplasty compared with those having CABG (66% vs 81%) and the rates were lower in both groups than the 91% survival with either procedure in the nondiabetic patients (3).

That said, the results of this study differ from the first cited study and find no worse rates of 6-month restenosis or 1-year clinical events with angioplasty in patients with diabetes. Recognizing the inconsistent results in this field, studies would seem to suggest that persons with diabetes do slightly worse in terms of restenosis and clinical follow-up than those without diabetes whether they have had angioplasty, atherectomy, or stent placement. No data exist that show whether tight control of hyperglycemia would offset this effect. A general recommendation that CABG is preferable to PTCA in patients with diabetes must await further desperately needed randomized trials.

Donald A. Smith, MD
Mount Sinai Medical CenterNew York, New York, USA


1. Stein B, Weintraub WE, Gebhart SP, et al. Circulation. 1995;91:979-89.

2. Carrozza JP Jr, Kuntz RE, Fishman RF, Baim DS. Ann Intern Med. 1993;118:344-9.

3. The Bypass Angioplasty Revascularization Investigation (BARI) Investigators. N Engl J Med. 1996;335:217-25.