Background: Directional atherectomy is a frequently used percutaneous revascularization strategy, but its long-term outcomes have not previously been compared with those of balloon angioplasty in a prospective trial.
Methods and results: The 1012 patients enrolled in the Coronary Angioplasty Versus Excisional Atherectomy Trial (CAVEAT I) were followed for at least 1 year after randomization. Analyses of predetermined end points were performed, including a detailed analysis of the 14 patients who died. At 1 year, 11 patients had died in the atherectomy group compared with 3 in the angioplasty group (2.2% versus 0.6%, P = .035), with an excess of out-of-hospital deaths (2.2% versus 0.2%, P = .01) and late cardiac deaths (1.6% versus 0%, P = .01). Univariate predictors of death included age, abrupt closure, periprocedural enzyme elevation, and peripheral vascular complications. There was no evidence that the excess of deaths after atherectomy was linked to perforation, ectasia, or deep resection. Cumulative rates of myocardial infarction were higher in those who had been randomized to atherectomy than in those randomized to angioplasty (8.9% versus 4.4%, P = .005) with a trend toward excess Q-wave and non-Q-wave infarctions. By multivariate analysis, atherectomy was the only variable predictive of the combined end point of death or myocardial infarction. No clinical or angiographic characteristics added to this index. Rates of repeat percutaneous intervention at the target site (24.4% after atherectomy versus 25.9% after angioplasty), coronary artery bypass surgery (9.3% versus 9.1%), hospitalization (50% versus 47.1%), and stroke (1% in both groups) were not significantly different.
Conclusions: Long-term follow-up of the 1012 patients randomized to atherectomy or angioplasty has revealed a statistically significant excess of deaths after directional atherectomy that was not evident at 6 months. This difference could be due to the chance occurrence of a low mortality rate in those randomized to angioplasty. The excess of myocardial infarctions after atherectomy remains statistically significant at 1 year. Further investigation is warranted to improve the safety of atherectomy.