Differences in restenosis propensity of devices for transluminal coronary intervention. A quantitative angiographic comparison of balloon angioplasty, directional atherectomy, stent implantation and excimer laser angioplasty. CARPORT, MERCATOR, MARCATOR, PARK, and BENESTENT Trial Groups.

With the increasing clinical application of new devices for percutaneous coronary revascularization, maximization of the acute angiographic result has become widely recognized as a key factor in maintained clinical and angiographic success. What is unclear, however, is whether the specific mode of action of different devices might exert an additional independent effect on late luminal renarrowing. The purpose of this study was to investigate such a difference in the degree of provocation of luminal renarrowing (or 'restenosis propensity') by different devices, among 3660 patients, who had 4342 lesions successfully treated by balloon angioplasty (n = 3797), directional coronary atherectomy (n = 200), Palmaz-Schatz stent implantation (n = 229) or excimer laser coronary angioplasty (n = 116) and who also underwent quantitative angiographic analysis pre- and post-intervention and at 6-month follow-up. To allow valid comparisons between the groups, because of significant differences in coronary vessel size (balloon angioplasty = 2.62 +/- 0.55 mm, directional coronary atherectomy = 3.28 +/- 0.62 mm, excimer laser coronary angioplasty = 2.51 +/- 0.47 mm, Palmaz-Schatz = 3.01 +/- 0.44 mm; P < 0.0001), the comparative measurements of interest selected were the 'relative loss' in luminal diameter (RLoss = loss/vessel size) to denote the restenosis process, and the 'relative lumen at follow-up' (RLfup = minimal luminal diameter at follow up/vessel size) to represent the angiographic outcome. For consistency, lesion severity pre-intervention was represented by the 'relative lumen pre' (RLpre = minimal luminal diameter pre/vessel size) and the luminal increase at intervention was measured as 'relative gain' (relative gain = gain/ vessel size). Differences in restenosis propensity between devices was evaluated by univariate and multivariate analysis. Multivariate models were constructed to determine relative loss and relative lumen at follow-up, taking account of relative lumen pre-intervention, lesion location, relative gain, vessel size and the device used. In addition, model-estimated relative loss and relative lumen at follow-up at given relative lumen pre-intervention relative gain and vessel size, were compared among the four groups. Significant differences were detected among the groups both with respect to these estimates, as well as in the degree of influence of progressively increasing relative gain, on the extent of renarrowing (relative loss) and angiographic outcome (relative lumen at follow-up), particularly at higher levels of luminal increase (relative gain). Specifically, lesions treated by balloon angioplasty or Palmaz-Schatz stent implantation (the predominantly 'dilating' interventions) were associated with more favourable angiographic profiles than directional atherectomy or excimer laser (the mainly 'debulking' interventions). Significant effects of lesion severity and location, as well as the well known influence of luminal increase on both luminal renarrowing and late angiographic outcome were also noted. These findings indicate that propensity to restenosis after apparently successful intervention is influenced not only by the degree of luminal enlargement achieved at intervention, but by the device used to achieve it. In view of the clinical implications of such findings, further evaluation in larger randomized patient populations is warranted.