Generalized model of restenosis after conventional balloon angioplasty, stenting and directional atherectomy

This study was designed to extend the results of a quantitative model originally developed for restenosis after stenting or atherectomy to include restenosis after conventional balloon angioplasty.

We have previously described a continuous regression model that explains late (6-month) lumen narrowing as the difference between the immediate gain and the subsequent normally distributed late loss in lumen diameter after Palmaz-Schatz stenting or directional atherectomy.

Lumen diameter was measured immediately before and after coronary intervention on 524 consecutive lesions including those treated by Palmaz-Schatz stenting (102), directional atherectomy (134) and conventional balloon angioplasty (288). Of these lesions, 475 (91%) underwent follow-up angiography 3 to 6 months after treatment. The immediate increase in lumen diameter produced by the intervention (immediate gain) and the subsequent reduction in lumen diameter between the time of intervention to follow-up angiography (late loss) were examined. Association between demographic or angiographic variables and continuous measures of restenosis (late lumen diameter or late percent stenosis) was tested with linear regression techniques; a traditional binary measure of restenosis (late diameter stenosis > or = 50%) was evaluated with logistic regression analysis.

Regression models relating late lumen diameter to the immediate lumen result were successfully fitted to all segments studied. According to these models, three indexes of restenosis (late lumen diameter, late percent stenosis and binary restenosis) were found to depend solely on the immediate lumen diameter after the procedure and the immediate residual percent stenosis, but not on the specific intervention used. Moreover, the late loss in lumen diameter was found to vary directly with the immediate gain provided by an intervention, and the "loss index" (a measure that corrects for differences in immediate gain) was uniform among all three interventions.

The quantitative model originally developed for restenosis after stenting or atherectomy may thus be generalized to include conventional balloon angioplasty. It shows that the apparent differences in restenosis among the three interventions studied are due solely to differences in the immediate result provided and not to differences in the behavior of subsequent late loss. Moreover, although the late loss in lumen diameter was found to correlate with differences in the immediate gain provided by an intervention, the "loss index" (a measure that corrects for differences in acute gain) was uniform across all three interventions. It is thus the immediate result (and not the procedure used to obtain that result) that determines late outcome after coronary intervention.