One hundred twenty-five lesions successfully treated with excimer laser coronary angioplasty at 3 centers were analyzed in a central core laboratory using detailed quantitative angiographic analysis. Sixty-seven narrowings had restenosis (greater than or equal to 50% diameter stenosis at restudy). Correlates of restenosis were as follows: base-line diameter stenosis was 79% in the restenosis group versus 71% in the group without restenosis (p = 0.0002), baseline minimal diameter stenosis was 0.55 mm in the restenosis group versus 0.72 mm in the group without restenosis (p = 0.006), final diameter stenosis was 40% in the restenosis group versus 32% in the group without restenosis (p = 0.002), lesion length greater than or equal to 7 mm was present in 43% of the restenosis group versus 21% in the group without restenosis (p = 0.009), and Thrombolysis in Myocardial Infarction trial flow 0 to 2 was 33% in the restenosis group versus 15% in the group without restenosis (p = 0.025). The strongest multivariate correlate of restenosis was the baseline diameter stenosis (p = 0.003). Whereas most predictors were not controllable, achieving a low residual diameter stenosis that is operator-dependent can favorable influence the restenosis rate.