To determine the incidence and predictors of abrupt closure (TIMI flow = 0 or 1) in patients undergoing new device angioplasty (NDA), we studied the hospital records and qualitative morphologic and quantitative angiographic methods of 1,983 patients with unfavorable lesion morphologies treated electively in native coronary arteries with intracoronary stents (N = 96), extraction atherectomy (TEC; N = 27). directional atherectomy (N = 563), rotational atherectomy (N = 775), or laser (N = 522). The overall incidence of abrupt closure was 3.9% (78/1983), with 2.1% (42/78) occurring during the procedure and 1.8% (36/78) occurring out-of-Iab (1 patient with both inand out-of-Iab abrupt closure). Among NDA, TEC (14.8%) and stents (8.3%) had relatively higher frequencies of abrupt closure compared to other NDAs (mean of 3.5%). Multivariate analysis of demographic and lesion characteristics revealed the following predictors associated with either in-lab or out-oflab abrupt closure:Independent PredictorsOdds Ratio95% C.I.p valueUnstable angina2.011.13-3.57<0.05Lesion length ≥ 10 mm2.001.26-3.19<0.005Angulation ≥ 45 °1.821.09-3.04<0.05C.I. = confidence interval C.I. = confidence interval The sole independent predictor for out-of-lab only abrupt closure was dissection (Odds Ratio = 3.11, p < 0.005). We conclude that:(1) The incidence of abrupt closure (both in- and out-of-lab) was 3.9% in patients undergoing elective new device angioplasty, with nearly equal distribution of in- and outof- lab occurrences: (2) Unstable angina. lesion length ≥ 10 mm, and angulation ≥ 45° predicted either in- or out-of-lab abrupt closure, whereas dissection was the sole significant predictor of out-of-lab only abrupt closure: and (3) both lesion-related (thrombus-containing lesions for extraction atherectomy) and device-related factors (stents with subacute thrombosis) added to the acceptable rates of abrupt closure with these new devices for lesions unfavorable for conventional angioplasty.