Correlation of angiographic late loss with neointimal coverage of drug-eluting stent struts on follow-up optical coherence tomography.

Minimal data have been published on the correlation between angiographic late loss (LL) and incomplete neointimal coverage of struts after drug-eluting stent (DES) implantation. Therefore, we evaluated the relationship between angiographic LL and the percentage of uncovered struts on follow-up optical coherence tomography (OCT) images, in all cross-sections of the lesions. From the OCT registry database, 219 lesions without restenosis after DES implantation were divided into tertiles based on angiographic LL: tertile I (LL ≤ 0.26 mm), tertile II (0.26 < LL < 0.59 mm), and tertile III (≥0.59 mm). Lesions with the percentage of uncovered struts in the highest quartile (≥75th percentile; >6.0%) were defined as highly uncovered; in an independent analysis, lesions without any uncovered strut(s) were defined as completely covered. Higher percentages of uncovered struts were observed in tertile I than in both tertile II and III (10.3 ± 12.8% vs. 4.2 ± 7.4% vs. 2.4 ± 5.1%, respectively; P < 0.001 for I vs. II and I vs. III). Angiographic LL correlated significantly with the percentage of uncovered struts on OCT (r = -0.340, P < 0.001). The best cut-off values of angiographic LL to predict highly uncovered and completely covered lesions were 0.29 mm (area under curves [AUC] = 0.723, P < 0.001) and 0.61 mm (AUC = 0.692, P < 0.001), respectively. Angiographic LL inversely and significantly correlated with the percentage of uncovered struts on OCT after DES implantation.