Background: We used intravascular ultrasound (IVUS) to (1) clarify the mechanisms of luminal loss after drug-eluting stent (DES) implantation and (2) classify morphological patterns of in-stent restenosis (ISR).
Methods and results: On the basis of IVUS-identified luminal narrowing (in-stent minimum lumen area <4 mm(2)), IVUS-defined ISR was classified as focal (luminal narrowing ≤10 mm in length), multifocal (≥1 focal lesions), and diffuse (luminal narrowing >10 mm in length) with or without stent edge involvement. Significant intimal hyperplasia (IH) was defined as IH area >50% of stent. Overall, 76 lesions had IVUS-defined ISR; 32 (42%) had stent underexpansion (minimal stent area <5 mm(2)); and 71 (93%) had IH area >50% of stent. Total stent length negatively correlated with minimal stent area (r=-0.613, P<0.001) as well as with stent area at the minimum lumen site (r=-0.436, P<0.001) but not with minimum lumen area (r=-0.084, P=0.472). Underexpansion was present at the minimum lumen site in 15 of 43 (35%) lesions with stent length >28 mm, even though there was significant IH in 34 (79%) lesions; conversely, in 32 of 33 (97%) lesions with stent length ≤28 mm, the minimum lumen site was not associated with stent underexpansion but significant IH. IVUS-defined focal ISR was most common (47%). Compared with focal ISR, normalized vessel, stent, lumen, and plaque volumes were smaller in diffuse and multifocal than focal ISR, with no difference in IH extent.
Conclusions: In most DES restenosis, IH was the dominant mechanism of ISR. Nevertheless, underexpansion associated with longer stent length remained an important preventable mechanism of ISR.