Scanning electron microscopy (SEM) photographs and quantitative analysis of rabbit iliac arteries implanted with overlapping sirolimus-eluting stents (SES), sirolimus-eluting stents plus estradiol (SES+ED), bare-metal stents (BMS), and Cypher drug-eluting stents (CDES). The dashed lines indicate boundaries of stent overlap. (A) En-face examination demonstrates near-complete endothelial coverage of SES, SES+ED, and BMS, with rare areas of exposed struts confined to the overlap region. However, there is marked absence of intact endothelium over stent struts in CDES, especially in the overlapping segment. High magnification of the nonoverlapping and overlapping sites (box) shows that the endoluminal surface of SES, SES+ED, and BMS remains covered by endothelium with very few inflammatory cells. However, CDES show widespread areas of exposed stent struts and incompletely endothelialized lumen with fibrin, adherent heterophils/eosinophils, macrophages, and platelets in the vicinity of stent struts. (B) Quantitative analysis of endothelial coverage: the SES, SES+ED, and BMS demonstrated greater endothelialization than CDES at both the nonoverlapping and overlapping segments of the stent. For CDES, there is also significantly less endothelialization in overlapping as compared with nonoverlapping areas, which does not occur in the ISAR (Individualizable drug-eluting stent System to Abrogate Restenosis) stents. *p = 0.003 versus SES, SES+ED, BMS. p = 0.02 versus nonoverlapping CDES.

(A) High magnification (inset) photomicrographs of overlapping stent struts demonstrate decreased hetero/eosinophil infiltration and fibrin deposition among SES, SES+ED, and BMS versus CDES. The CDES show pronounced inflammatory cell deposition and fibrin accumulation (arrows). Bar graphs demonstrate a significant increase in the percentage of struts surrounded by eosinophils in overlapping segments (B) and fibrin in both nonoverlapping and overlapping segments (C) in CDES versus the other nonpolymeric stents. The degree of giant cell accumulation (D) was different between nonoverlapping and overlapping segments for SES+ED and CDES. *p < 0.0001 versus SES, SES+ED, BMS. p = 0.0001 versus nonoverlapping CDES. **p < 0.0002 versus SES, SES+ED, BMS. p 0.02 for overlapping versus nonoverlapping segments. Abbreviations as in Figure 1.

(A1 to D1) Representative low-power photomicrographs of all 4 stent groups—SES, SES+ED, BMS, and CDES—within the nonoverlapping stent segment. (A2 to D2) The neointima of overlapping SES, SES+ED, and BMS consists of an organized layer of smooth muscle cells in a proteoglycan-rich matrix. In contrast, CDES implanted arteries demonstrate impaired vascular healing with little neointimal growth at the overlap site (note that the struts of the innermost stent are mostly uncovered). (A3 to D3) High-power views from boxed areas in A2 to D2 of overlapping stent strut sections (Movat Pentachrome stain). Occasional giant cells populate the immediate vicinity of overlapping stent struts in all groups. (A4 to D4) Fibrin-rich areas were common in the area neighboring overlapping CDES struts, particularly those uncovered by an appreciable neointima (hematoxylin and eosin stain). Note the remnants of the polymer in the area once occupied by CDES struts (arrows), which are absent from SES, SES+ED, and BMS. Abbreviations as in Figure 1.

Relative protein expression levels of (A) chemokines (macrophage-derived chemokine [MDC], macrophage colony-stimulating factor [MCSF]) and growth factors (vascular endothelial growth factor [VEGF] and insulin-like growth factor [IGF]-1) and (B) various cytokines (interleukin [IL]-2, -4, -8, and -10) from cultured nonpolymeric SES, SES+ED, and BMS implanted arteries at 14 days as illustrated by the box and whisker plots. Values represent ratios calculated as protein expression levels of SES or SES+ED over BMS. Abbreviations as in Figure 1.