Author + information
- Maciej Pruski Jr.1,
- Aleksandra Blachut2,
- Mateusz Kachel1,
- Magdalena Michalak3,
- Adam Janas3,
- Carlos Fernandez4,
- Paweł Buszman1,
- Krzysztof Milewski1 and
- Piotr Buszman1
- 1Center for Cardiovascular Research and Development, American Heart of Poland, Katowice, Poland
- 2Center for Cardiovascular Research and Development American Heart of Poland, Katowice, Poland
- 3Center of Cardiovascular Research and Development, American Heart of Poland, Katowice, Poland
- 4Center for Cardiovascular Research and Development, American Heart of Poland, Katowice, Poland
Although optical coherence tomography (OCT) provides excellent resolution, the indentification of endothelialisation and complete stent healing still cannot be confirmed, thus limiting the clinical value of this technology. We aimed to establish the determinants of complete healing in OCT by validation with histopathology in the experimental model.
In total 106 stents (82 DES, 24 BMS) were implanted in porcine coronary arteries either for 28 or 90 day follow-up. Following terminal imaging with OCT animals were sacrificed and stented segments harvested for pathological analysis. Overall, 273 OCT frames and stent cross sections from pathology were matched. In pathology, a complete stent healing was defined as endothelialization >90% (allowed due to possible preterminal injury from OCT procedure), no fibrin, highest smooth muscle score, inflammation score <1 and no restenosis (>50% Diameter Stenosis) in pathology. Strut coverage in OCT was classified as embedded, malapossed, protruding covered and uncovered.
A cutoff neointimal thickness (NT) value of >0.22 mm in OCT was found on ROC curve analysis to indicate complete healing (AUC: 0.7 sensitivity 52.8%, specificity 76.4%; p<0.01). Sections with fully embedded struts were more often considered healed (47.5% vs. 16,9% ; p<0.01), had higher endothelization score (2.37 vs. 2.08, p=0.004), lower fibrin score (0.53 vs. 0.85, p=0.0003), higher smooth muscle score (2.24 vs. 2.0, p=0.01), and higher injury score (0.77 vs 0.33, p<0.0001) when compared to sections with protruding or malapossed struts. No significant difference was found for inflammation score (0.44 vs. 0.39, p=0.5). Sections containing only embedded struts compared to sections with protruding covered struts had significantly lower fibrin score (0.53 vs 0.9, p=0.005) and smooth muscle score (2.23 vs 2.03, p=0.03).
Our results suggest that only stent sections with all struts embedded and NT>0.22 mm have high probability of complete healing in pathology, when compared to sections with thinner neointima and protruding or malapposed struts.
IMAGING: Imaging: Intravascular