Peristrut microhemorrhages: a possible cause of in-stent neoatherosclerosis?

BACKGROUND: In-stent neoatherosclerosis is characterized by the delayed appearance of markers of atheroma in the subintima, but the pathophysiology underlying this new disease entity remains unclear. METHODS AND RESULTS: We collected 20 human coronary artery stents by removal from explanted hearts. The mean duration of stent implantation was 34 months. In all samples, neoatherosclerosis was detected, particularly in peristrut areas. It consisted of foam cells and cholesterol clefts, with or without calcification, associated with neovascularization. Iron and glycophorin-A were present in peristrut areas, as well as autofluorescent ceroids. Moreover, in response to neoatherosclerosis, tertiary lymphoid organs (tissue lymphoid clusters) often developed in the adventitia. Some of these features could be reproduced in an experimental carotid stenting model in rabbits fed a high-cholesterol diet. Foam cells were present in all samples, and peristrut red blood cells (RBCs) were also detected, as shown by iron deposits and Bandeiraea simplicifiola isolectin-B4 staining of RBC membranes. Finally, in silico models were used to evaluate the compliance mismatch between the rigid struts and the distensible arterial wall using finite element analysis. They show that stenting approximately doubles the local von Mises stress in the intimal layer. CONCLUSIONS: We show here that stent implantation both in human and in rabbit arteries is characterized by local peristrut microhemorrhages and finally by both cholesterol accumulation and oxidation, triggering together in-stent neoatherosclerosis. Our data indicate that these processes are likely initiated by an increased mechanical stress due to the compliance mismatch between the rigid stent and the soft wall.

Monitoring of arterial wall remodelling in atherosclerotic rabbits with a magnetic resonance imaging contrast agent binding to matrix metalloproteinases.

AIMS: P947 is a gadolinium-based magnetic resonance imaging (MRI) contrast agent with high affinity for several matrix metalloproteinases (MMPs) involved in arterial wall remodelling. We tested whether the intensity of enhancement detected in vivo in the arterial wall with P947 and MRI correlates with actual tissue MMP-related enzymatic activity measured in a rabbit atherosclerotic model subjected to dietary manipulations. METHODS AND RESULTS: Aortas of 15 rabbits in which atherosclerotic lesions were induced by balloon angioplasty and 4 months of hypercholesterolaemic diet were imaged at 'baseline' with P947-enhanced MRI. Atherosclerotic rabbits were divided into three groups: five rabbits were sacrificed ('baseline' group); five rabbits continued to be fed a lipid-supplemented diet ('high-fat' group); and five rabbits were switched from atherogenic to a purified chow diet ('low-fat' group). Four months later, a second P947-enhanced MRI was acquired in the 10 remaining rabbits. A significantly lower signal was detected in the aortic wall of rabbits from the 'low-fat' group as compared with rabbits from the 'high-fat' group (21 ± 6 vs. 46 ± 3%, respectively; P = 0.04). Such differences were not detected with the contrast agent P1135, which lacks the MMP-specific peptide sequence. In addition, the intensity of aortic wall enhancement detected with MRI after injection of P947 strongly correlated with actual MMP-2 gelatinolytic activity measured in corresponding aortic segments using zymography (r = 0.87). CONCLUSION: P947-enhanced MRI can distinguish dietary-induced variations in MMP-related enzymatic activity within plaques in an experimental atherosclerotic model, supporting its utility as a clinical imaging tool for in vivo detection of arterial wall remodelling.