Although very late stent thrombosis (VLST) after drug-eluting stent (DES) implantation remains a major concern, the precise mechanisms have not been clarified. An association between late acquired incomplete stent apposition (ISA) and VLST of DES has been suggested by several intravascular ultrasound studies demonstrating very high prevalence of ISA in the setting of VLST. To clarify the pathological mechanisms of VLST, we investigated vascular responses of coronary arteries of VLST cases after DES implantation.
Drug-eluting stents (DESs) have dramatically reduced angiographic restenosis and clinical need for repeat revascularization procedures [
The pathological findings from patients who died of late DES thrombosis have demonstrated that delayed arterial healing characterized by incomplete reendothelialization is an important underlying factor [
(a) The left panels show angiograms of a stable angina patient who underwent successful stent implantation in the native left anterior descending coronary artery with SES and the first diagonal branch with BMS. More than 11 months after stenting, severe thrombotic occlusion occurred at the proximal SES site, 1 week after cessation of antiplatelet medication. Despite complete revascularization by thrombus aspiration, the patient died of multiorgan failure 1 week later. (b) Macroscopically, neointimal coverage of the SES struts was scarcely visible (*blood clot formed at agonal stage). In contrast, complete coverage by neointima was observed at the BMS site. (c) Microscopic observation demonstrated no obvious endothelialized struts at the SES site, even at the portion where a mild proliferative response of smooth muscle cells was evident. At the luminal surface of these nonendothelialized struts, the remnants of fibrin-rich thrombus were visible. (d) In the BMS segment, however, relatively large neointimal coverage of stent struts was observed. Furthermore, complete re-endothelialization was clearly visible.
Although a correlation has been observed between uncovered DES struts and LST, in our pathological studies of Japanese patients, considerable number of cases showed neointimal coverage with reendothelialization of a great deal of the DES struts, especially in simple lesions beyond 1 year (Figure
Histological sections of the SES segments harvested beyond 1 year (upper panels) and 2 years (lower panels) after stenting. Note complete neointimal coverage composed of smooth muscle cells with obvious reendothelialization.
Late ISA has been observed on follow-up intravascular ultrasound (IVUS) in patients who received sirolimus-eluting stents (SESs) [
Histological sections of the plaque-free regions with SES implantation after 11.5 months. Note remarkable depletion of medial smooth muscle cells (arrowheads) just underneath the SES struts (*).
Thus, late ISA by focal positive vessel remodeling caused by medial necrosis may be responsible for LST and/or VLST after DES implantation.
Peri-stent contrast staining (PSS) was defined as contrast staining outside the stent contour extending to ≥20% of stent diameter measured by quantitative coronary angiography. PSS found within 12 months after SES implantation appeared to be associated with subsequent VLST [
(a) These angiograms demonstrated that the SES stented site showed a tendency toward irregularly shaped coronary ectasia (peri-stent contrast staining; PSS) and finally saccular aneurysm formation 2 years after SES implantation, followed by VLST 3 months later. (b) Pathological examination demonstrated focal strut malapposition with aneurysmal dilatation and partially occlusive mural thrombus. (c) In addition, extensive inflammation, consisting primarily of lymphocytes and eosinophils, with a focal giant cell reaction, was evident at the stented site. An asteroid body with intense foreign body granulomatous inflammation was also visible (arrow). Such localized hypersensitivity vasculitis existed primarily around the struts and extended to the intima and adjoining media and adventitia.
Recent studies have identified immune cells and mediators at work in atheroma, implicating inflammatory mechanisms in disease development [
Histological sections from a patient with SES implantation 2 years antemortem. Note early necrotic core formation with pronounced foamy macrophages and circumferential cholesterol clefts (arrows) around the struts.
These foamy macrophages (arrowheads) clearly showed immunoreactivity to CD68, MMP-3, and MMP-9. Positive staining for these MMPs was also observed in the extracellular space.
Micrographs of an aspirated specimen from the VLST site in the DES-implanted segment. Note the intimal disruption and adherent thrombus. Numerous macrophages and cholesterol clefts (arrows) were also visible in both the neointima and the thrombus.
Thus, DES can induce atherosclerotic and thrombogenic lesions with a significantly higher incidence and earlier than with BMS.
LST may be more frequently related to incomplete healing and/or inadequate neointimal coverage with poor re-endothelialization. However, in the cases of VLST, several pathological studies have suggested a causal relationship between the inflammatory responses to the durable polymer and VLST, provoking late ISA and accelerated atherosclerosis followed by neointimal disruption. Histopathologic differences (the prevalence of eosinophils, giant cells, and fibrin) among DES platforms have been observed; this may reflect unique responses to the specific polymer/drug.
Thus, until novel DES using superbly biocompatible and/or biodegradable polymers becomes available, we still need to be cautious and carefully keep surveying these devices.
The author has no conflict of interest to declare.