Stent apposition in late and very late stent thrombosis

Thrombosis of an intracoronary stent is a sudden, devastating, rare event. Timing of stent thrombosis (ST) may be acute (0–24 h after implantation), subacute (>24 h to 30 days), late (>30 days to 1 year), or very late (>1 year). When ST occurs, it typically manifests as acute myocardial infarction (MI) or sudden death 1. Although both bare metal stents (BMS) and drug-eluting stents (DES) are at risk for ST, late and very late events are of particular concern for DES. 2006 reporting of this extended risk in Basel Stent Kosten-Effekivitäts Trial-Late Thrombotic Events 1, coupled with pathologic observations of delayed healing in DES with late ST 2, fueled panic in the cardiology community, fall in DES use and a science advisory underscoring importance of uninterrupted dual-antiplatelet therapy (DAPT) after DES implantation 3. Subsequent introduction of newer DES with lower ST risk has rekindled discussion as to the appropriate duration of DAPT after DES implantation, with recent observational data showing safety of discontinuation or physician-guided interruption of DAPT earlier than 12 months after stenting 4. As this conversation proceeds, we need to better understand the mechanisms of late and very late ST. In this issue of Catheterization and Cardiovascular Interventions, Pesarini and colleagues make an important contribution through their report of rigorous intravascular ultrasonography (IVUS) in 34 consecutive patients with late or very late ST. Patients with DES ST (n = 25, 17 with first-generation DES) tended to present with ST-segment elevation MI (STEMI, 60%) and thrombotic total occlusion (64%), with frequent incomplete stent apposition (ISA, 56%), particularly among those patients still on DAPT. In contrast, patients with BMS ST (n = 9) were less likely to exhibit STEMI, total occlusion or ISA, and more likely to show rupture of neoatherosclerotic plaque. To permit unadulterated assessment of stent morphology at the time of ST, study protocol required IVUS performance immediately after angiography and before balloon dilatation. Although the authors are to be congratulated in the name of science, for a clinician who has cared for patients in the throes of ST, this interruption in reperfusion therapy may give some pause. In cases where intraluminal thrombus caused total occlusion, operators were permitted to perform aspiration thrombectomy prior to IVUS. One patient died during catheterization; remarkably, all other included subjects survived hospitalization, with one death and two recurrent ST at 12-month follow-up. These favorable outcomes of the index hospitalization for ST speak to selection bias in this analysis; inclusion required survival through angiography and ability to cross the culprit lesion with an IVUS catheter prior to balloon dilatation. Indeed, six cases were excluded due to inability to cross the lesion or other technical difficulties. This may have prohibited IVUS in patients with different lesion morphologies. A key contribution of this study is observation of ISA in a majority of patients with DES ST while on DAPT. This hypothesis-generating finding adds to controversy over the relationship between ISA and ST 5. It is of interest, as authors acknowledge, that ISA observed in conjunction with DES ST is often absent at DES implantation and rather develops over time, in part through positive vascular remodeling. In this respect, is ISA a maker of ST or rather a marker of atherosclerotic progression and associated vulnerability to thrombosis? Most cases of late and very late DES ST in this study occurred in patients with first-generation DES. It is not certain that findings apply to the modern catheterization laboratory, where newer DES predominate, and now bioabsorbable and self-apposing stents (www.clinicaltrials.gov, NCT01732341) are under active investigation. Attention to the relationship between ISA and ST, using not only IVUS but also the higher-resolution optical coherence tomography, is needed with these newer stent platforms. Patients with late or very late DES ST after discontinuation of DAPT were distinguished in this study by complex lesions and stent undersizing or underexpansion. Although authors infer that DAPT was effective in preventing ST in these cases, it is remarkable that, in 9 of 11 cases, ST occurred more than 12 months after thienopyridine cessation. Although there may be a role in practice for prolonged or even indefinite DAPT in select cases based on lesion morphology or stenting quality, the criteria for identifying these cases are ill-defined, and the number needed to treat to prevent ST is likely high. As efforts proceed to define optimal DAPT duration after DES implantation, can we better stratify risk for late and very late ST? The results of this analysis provide important clues to coronary structural correlates of ST. But whether these findings are makers of ST or rather markers for a separate thrombotic vulnerability requires further study, in particular with newer stent platforms.