Contemporary occurrence of stent thrombosis in clinical practice: Better never than late
J. D. Foley,D. Moliterno
DOI: https://doi.org/10.1002/ccd.24343
IF: 2.3
2012-03-01
Catheterization and Cardiovascular Interventions
Abstract:Stent thrombosis (ST) at any time following percutaneous coronary intervention (PCI) is an uncommon yet life-threatening event. The true frequency of ST, particularly beyond the early weeks to months postprocedure, has been challenging to accurately quantify for a variety of reasons [1]. After the proposal and adoption of a series of formal definitions by the Academic Research Consortium (ARC) in 2007 [2], data emerged reporting frequencies of 1% for subacute ST, 0.5% for late ST, and 0.2–0.5% per year, thereafter, for very late ST (VLST) [3,4]. Independent risk factors for ST that have consistently emerged include early discontinuation of dual antiplatelet therapy (DAPT), complexity of the original PCI, and some underlying comorbidities, such as renal failure and diabetes mellitus. The reported consequences of ST are devastating with pooled data from several studies showing a composite mortality and nonfatal myocardial infarction rate of 60–80% at the time of ST or during the first week after the event [5,6]. Information regarding outcomes of ST with modern day percutaneous treatment strategies is limited. In this issue of Catheterization and Cardiovascular Interventions, Yeo et al. [7] add to the literature by reporting outcomes among 153 patients who had 165 episodes of ST within the University of California affiliate hospital network over a 5-year period (2005–2010). Data were collected retrospectively in the first half of this registry interval and prospectively thereafter using the hospital cardiac catheterization laboratory database, medical records, direct review of all angiograms, and the Social Security Death Index. Using the ARC definitions for ST, the distribution of these cases was roughly as follows: acute 4%, subacute 22%, late 18%, and very late 51%—although the exact time or interval at risk beyond 1 year was not available, and 6% of patients are not categorized or reported. As mentioned, noncompliance or premature discontinuation of DAPT is known to be associated with ST, and noteworthy in this report is that 28% of patients with subacute ST and 41% of patients with late ST were not taking DAPT. Surprisingly, nearly onethird of patients with VLST were not on either aspirin or a thienopyridine at the time of stent thrombosis. A majority of these ST cases presented with ST-segment-elevation myocardial infarction (63%), and at least half of all cases had previously received a drug-eluting stent (DES)—again data were incomplete for stent type. Novel information from this report involves procedural treatment regimens: 82% received a glycoprotein IIb/IIIa inhibitor; 52% had aspiration thrombectomy; and 61% received another stent, mainly DES. At 1-year follow-up, 18 deaths (15%) had occurred; five additional deaths occurred beyond 1 year and before study completion. These mortality rates are generally lower than previously reported. Multivariate analysis revealed that only diabetes mellitus and bifurcation disease were independently associated with an increase in in-hospital mortality. Yeo et al. describe a high proportion of VLST in their entire cohort, although this is plausible if the individual follow-up duration was long. The increasingly prevalent finding of VLST in the setting of prior DES implantation is an area whose true etiology has yet to be elucidated, and regulatory agencies are requiring very long-term follow-up information from device manufacturers. Possible mechanisms for VLST include ongoing positive arterial remodeling, late acquired incomplete stent apposition, persistent absence of endothelialization, and chronic inflammation associated with permanent polymers. Both patient-related and procedure-related characteristics have been implicated as reasons for ST, regardless of the timing. Yeo et al. describe bifurcation lesions and diabetes mellitus as the independent predictors of ST in their cohort. Likewise, the high percentage of