Late Coronary Stent Thrombosis

Abstract

HomeCirculationVol. 116, No. 17Late Coronary Stent Thrombosis Free AccessReview ArticlePDF/EPUBAboutView PDFView EPUBSections ToolsAdd to favoritesDownload citationsTrack citationsPermissions ShareShare onFacebookTwitterLinked InMendeleyReddit Jump toFree AccessReview ArticlePDF/EPUBLate Coronary Stent Thrombosis Stephan Windecker, MD and Bernhard Meier, MD Stephan WindeckerStephan Windecker From the Department of Cardiology, University Hospital Bern, Bern, Switzerland. Search for more papers by this author and Bernhard MeierBernhard Meier From the Department of Cardiology, University Hospital Bern, Bern, Switzerland. Search for more papers by this author Originally published23 Oct 2007https://doi.org/10.1161/CIRCULATIONAHA.106.683995Circulation. 2007;116:1952–1965Introduced >20 years ago,1 coronary artery stents have improved the safety and particularly the efficacy of percutaneous coronary interventions (PCIs).2 Abrupt vessel closure, complicating 6% to 8% of balloon angioplasty procedures, was associated with a 5% mortality, 40% rate of myocardial infarction (MI), and 40% rate of emergency coronary artery bypass grafting.3 Stents significantly reduced these adverse events (Figure 1).4,5 The reduction of restenosis afforded by bare metal stents (BMS) was modest (30% to 40%). Repeat revascularization still occurred in 15% to 20% of cases.6 Drug-eluting stents (DES) with antiproliferative drugs attached via polymers on the stent surface to minimize smooth muscle proliferation have reduced restenosis and rates of target lesion revascularization by 50% to 70% compared with BMS across nearly all lesion and patient subsets.7 Initially8 and again more recently,9–16 safety concerns were raised about DES, particularly about late stent thrombosis (ST). Download figureDownload PowerPointFigure 1. Outcome of 112 patients with abrupt vessel closure in the prestent balloon angioplasty experience of a registry (1985 to 1986; total, 1801 patients)3 and 339 patients with abrupt vessel closure during balloon angioplasty treated with BMS (Palmaz-Schatz) (total, 4596 patients).5 CABG indicates coronary artery bypass grafting.Historical PerspectiveST was a bane of stent implantation from the beginning. The initial experience with Wallstents in the late 1980s was overshadowed by ST rates approaching 24%.17 Subsequent series with Palmaz-Schatz and Gianturco-Roubin stents (still predominantly bailout stenting) observed ST in 6% to 12% of cases.4,18 The postprocedural antithrombotic regimen at the time consisted of aspirin, often in conjunction with oral anticoagulation. Dual antiplatelet therapy of aspirin and the thienopyridine ticlopidine in conjunction with a shift from bailout to elective stenting resulted in a significant reduction of ST to <2%.5,19 Earlier oral antiplatelet drug loading and glycoprotein IIb/IIIa antagonists further diminished ST. Table 1 provides an overview of ST reported in contemporary trials of BMS and DES, ranging from 0.1% to 3.1%.11,20–38 The wide range in the incidence of ST across trials with both BMS and DES is explained by differences in definitions, length of follow-up, antithrombotic drug regimens, and complexity of patients and lesions. Table 1. Coronary STStudyYearStent TypeTotal Population, nST, n (%)Definition of STThienopyridine (%)TimeBMS Karrillon et al201996BMS290051 (1.8)Definite and probableTiclopidineEarly Moussa et al211997BMS100119 (1.9)Definite and probableNone (25)EarlyTiclopidine (75) Schühlen et al221998BMS283365 (2.3)DefiniteTiclopidine (80)EarlyTiclopidine (75)None (2) De Servi et al231999BMS93914 (1.5)DefiniteTiclopidineEarly Cutlip et al242001BMS618653 (0.9)Definite and probableTiclopidineEarly Serruys et al252001BMS60017 (2.8)DefiniteTiclopidineEarly Heller et al262001BMS185534 (1.8)DefiniteTiclopidineEarly and late Orford et al272002BMS450923 (0.5)Definite and probableTiclopidine, clopidogrelEarly Wang et al282002BMS119120 (1.7)DefiniteTiclopidineEarly and late Wenaweser et al292005BMS605895 (1.6)DefiniteTiclopidine, clopidogrelEarly and late Lee et al302005BMS15979 (0.5)DefiniteClopidogrelEarly Lee et al302005BMS14151 (0.1)DefiniteClopidogrel and cilostazolEarlyDES Serruys et al311998Hepacoat4141 (0.2)DefiniteTiclopidineEarly Mehran et al322003Hepacoat2002 (1)Definite and probableNoneEarly Ong et al332005SES, PES200620 (1.0)DefiniteClopidogrelEarly31 (1.6)Definite and probableClopidogrelEarly Iakovou et al342005SES, PES222929 (1.3)Definite, probable, and possibleClopidogrel, ticlopidineEarly and late Kuchulakanti et al352006SES, PES297438 (1.3)DefiniteClopidogrelEarly and late Rodriguez et al362006SES, PES2255 (2.2)DefiniteClopidogrelEarly, late, and very late7 (3.1)Definite and probableClopidogrel Urban et al372006SES15 157126 (0.9)Definite and probableClopidogrelEarly and late Park et al382006SES, PES191115 (0.8)Definite, probable, and possibleClopidogrelEarly, late, and very late Daemen et al112007SES, PES8146152 (1.9)DefiniteClopidogrelEarly, late, and very lateDefinition of STST, typically encountered early postoperatively, generally translates into a clinical syndrome consisting of acute onset of chest pain with ischemic ECG changes in the target vessel territory. The proof is thrombotic stent occlusion (angiography or autopsy). Definitions of ST range from “angiographically proven” to “clinically suspected” ST with the inclusion of MI involving the target vessel to unexplained death (within 30 days or anytime). Although the first definition characterizes a well-defined mechanism limited to a select patient population undergoing angiography at the time of ST and hence underestimates the true incidence of ST, the others include events related to disease progression and arrhythmia and therefore overestimate the true incidence. Accounting for these limitations, an academic research consortium proposed a new standardized definition of ST (Table 2).39 It is based on 2 principles: level of certainty that ST is underlying mechanism of adverse event and time of adverse event relative to index procedure. Table 2. Definition of ST as Proposed by the Academic Research Consortium39*The incidental angiographic documentation of stent occlusion in the absence of clinical signs or symptoms (silent occlusion) is (for this purpose) not considered a confirmed stent thrombosis.Definite ST Definite stent thrombosis is diagnosed when either angiographic or pathological confirmation is present Angiographic confirmation of ST* The presence of a thrombus originating in the stent or in the segment 5 mm proximal or distal to the stented region and at least one of the following criteria within a 48-h time window: Acute onset of ischemic symptoms at rest (typical chest pain >20 min) New ischemic ECG changes suggestive of acute ischemia Typical rise and fall in cardiac biomarkers Pathological confirmation of stent thrombosis Evidence of recent thrombus within the stent determined at autopsyProbable ST Clinical definition of probable ST is diagnosed after intracoronary stenting in the following cases Any unexplained death within the first 30 d Regardless of the time after the index procedure, any MI that is related to documented acute ischemia in the territory of the implanted stent without angiographic confirmation of ST and in the absence of any other obvious causePossible ST Clinical definition of possible ST is diagnosed with any unexplained death from 30 d after intracoronary stenting until the end of trial follow-upDefinite ST (highest level of certainty) requires either angiographic or postmortem evidence of thrombotic stent occlusion. Probable ST encompasses any unexplained death within 30 days of stent implantation or any MI in the territory of the implanted stent regardless of time. Possible ST includes any unexplained death beyond 30 days until the end of follow-up. Probable ST and possible ST have been added to the definition of ST because they provide higher sensitivity in detecting safety signals. Conversely, they are less specific and therefore critically dependent on detailed data collection about the cause of death or MI to avoid overreporting of ST. Although it has been suggested that the composite of definite and probable ST represents a good balance of specificity and sensitivity, reporting of definite and overall rates with careful adjudication of late unexplained deaths has been encouraged.The second classification principle is based on the time of the adverse event relative to the index procedure (Figure 2). Early ST refers to the first 30 days after stent implantation and is further stratified into acute (<24 hours) and subacute (24 hours to 30 days). Late ST defines the time interval between 1 month and 1 year after stent implantation; very late ST includes any event beyond 1 year. The rationale of this classification is to account for different pathophysiological mechanisms that may be at work at various times. Download figureDownload PowerPointFigure 2. Timing of ST. Early ST refers to <30 days; late ST, to 1 month to 1 year; and very late ST, to >1 year after stent implantation.An additional level of information is provided by reporting whether ST occurred in the context of an intercurrent target lesion revascularization. Thus, censoring of adverse events once intercurrent revascularization procedures occurred may disadvantage devices with lower (ie, DES) compared with devices with higher (ie, BMS) reintervention rates. To avoid this form of selective reporting, it is recommended that both primary ST rates (without intercurrent target lesion revascularization) and secondary ST rates (with intercurrent target lesion revascularization) be provided. Primary ST rates identify safety signals related to the originally implanted device, whereas secondary ST rates provide information on the overall treatment strategy, taking into account the impact of frequency of repeat revascularization procedures.Frequency and Time of STEarly ST (0 to 30 days) is encountered with a similar or even somewhat lower frequency after DES compared with BMS. A 30-day definite ST rate of 1.2% with BMS (506 patients), 1.0% with sirolimus-eluting stents (SES; 1017 patients), and 1.0% with paclitaxel-eluting stents (PES; 989 patients) was found in a sequential cohort comparison.33 A meta-analysis of 6 studies comparing BMS with SES in 2963 patients reported ST rates at 30 days of 0.5% with SES and 0.6% with BMS, respectively (relative risk [RR], 0.76; 95% confidence interval [CI], 0.30 to 1.88; P=0.55).40 A pooled analysis of 5 trials using PES in 3513 patients showed early ST rates of 0.5% with PES compared with 0.6% with BMS (RR, 0.74; 95% CI, 0.31 to 1.80; P=0.51).41 Finally, a recent analysis of various definitions of ST observed no difference between SES, PES, and BMS in the propensity for early ST.16Although late ST (>1 month to <1 year) had already been observed during the BMS era, it was largely ignored. Four registries comprising 9465 BMS patients reported rates of late ST ranging from 0% to 1% (average, 0.5%) and representing nearly one third of overall ST events (average, 1.6%) (Figure 3).26,28,29,42 Confirmation of this observation came from necropsy findings in a series of 13 patients succumbing to late ST after BMS implantation.43 Stenting across ostia, plaque disruption in the peristent region, extensive plaque prolapse caused by very lipid rich plaque, and in-stent restenosis were pathological correlates of late ST. In a meta-analysis, no differences existed in the overall (0.6% versus 0.5%; odds ratio, 1.05; 95% CI, 0.51 to 2.15; P=1.00) and late (0.2% versus 0.3%; odds ratio, 99; 95% CI, 0.35 to 2.84; P=1.00) incidence of ST between DES (2602 patients) and BMS (2428 patients).44 Similarly, no significant differences existed in the incidence of late ST between SES, PES, and BMS according to various definitions as proposed by the academic research consortium.16 Whether the 2 US Food and Drug Administration–approved DES differ with respect to early and late ST has been scrutinized in several randomized trials directly comparing SES and PES and in a meta-analysis.45 No significant differences were detected for up to 1 year of follow-up (Figure 4). Taken together, BMS and DES (SES or PES) show similar rates of ST for up to 1 year after the index procedure. ST occurred within the expected range in a broad spectrum of patients and lesions, but average dual antiplatelet therapy was longer after DES than BMS. Download figureDownload PowerPointFigure 3. Overall and late ST in 4 observational studies with BMS investigating ST beyond 30 days.26,28,29,42Download figureDownload PowerPointFigure 4. Risk of ST in 9 trials directly comparing SES and PES with follow-up to 1 year. Updated from Reference 45.Case reports, observational studies, extended follow-up of trials comparing DES with BMS, and meta-analyses of randomized trials have corroborated that very late ST (>1 year) is more common with DES than BMS. The predisposition to ST of DES >1 year after implantation has been anticipated as design inherent.8 It was first documented in 4 patients who discontinued antiplatelet therapy or underwent a surgical procedure.46 Follow-up results in 826 patients assigned on a random daily basis to treatment with either BMS (n=281) or DES (n=546) indicated an increased risk of cardiac death (1.2% versus 0%; P=0.09) and MI (4.1% versus 1.3%; P=0.04) with DES during the time period of 7 to 18 months (a time when clopidogrel treatment had been discontinued),10 although overall rates of cardiac death (2.8% versus 2.5%; P=NS) and MI (6.1% versus 5.3%; P=NS) were similar at 18 months. The frequency and timing of definite ST encountered with first-generation DES have been examined in 8146 patients treated at 2 academic centers that adopted default DES use in routine clinical practice in 2002.11 Whereas the incidence of early ST (1.1%) was similar to previous reports, the incidence of late ST continued at a steady rate of 0.6% per year for up to 3 years of follow-up (Figure 5). Pooled analysis of 4 randomized trials (1748 patients) comparing SES and of 5 randomized trials (3513 patients) comparing PES with BMS revealed similar rates of protocol-defined ST up to 1 year but significantly more very late ST (SES versus BMS: 0.6% versus 0%, P=0.03; PES versus BMS: 0.7% versus 0.2%, P=0.03).12 After readjudication of all ST events in these trials according to each of the newly proposed academic research consortium categories, differences in the incidence of very late ST diminished but were still apparent.16 Another systematic review of 14 trials comparing SES with PES revealed no difference in the overall incidence of protocol-defined ST (SES, 1.5% versus BMS, 1.3%; P=0.75), but very late ST was more frequent with SES (0.3% versus 0.04%; P=0.02).14 Finally, a meta-analysis of 6675 patients with follow-up ranging from 8 to 48 months reported no difference in the overall incidence of ST between BMS and DES (0.10% versus 0.07%; RR, 1.03; 95% CI, 0.63 to 1.68; P=0.91) but a significantly higher rate of very late ST in disfavor of DES (0.5% versus 0%; RR, 5.02; 95% CI, 1.29 to 19.52; P=0.02).9 Accordingly, very late ST is a distinct clinical entity complicating the use of first-generation DES while being exceedingly rare after BMS. It remains to be determined whether the yearly rate of 0.2% to 0.6% persists beyond 3 years, whether DES in off-label settings is associated with higher rates of very late DES, and whether newer-generation DES have a more favorable risk profile. Download figureDownload PowerPointFigure 5. Frequency distribution over time of ST in 152 of 8146 consecutive DES patients. Early ST cases are gray (91 patients); late ST cases (61 patients) are black. The line shows the cumulative number of events over time. Reproduced from Daemen et al,11 copyright © 2007, with permission from Elsevier.Clinical Sequelae of STST results in abrupt closure of the stented artery with the associated risk of MI and death.24 The impact of ST is influenced by the myocardial area at risk, its viability, the degree of instantly recruitable collaterals, and rapid reperfusion therapy. Moreover, ST may be responsible for late complications of MI, including heart failure, arrhythmias, or mechanical complications.Mortality after ST is high. A pooled analysis of multicenter BMS trials24 and a single-center registry of 6058 BMS patients29 both observed a mortality of 7% at 30 days after definite ST. Mortality at 30 days after definite ST in DES amounted to 9% in a registry of 8146 patients11 and to 19% in a series of 2974 patients.35 A recent report of definite or probable ST in randomized clinical trials of DES versus BMS revealed similar rates of mortality for both stent types (SES versus BMS, 31% versus 33%; PES versus BMS, 32% versus 28%).16 Case fatality rates may vary across studies, depending on the definition of ST (Figure 6). Thus, rates of MI are quite similar in series of definite only and definite/possible/probable ST, whereas mortality ranges between 11% (definite ST) and 45% (definite, probable, and possible ST) at 6 to 9 months of follow-up.11,34,35 The difference in case fatality of ST is related to the broader and therefore less specific inclusion of probable and possible as opposed to only definite ST cases. Download figureDownload PowerPointFigure 6. Levels of certainty in the diagnosis of ST in DES and impact on mortality and MI. A, Definite ST in 152 of 8146 consecutive patients.11 B, Definite ST in 38 of 2974 consecutive patients.35 C, Definite, probable, or possible ST in 29 of 2229 consecutive patients.34Most ST patients develop MI (66% to 100% with DES11,16,35 and 60% to 87% with BMS) with no differences between DES and BMS.16,24,29 The consequences of ST may be grave in patients in whom multiple stents in different vessels occlude simultaneously, as has been observed in 7 of 152 patients (5%) with DES ST.11 Patients with suspected ST may show a thrombus without flow impairment resulting from spontaneous or drug-facilitated lysis and may suffer minimal or no myocardial injury. Patients suffering from ST are at significant risk of recurrent thrombotic stent occlusion; 11 of 95 patients (12%) with BMS ST and 3 of 152 patients (2%) with DES ST had recurrent ST after a first event.11,29 Impaired collateral flow after implantation of DES compared with BMS (collateral flow index, 0.15±0.10 versus 0.22±0.14; P<0.01) during 6 months of follow-up was suggested,47 but the clinical significance of this observation in patients with ST remains unclear. Definite early or late ST after DES implantation in 152 patients had comparable rates of death (13.2% versus 8.2%; P=0.24) and major adverse cardiac events (77% versus 75%; P=0.99).11The US Food and Drug Administration convened a meeting of its Circulatory System Devices Advisory Panel on December 7 and 8, 2006, to address the concern of whether the slight excess of very late ST with DES compared with BMS might be associated with increased rates of death or MI during long-term follow-up. The clinical outcomes of randomized clinical trials comparing SES and PES with BMS were analyzed in several systematic reviews and revealed no significant differences relative to death (PES, 6.1% versus BMS, 6.6%; hazard ratio [HR], 0.94; 95% CI, 0.70 to 1.26; P=0.6812; SES, 6.0% versus BMS, 5.9%; HR, 1.03; 95% CI, 0.80 to 1.30)14 or MI (PES, 7.0% versus BMS, 6.3%; HR, 1.06; 95% CI, 0.81 to 1.39; P=0.6612; SES, 9.7% versus BMS, 10.2%; HR, 0.97; 95% CI, 0.81 to 1.16)14 during long-term follow-up to 4 to 5 years.13 It was hypothesized that the small increase in very late ST with DES was balanced by a somewhat smaller early ST rate, less frequent need for repeat revascularization procedures, and fewer associated complications compared with BMS. This notion is supported by the observation of a higher rate of late and very late ST after readjudication of previously censored ST cases in patients allocated to treatment with BMS after intercurrent revascularization procedures.16 Yet a large-scale registry of 6033 DES and 13 738 BMS patients in Sweden reported similar rates of mortality (propensity score–adjusted Cox regression analysis: RR, 0.94; 95% CI, 0.83 to 1.06) and MI (RR, 0.94; 95% CI, 0.77 to 1.03) for up to 6 months, followed by an excess risk of death (RR, 1.20; 95% CI, 1.05 to 1.37) and MI (RR, 1.12; 95% CI, 0.95 to 1.32), using landmark analyses during follow-up to 3 years.15 Although groups of DES and BMS differed widely with respect to cardiovascular risk factors such as diabetes (DES, 24% versus BMS, 16%), number of stents, stent diameter, stent length, and target lesion location, limiting the value of adjustments made by propensity score analysis, concerns persist that the use of DES in more complex patient and lesion subsets not represented in the randomized clinical trials may be associated with higher adverse event rates.Risk Factors of STST is a multifactorial problem related to patient, lesion, and procedural factors and to the coagulation system and response to antiplatelet therapy (Table 3). Early ST has been viewed as a problem originating from the procedure itself. Schühlen et al,22 using a classification and regression tree, identified residual dissections, followed by length of the stented segment, as the most important predictors of ST within 30 days with BMS. The importance of residual dissections in the DES era has been reiterated when observing an increased risk of ST (6.3% versus 1.3%; P=0.01) and major adverse cardiac events (18.5% versus 11.2%; P=0.07) at 6 months in patients with or without residual dissections.48Table 3. Multifactorial Origin of STPatient factors Thickness and robustness of neointimal stent coverage Drug response/interactions Gene polymorphism Left ventricular function Acute coronary syndrome Renal failure Diabetes mellitusAntithrombotic and anticoagulation therapy Coagulation activity Inhibition of platelet aggregationProcedural factors Dissection Incomplete stent apposition Stent expansionLesion factors Vessel size Lesion length Thrombus Plaque characteristics BifurcationDevice factors Stent surface Drugs PolymerST is mediated predominantly by platelet-rich thrombi and hence platelet aggregation. ADP-induced (65±3% versus 51±2%; P<0.001) and shear-induced (40.9±12.2% versus 18.2±18%; P=0.013) platelet aggregation has been found to be increased in patients with ST compared with control subjects, suggesting increased intrinsic platelet reactivity.49,50 Moreover, impaired response to antiplatelet therapy with aspirin, not correctable by the addition of clopidogrel, was documented in patients suffering from ST.51 Whether differences in platelet reactivity and response to antiplatelet therapy affect late and very late ST is unknown.Discontinuation of antiplatelet therapy has emerged as one of the most important predictors of ST. Aspirin withdrawal was responsible for admission with an acute coronary syndrome in 51 of 1236 patients (4%) with a mean delay between aspirin cessation and hospitalization of 10±2 days.52 Notably, 10 of 51 patients (19%) presented with late and very late ST at a mean of 16±7 months after BMS implantation. Discontinuation of antiplatelet therapy was related in descending order to patient noncompliance, dental procedures, surgical procedures, and bleeding. Predictors of the composite of definite, probable, and possible ST up to 9 months after DES implantation were identified in a cohort study of 2229 patients.34 The strongest predictor of early (HR, 161; 95% CI, 26 to 998; P<0.001), late (HR, 57; 95% CI, 15 to 220; P<0.001), and overall ST (HR, 90; 95% CI, 30 to 270; P<0.001) was premature discontinuation of antiplatelet therapy: both aspirin and thienopyridine in 4 patients and thienopyridine alone in 1 patient. Similarly, discontinuation of thienopyridines was more prevalent in patients suffering from ST after DES implantation compared with control subjects (37% versus 11%; P<0.0001) and emerged as an independent predictor of overall ST in a population of 2974 patients.35 It is noteworthy that most of these data are based on only a few events, that compliance with antiplatelet therapy is difficult to assess, and that even continued dual antiplatelet therapy provides but an imperfect safety net.53 Patients may still develop early and late ST despite adherence to the prescribed antiplatelet regimen. Dual antiplatelet therapy was taken by 14 of 61 patients (23%) suffering from late ST after DES implantation, whereas only 16 (26%) were off antiplatelet therapy at the time of late ST (Figure 7).11 ST occurred late in 31 patients while on aspirin monotherapy, and most of them (30 of 31; 97%) experienced the event well after the recommended period of clopidogrel. Taken together, these data indicate that dual antiplatelet therapy is important but is no panacea for the prevention of ST with both BMS and DES, that discontinuation of either aspirin or clopidogrel should be avoided (particularly during the first 6 to 12 months after the index procedure), and that cessation of antiplatelet therapy is related mostly to noncompliance and surgical procedures.54,55Download figureDownload PowerPointFigure 7. Antiplatelet treatment at the time of DES thrombosis in 152 patients. Proportion of patients with early (left column; 91 patients) and late (right column; 61 patients) ST, respectively, treated with dual, single, or no antiplatelet therapy. Data from Daemen et al.11Similar to previous the experience with BMS, ST complicating DES implantation is influenced by several additional patient, lesion, and procedural factors. In a series of 15 157 patients treated with SES and followed up prospectively for 1 year, ST occurred in 126 patients (0.9%: early, 0.7%; late, 0.2%), and multivariate analysis identified impaired postprocedural flow impairment, insulin-dependent diabetes, calcification, total occlusions, acute coronary syndrome, and number of treated lesions as predictors of overall ST.37 Renal failure, bifurcation lesions, total stent length, and diminished left ventricular function were additional clinical predictors of ST with DES in subsequent reports.34,35 Intravascular ultrasound in 15 patients with ST compared with 45 control subjects revealed smaller minimal stent area (4.3±1.6 versus 6.2±1.9 mm2; P<0.001), reduced stent expansion (65±18% versus 85±14%; P<0.001), and residual edge stenosis (67% versus 9%; P<0.001) to be more common in the ST patients than the control subjects.56Pathogenesis of STVirmani and colleagues57 first described a case of local hypersensitivity reaction with extensive vasculitis of intima, media, and adventitia consisting predominantly of lymphocytes and eosinophils in a patient suffering very late DES thrombosis. Histopathological analysis revealed aneurysmal dilatation of the vessel wall within the stented segment with evidence of stent malapposition and thick fibrin thrombus between the stent and the arterial wall. Clinical evidence of hypersensitivity reactions stems from a registry,58 with 17 of 5783 patients reporting hypersensitivity symptoms probably or certainly related to DES.In a recent necropsy comparison of 23 DES cases with 25 BMS cases (>30 days after the index procedure), delayed healing manifested by persistent fibrin deposition and incomplete reendothelialization emerged as an important discriminator between BMS and DES.59 Endothelialization (27±26% versus 66±25% versus 90±21%) was reduced whereas fibrin scores (3.0±0.9 versus 1.9±1.1 versus 0.9±0.8) were increased in DES patients with late ST compared with patients with both patent DES and BMS. Endothelialization was nearly complete in BMS specimens examined beyond 6 months, whereas incomplete endothelialization in DES specimens persisted beyond 40 months (Figure 8). Fourteen of 21 DES patients suffered late ST, which was related to delayed healing in all patients; in addition, the following pathological mechanisms were identified: chronic inflammation/hypersensitivity reaction, stenting over major side branches or bifurcation stenting using the crush technique, malapposition related to positive arterial remodeling or incomplete stent expansion, in-stent restenosis with superimposed thrombus, and penetration of necrotic core through stent struts. The results of this analysis are confounded by selection bias because patients with late DES thrombosis were more likely to undergo autopsy for suspected late thrombosis, whereas patients with BMS were more likely to be referred for other reasons. Download figureDownload PowerPointFigure 8. Healing after BMS and DES in humans. Left, Degree of endothelialization of thrombosed DES vs patent DES and BMS as obtained from an autopsy study of 23 DES and 25 BMS. Right, Percentage of endothelialization as a function of time after the index procedure. Dashed line indicates BMS; solid line, DES. Reproduced from Joner et al,59 copyright © 2006, with permission from Elsevier.Several clinical investigations support the notion of reduced or dysfunctional endothelialization after DES implantation. Kotani et al60 compared stent strut coverage between BMS (n=22) and SES (n=15) 3 to 6 months after the procedure using intracoronary angioscopy. Struts of SES were not (grade 0, 20%) or were minimally (grade 1, 67%) covered, whereas BMS showed complete coverage in all cases (grade 2, 13%; grade 3, 87%). Although angioscopy is unable to provide histological evidence of endothelialization, it is not farfetched to conclude that both neointimal regrowth and reendothelialization were impaired by DES. Physiological evidence of dysfunctional endothelium comes from studies assessing vasomotion