Platelet adhesion, activation and aggregation play key roles in the initiation of intracoronary thrombosis that causes acute coronary syndromes (ACS) and ischaemic complications following coronary artery interventions, including recurrent myocardial infarction (MI) and stent thrombosis (ST). Indeed, the most feared complication related to coronary stent placement is ST.
ST might occur after implantation of either a bare metal stent (BMS) or a drug-eluting stent (DES).1,2 The risk of early ST is similar between BMS and DES, but very late ST might occur more frequently in patients receiving first-generation DES.3,4 Although a relatively low-frequency event, ST is associated with a high risk of MI (~50–70 %), repeat revascularisation and death (~20–40 %).2,5,6
Various procedure-, lesion- and patient-related factors have been implicated in the pathophysiology of ST (see Table 1). Emerging data suggest that ST-elevation myocardial infarction (STEMI) presentation is one of the most powerful predictors of ST. The high frequency of ST after primary percutaneous coronary intervention (PCI) for STEMI may be related to stent placement under increased inflammatory and prothrombotic conditions, underlying plaque composition of culprit lesion as well as thrombotic burden. Currently, the most widely accepted definition and classification of ST was proposed by the Academic Research Consortium (ARC).7 This uniform definition incorporates timing as well as diagnostic certainty (see Table 2).
This article will focus on early ST after primary PCI in STEMI and its implications on clinical practice.
Coronary ST can occur after implantation of both BMS and DES.8 Early ST is reported to occur with an incidence ranging from 0.5–2.0 %.5,9–11 These early events may be related to lesion-related and procedural factors such as edge dissection, residual thrombus or tissue protrusion, compromised flow, stent underexpansion or a combination of these.10,12
Moreover, risk of ST is directly related to the acuteness of clinical presentation, independent of stent type.13,14 In the Swedish coronary angiography and angioplasty registry (SCAAR), the risk of ST in STEMI was increased 2.5-fold relative to patients without STEMI.15
In the Harmonizing outcomes with revascularization and stent in acute myocardial infarction (HORIZONS AMI) trial, the rate of early ST (within 30 days) was 2.5 %,16 which was substantially higher than the 0.1–0.6 % early ST rates reported from the randomised controlled trials of DES in patients with stable coronary artery disease17,18 and higher than the 1.4 % early ST rate reported in moderate- and high-risk patients with non–STEMI in the Acute catheterization and urgent intervention triage strategy (ACUITY) trial.16 Similarly, Montalescot et al. demonstrated an early ST incidence of 2.5 % in patients who had undergone primary PCI for STEMI and receiving aspirin and clopidogrel.19 Finally, the Dutch stent thrombosis registry reported that out of 5,842 patients with STEMI undergoing primary PCI approximately 3.5 % presented with definite early ST.20
The high frequency of 2.5–3.5 % of early ST after STEMI may be related to the underlying plaque composition responsible for MI, the inflammatory and prothrombotic milieu, the thrombotic material and the subsequent local vascular responses to stents. Moreover, reduced epicardial flow, vasoconstriction and concerns about embolising thrombus and friable atheroma may lead to stent undersizing, with incomplete stent apposition and underexpansion during the index procedure. Another mechanical problem in patients with ACS might arise from trapped thrombus between the stent struts and the vessel wall which, in turn, might contribute to malapposition after thrombus resolution, increasing the risk of ST.21 Indeed, large thrombus burden during primary PCI for STEMI has been shown to be an independent predictor of subsequent ST.22 Finally, certain studies report on ST rates after an ‘optimal angiographic result’, whereas others report on ST in an ‘all-comer population’. The HORIZONS AMI study followed the latter and thus reported high rates of ST.
ST presents most commonly as an acute MI.23,24 Myocardial re-infarction due to ST is associated with larger thrombus burden accompanied by extensive distal embolisation and lower rates of successful catheter-based PCI reperfusion.22–24 An Italian multicentre registry treating 110 patients with ST using an invasive approach reported that optimal reperfusion could only be achieved in approximately two-thirds of patients resulting in a high overall rate of 21 % major adverse cardiac and cerebrovascular events and a mortality rate of 12 % at 30 days.24
Similar outcomes were reported from a Spanish multicentre registry of 301 angiographic-documented DES-associated ST with in-hospital mortality rates of 11.6 %.5 Finally, in a series of 431 definite ST cases, subsequent death or recurrent ST occurred in 18 % in a 30-day follow-up and patients remained at increased risk for recurrent ST and death for up to three years with over one-quarter of patients meeting this endpoint.23
It appears that adverse clinical outcomes of ST are aggravated if the index procedure was performed for treatment of ACS as shown by the ACUITY trial, which documented that early definite or probable ST was associated with mortality in more than one-quarter of patients, MI in nearly 80 % of patients and the need for repeat revascularisation in two-thirds of patients at 30 days.25
A post-hoc analysis of the HORIZONS AMI trial found that one-third of all ST events during the three-year follow-up occurred during the index hospital stay and that patients with in-hospital compared to out-of-hospital ST had significantly greater one-year mortality (27.8 versus 10.8 %, p<0.01).26 Moreover, subgroup analysis indicated that within the in-hospital ST group, mortality after acute ST was 7.1 % and after subacute ST 50 % (p=0.0004). Multivariate analysis confirmed a significant increased one-year mortality in patients with in-hospital compared to out-of-hospital ST (adjusted HR 4.62, 95 % CI 1.98–10.77, p<0.01). Thus, subacute ST after primary PCI in STEMI may be associated with a particular poor prognosis compared to in-patients with ST after an elective stent procedure, presumably due to the successive loss of myocardium in an already injured heart. The lower mortality after acute ST can be explained by the proximity of the interventional team and the availability of an interventional suite shortly after an acute MI PCI was performed. Overall, this contemporary report found a mortality rate of around 10 % after out-of-hospital ST, which has been lower than the historical expectations.
Predictors of ST have been studied in registries and post-hoc analyses from randomised clinical trials. Table 3 summarises independent predictors associated with the occurrence of early ST.
Early ST was associated in both BMS and DES with procedure-related factors such as total stent length, multiple stents, stent underexpansion, slow flow, dissection and co-morbidities such as impaired left ventricular systolic function and diabetes and, most importantly, ACS at presentation as well as premature antiplatelet therapy discontinuation.1,2,14,27,28
Registry data of 14,120 patients undergoing either elective or emergent DES–PCI identified as independent predictors for early ST ‘presentation with acute coronary syndrome’, ‘presentation with ST-segment elevation myocardial infarction (STEMI)’, ‘renal failure’, ‘stent in left anterior descending artery’, ‘stent length’ and ‘diabetes’.5
Another registry focusing on patients receiving DES or BMS in primary PCI for STEMI reported similar factors which were independently associated with early ST with the lack of clopidogrel therapy at the time of ST being most strongly associated with subacute ST.20
The post-hoc analyses of the HORIZONS AMI trial identified in patients who underwent primary PCI for STEMI with either BMS or DES the factors ‘angiographic ulceration’, ‘impaired baseline flow’ and ‘insulin-treated diabetes’ as independent predictors of early ST while ‘clopidogrel loading dose 600 mg’ and ‘early treatment with heparin’ were associated with a reduced hazard ratio for early ST.16
Intravascular ultrasound (IVUS) allows detailed qualitative and quantitative mural arterial coronary imaging. Studies investigating the clinical outcomes of IVUS-guided primary PCI in patients with STEMI show conflicting results.29–31 However, the strongest IVUS predictor for ST in patients with stable angina was a small stent lumen as well as residual inflow/outflow disease.10,21,32 Moreover, geometrical properties of drug-eluting stent-platforms have been associated with ST.33 The IVUS substudy of the HORIZONS AMI trial found in patients undergoing primary PCI for STEMI that the two strongest predictors of definite/probable early ST were minimum lumen cross-sectional area (because of tissue protrusion after stenting) and inflow/outflow disease (due to residual stenosis or dissection) as well as a higher incidence of TIMI grade 0/1 flow at baseline and after stenting.12
Impact of Stent Type and Antithrombotic Therapy
Although there is a low risk of ST in stable patients (approximately 1 % up to three years in both BMS and DES), occurrence is accompanied with a four-fold increase in mortality.
However, in patients undergoing primary PCI for STEMI, randomised trials comparing first-generation DES to BMS reported ST rates of >3 % at one year without difference between DES and BMS.34–36 Meta-analyses were in accord with these findings showing no difference in early ST rates between DES and BMS used in primary PCI though TVR rates were consistently lower in the DES group.37,38 The ACUITY trial reported no differences in early ST between DES- and BMS-treated patients with ACS.25 It appears that ST rates are lower with second-generation everolimus-eluting stents compared to first-generation DES or zotarolimus-eluting stents with follow-ups ranging between nine and 48 months,39 with insufficient data in the setting of STEMI to draw definite conclusions. Moreover, whether newer generation drug-eluting stents or future bioabsorbable or other stent designs may influence the incidence of early ST after primary PCI remains a question for further investigation.
Currently bivalirudin is considered the anticoagulant of choice for primary PCI. The HORIZONS–AMI trial demonstrated that use of bivalirudin rather than unfractionated heparin (UFH) plus a glycoprotein IIb/IIIa inhibitor (GPI) in patients with STEMI undergoing primary PCI markedly reduced haemorrhagic complications, transfusions and thrombocytopenia and, as a result, significantly improved overall and cardiac survival.40,41 Nonetheless, it was previously reported that ST was increased with bivalirudin use within 24 hours, although not by 30 days or one year,40,41 and that ST in patients treated with bivalirudin occurred significantly less frequently than with UFH plus GPI after the 24-hour period.16 The early events with bivalirudin in this trial were evident within five hours of the procedure and were unrelated to clopidogrel loading dose, but were attenuated in patients receiving pre-randomisation heparin (in both bivalirudin and glycoprotein IIb/IIIa inhibitor groups). This raised the hypothesis that the vulnerable time frame with respect to acute ST may arise due to the rapid pharmacodynamics reversal of bivalirudin (occurs within one to two hours of stopping the infusion) and the slow onset of platelet inhibition after oral clopidogrel loading dose (takes up to six hours from oral loading dose) especially in combination with suboptimal angiographic results frequently seen in the setup of acute MI. Therefore, more intense and expanded anticoagulation (e.g. early heparin treatment in the ER or bivalirudin infusion for few hours post-PCI) and antiplatelet therapy (loading with the fast-acting antiplatelets prasugrel or ticagrelor) in this early vulnerable time phase might abate the excess of acute ST occurrence seen under bivalirudin use.
Dual antiplatelet therapy (DAPT) with aspirin and a thienopyridine is currently recommended for PCI patients. The focus of DAPT has grown in particular under the concern of ST. Currently DAPT therapy with aspirin and clopidogrel is the standard care for the majority of patients undergoing PCI irrespective of stent type. However, the main caveat of clopidogrel therapy arises through its pharmacokinetic properties by being a prodrug – hence it requires the CYP2C19 enzyme in order to be converted into its active metabolite. A growing number of studies have linked gene polymorphisms encoding the CYP2C19 allele as well as poor antiplatelet response to clopidogrel to adverse clinical outcomes, particularly coronary ischaemia and ST.42–45 Moreover, the potential reasons for reduction of early ST in STEMI using a higher loading dose of clopidogrel are not difficult to discern. It has been shown that a 600 mg dose compared with a 300 mg loading dose of clopidogrel consistently leads to greater and more rapid platelet inhibition46 and moreover has been associated with lower ischaemic event rates in other STEMI studies in addition to HORIZONS–AMI,47 particularly in subacute ST.
In patients with ACS and STEMI more rapid and potent blockade of the ADP platelet receptor with prasugrel or ticagrelor further reduce ischaemic events including early ST, although this benefit is achieved at the cost of increased risk of major bleeding not related to surgery.48,49 Nonetheless, the overall risk-benefit ratio was favourable for prasugrel within the subset of patients originally treated for acute STEMI.19
Interestingly, the marked reduction in ST by prasugrel compared to clopidogrel in patients with moderate- to high-risk ACS with planned PCI were independent of stent type and was concordant in regards to both early and late ST.48 Of note, bivalirudin has not to date been studied with either prasugrel or ticagrelor. Although such a combination should theoretically provide synergistic benefits with regard to ischaemic outcomes it remains of great interest to investigate the net clinical benefit after analysing the accompanied bleeding risk under such a combination.
Observational studies have uniformly documented that premature thienopyridine discontinuation is strongly associated with ST.2,20,50 In addition, registry data reported that the risk of adverse clinical events including death appeared greatest for patients undergoing primary PCI for STEMI who were noncompliant with clopidogrel during the first 30 days post-procedure.51 Therefore, compliance with the antiplatelet regime after DES placement – in particular because it is one of the potentially modifiable risk factors – needs to be emphasised. However, data derived from a registry reporting clopidogrel discontinuation as the strongest independent predictor of ST after DES placement found by analysing the estimate of the population-attributable risk percentage, that 68 % of ST cases in this study population were attributable to factors other than compliance with clopidogrel.28,51 Nonetheless, attention to various factors influencing patient adherence to prescribed DAPT is of fundamental importance.
Although early ST is a rare complication after stent placement in the general PCI population, patients undergoing primary PCI for STEMI have an increased risk of ST which is associated with morbidity and mortality. Great care should be given to technical details in order to optimise stent implantation and deployment, particularly in complex lesions. Moreover, the individualised patient risk-benefit ratio for ST versus bleeding risk warrants consideration when potent platelet inhibition using prasugrel or ticagrelor is desired.