European Heart Journal Advance Access originally published online on September 21, 2005
European Heart Journal 2005 26(23):2493-2519; doi:10.1093/eurheartj/ehi455
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Peri-procedural myocardial injury: 2005 update
Department of Internal Medicine, Mayo Clinic Rochester, 200 First Street S.W., Rochester, MN 55905, USA
Received 15 December 2004; revised 4 July 2005; accepted 25 July 2005; online publish-ahead-of-print 21 September 2005.
* Corresponding author. Tel: +1 507 255 5890; fax: +1 507 255 1824. E-mail address: herrmann.joerg{at}mayo.edu
 |
Abstract |
|---|
Top
Abstract
Definition and diagnosis of...During the past three decades, percutaneous coronary intervention has become one of the cardinal treatment strategies for stenotic coronary artery disease. Technical advances, including the introduction of new devices such as stents, have expanded the interventional capabilities of balloon angioplasty. At the same time, there has been a decline in the rate of major adverse cardiac events, including Q-wave acute myocardial infarction, emergency coronary artery bypass grafting, and cardiac death. Despite these advances, the incidence of post-procedural cardiac marker elevation has not substantially decreased since the first serial assessment 20 years ago. As of now, these post-procedural cardiac marker elevations are considered to represent peri-procedural myocardial injury (PMI) with worse long-term outcome potential. Recent progress has been made for the identification of two main PMI patterns, one near the intervention site (proximal type, PMI type I) and one in the distal perfusion territory of the treated coronary artery (distal type, PMI type II) as well as for preventive strategies. Integrating these new developments into the wealth of clinical information on this topic, this review aims at giving a current perspective on the entity of PMI.
Key Words: Angioplasty • Atherectomy • Cardiac troponin • Creatine kinase • Embolization • Myocardial infarction • Stents
Ever since the first coronary balloon angioplasty by Grüntzig1 in 1977, percutaneous coronary intervention (PCI) has been recognized as a valuable strategy in the management of patients having coronary artery disease (CAD). Owing to improvement in interventional techniques and expertise as well as the introduction of new devices such as stents, the incidence of major peri-procedural ischaemic complications [Q-wave acute myocardial infarction (AMI), coronary artery bypass grafting (CABG), and cardiac death] was reduced from initially 9% to currently less than 2%.2,3 However, the incidence of post-procedural cardiac marker elevation has not substantially declined since its first serial assessment by Oh et al.4 in 1985. In analogy to AMI, these marker elevations are now considered to reflect peri-procedural myocardial injury/infarction (PMI).5
A PubMed/Medline search was performed to identify peer-reviewed original and review articles, which were published over the past 27 years with the main indexing terms PCI, angioplasty, atherectomy, stents, marker, creatine kinase, cardiac troponin, ECG, chest pain, and outcome. Observational studies on incidence and randomized studies on prevention of PMI were listed and reviewed with the incorporation of relevant consensus statements from the European Society of Cardiology, the American College of Cardiology, the American Heart Association, and the Society for Cardiac Angiography and Interventions. Integrating the wealth of information available, this review aims at giving a current perspective, a 2005 update, on the entity of PMI.
|
Definition and diagnosis of PMI |
|---|
The term ‘myocardial injury’ has been used to describe any kind of impairment of regular myocardial homeostasis, which can lead to reversible or irreversible alteration in myocardial function and structure. A number of different aetiologies are united in this terminology, including mechanical, infectious, toxic, and metabolic causes. In the setting of PCI, myocardial injury is mainly metabolic in origin, relating to ischaemia. The fact that currently applied diagnostic methods do not reflect all myocardial alterations taking place in quantity and quality is important. Katoh et al.,6 for example, highlighted the iceberg phenomenon of peripheral venous cTnT concentrations in their discrepancy to coronary sinus cTnT concentrations following percutaneous transluminal coronary angioplasty (PTCA). Furthermore, experimental studies indicated that reversible injury might yield positive cardiac marker results.7,8 However, Johansen et al.9 were able to demonstrate that the majority of cTnT elevations within 24 h after PCI persist at 96 h after PCI, indicating ongoing release of cTnT from the contractile apparatus and hence irreversible myocardial injury.10 In addition, two elegant magnetic resonance (MR) imaging studies identified areas of myocardial infarction as the anatomic substrate of even minor cardiac serum marker elevations after intervention.11,12 Hence, PMI is synonymous with peri-procedural myocardial infarction in most cases of post-procedural cardiac serum marker elevation. In line with current ESC/ACC guidelines, any cardiac serum marker elevation above the upper limit of normal (ULN) in a rise and fall pattern on serial blood sampling after PCI qualifies to make the diagnosis of PMI.13 Given its high specificity and sensitivity, cTnI and cTnT are the preferred serum markers of myocardial injury with peak values to be expected at 24–48 h after PCI.13 Mass assay-quantified CKMB remains as the preferred alternative with peak values at 24 h after PCI.
|
Incidence of PMI |
|---|
Over the past 20 years, more than 60 studies have assessed the incidence of PMI with all major interventional devices, not including those with positive cardiac marker acute coronary syndrome (ACS), renal insufficiency, or primary therapeutic consideration, which are discussed separately (Table 1). As demonstrated by studies that used a multilevel approach, incidence can vary remarkably depending on the choice of biomarker, marker assay, and cutoff value. Frequency of blood analyses is another confounding factor, which can explain incidence differences among the different studies. In non-selective, multimarker series, outlined in Table 1, post-procedural elevation of CKMB mass, cTnT, and cTnI>ULN can be noted in 0–47, 7–69, and 5–53% (on average 23±12, 23±11, and 27±12%), respectively.
|
|
Risk factors for PMI |
|---|
Patient-related risk factors
A total of 10 of the studies listed in Table 1 outlined that seemingly more extensive CAD is associated with a higher frequency and a larger extent of PMI. Overall, the presence of multivessel CAD at the time of intervention increases the risk of PMI by 1.3–1.8-fold in adjusted series. Kini et al.36 identified systemic atherosclerosis as an even stronger clinical predictor of PMI than multivessel and diffuse CAD [odds ratios (ORs) 1.89 vs. 1.31 and 1.41, respectively].
Older age is the only cardiovascular risk factor frequently associated with PMI yet with ORs just above 1. For comorbidities, only one of the studies listed in Table 1 indicated that the incidence of any CKMB elevation was overall highest in patients with renal failure (33.3 vs. 18.7% in the entire cohort).36 It has to be pointed out, though, that most of the mentioned studies excluded patients with renal failure because of its potentially confounding impact on cardiac marker assessment. Supplemental data were provided by Gruberg et al.82 showing that the incidence of non-Q-wave PMI (CKMB>5x ULN) was higher in patients with chronic and end-stage kidney diseases than in patients with normal renal function (19.0 and 17.6 vs. 13.8%, P<0.0001). In saphenous vein graft (SVG) intervention, in particular, the incidence of non-Q-wave PMI (CKMB>3x ULN) increased by a factor of 1.6, as estimated creatinine clearance decreased from >70 to <30 mL/min.83 However, other studies remained less conclusive in this regard.84 Anaemia (<12 g/dL in women and <13 g/dL in men) was recently identified as another comorbidity to increase the risk for PMI with an adjusted OR of 1.35 (P=0.02).85 Yet, the overall incidence of PMI was low and there was no significant trend over the haemoglobin quartiles in this study.85
Unstable angina pectoris (UAP) was pointed out as another important clinical predictor for PMI with an average adjusted OR of 1.5. Of note, patients with UAP and cardiac marker elevation upon admission, known to have worse outcome, were excluded in most of these studies given the anticipated difficulties to distinguish between the natural course of evolving AMI and PCI-related myocardial injuries. It is worth mentioning that subanalysis data from the IMPACT-II trial included 11% of patients with pre-procedural cardiac marker elevation in the PMI analysis,23 and these patients were found to have a higher frequency and greater extent of post-procedural cardiac marker elevation.23 A recent subanalysis of ACS-focused trials noted a reciprocal linearity between the median time to PCI and the extent of post-procedural CKMB elevation and more extensive post-procedural CKMB elevation among patients with enrolment infarction.29 Overall, 44% of ACS patients without pre-procedural CKMB elevation were found to have a post-procedural CKMB elevation, mainly of milder degree.29 This number is similar to the percentage of UAP patients who will have an increase in cTnI serum concentration by at least 0.1 ng/mL from admission values (35.6%) or even any re-elevation of cTnI after intervention as defined within a 24-h post-procedural window (38.6%).86,87 In contrast, less focused studies on outcome between unstable and stable angina patients provided less uniform results.
Pre-procedural elevation of C-reactive protein (CRP) serum concentration has recently emerged as a clinical predictor of PMI. Saadeddin et al.88 noted pre-procedural high-sensitive C-reactive protein (hsCRP) elevation in 41% of 85 patients with stable angina undergoing PCI, associated with a 2.27-fold higher risk of developing PMI, defined by post-procedural cTnI elevation of >2 ng/mL. In one of the largest register-based studies in the field, Ellis et al.39 confirmed a nearly 4x higher pre-procedural hsCRP level among patients developing post-procedural CKMB elevation. Even before the hsCRP era, Buffon et al.89 noted intraprocedural and in-hospital events exclusively in nearly one out of four patients with a pre-procedural CRP serum concentration of >0.3 mg/dL, mainly thrombosis and threatened or complete acute closure but also in-hospital MI. Likewise, Chew et al.90 found a remarkable separation in the 30-day event-free survival curves within the first day after PCI based on a pre-procedural CRP serum concentration
or <0.3 mg/dL, which was largely driven by the myocardial infarction rate. Upon further stratification, there was a dichotomous difference in the 30-day event rate of MI between the lowest (<0.16 mg/dL) and all other pre-procedural CRP quartiles (3.9 vs. 10.1–11.2%). Eventually, a linear correlation between pre-procedural CRP serum concentrations and peak post-procedural CKMB serum activity was noted in the GENERATION study.27 The significance of an inflammatory state was supplemented by subanalysis findings from the EPIC, EPILOG, and EPISTENT trial, showing that the incidence of post-procedural CKMB increase was 35% higher (7% absolute) in patients with a pre-procedural white blood cell count >9.5x106/L.91
Lesion-related risk factors
Compared with restenoses, intervention of de novo lesions is associated with a higher risk of PMI (ORs 1.6–1.8). Under no circumstance can this difference be more clearly seen than with SVG disease. Although SVG disease constitutes a form of accelerated atherosclerosis, associated with a high risk of PMI (adjusted ORs 2.2–2.4), a markedly reduced PMI incidence can be seen with intervention of SVG in-stent restenoses.92–94 Histology-based studies highlighted the differences in lesion composition, which most likely underlie these observations. In distinction from the cell and lipid accumulation in primary atherosclerosis, leading to friable atheroma formation, extracellular matrix accumulation, leading to firm neointima formation, characterizes restenosis, particularly in-stent restenosis.95 Lesion characteristics of de novo lesions that seem to pre-dispose to PMI include eccentric features, greater plaque and thrombus burden, evidence of plaque rupture, and adjacent major side branches.
Procedure-related risk factors
The greater PMI potential of more complex lesions is in part linked to the greater pre-disposition to procedural complications. As outlined in Table 2, side-branch occlusion has been associated with PMI most frequently with adjusted ORs in the range from 1.7 to 7.9. Dissection is the next most frequently noted procedural complication with adjusted ORs from 1.2 to 1.8, relating to threatened/acute vascular closure, which by itself bears an adjusted ORs between 1.9 and 8.0. No-reflow/slow flow is one of the most potent predictors of PMI with adjusted ORs from 4.5 to 5.8. Similarly, distal embolization is one of the strongest predictors of PMI with adjusted ORs in the range from 4.4 to 6.0.
|
In addition to the above, more complex lesions may necessitate more complex intervention. Rather than specific intervention parameters such as number, pressure, and duration of balloon inflations as well as overall procedural time and utility use, the type of intervention seems to matter. The CAVEAT-I trial was the first study to indicate a higher PMI incidence with directional coronary atherectomy (DCA), which was confirmed in the BOAT trial and conveyed to the general new device arena in the IMPACT-II trial.23–25 A number of studies, aiming at a direct comparison, subsequently confirmed an about 2x higher risk of PMI with atherectomy procedures and an about 1.2x higher risk of PMI with stenting compared with PTCA. A significantly higher incidence of PMI with multivessel intervention remains less well defined.37
In summary, it seems to be the interplay between the extent of CAD, the extent of PCI, and the extent of myocardial compromise, which determines the occurrence of PMI. Angiographic complications are the most frequently listed and strongest predictors of PMI yet occur only in the minority of procedures, leaving the majority of post-procedural cardiac marker elevations ‘cryptogenic’.
|
Pathophysiology of PMI |
|---|
Although with qualitative differences between them, either type of coronary intervention causes local plaque distribution and trauma, which bears occlusion potential on the epicardial and myocardial microvascular levels. MR imaging confirmed these two, newly classifiable, basic patterns of PMI: type I (proximal type), which is in proximity to the target lesion and may be due to side-branch occlusion, and type II (distal type), which is in the distal perfusion territory of the treated coronary artery and mainly due to structural and functional microvascular obstructions (Figure 1).11,12 The latter accounts for 50–75% of all PMIs, relating to the following mechanisms.
|
Embolism of atheromatous and thrombotic debris
Volumetric intravascular ultrasound studies have added greatly to our understanding of the mechanisms involved in lumen enlargement by various interventional techniques. In contrast to PTCA, which involves plaque redistribution alone, both stenting and rotablation involve an additional decrease in plaque volume due to plaque embolization, compression, or fragmentation, which may explain the higher incidence of PMI observed with these two techniques.96–98 Furthermore, these findings pointed towards the significance of the pre-procedural plaque burden for PMI in addition to the relative aggressiveness of the catheter-based procedure.30 Moreover, Prati et al.99 were able to show that post-procedural CKMB relates to intraprocedural reduction of plaque volume, pointing towards the significance of atheroembolization in the pathophysiology of PMI.100,101 Formal support for this concept was given by autopsy-based studies, outlining plaque debris in the myocardial microvasculature after PCI.102 More recently, clinical studies with distal protection devices confirmed mobilization of plaque material with almost any type of intervention and no significant difference between SVG and native coronary artery lesions.103–107 The average volume of retrieved debris was 20 µL. The average particle sizes were 200x80 µm with the PercuSurge balloon occlusion system and 520x240 µm with the AngioGuard filter system. These differences may relate to particle adhesion with filter device use although there is a remarkable heterogeneity in major particle size, ranging from <56 to >600 µm, among the different studies in general. Of further note, available reports have not yet shown a significant correlation between angiographic plaque area and extent of particle retrieval. Qualitatively, amorphous material accounts for most of the debris with a ratio of 3:1 for thrombotic vs. atheromatous debris; neutrophils and foam cells constitute the majority of the few cells in the retrieval.
Platelet activation and thrombosis
In a very elaborate experimental study, Bonderman et al.108 demonstrated that in vivo plaque disruption by PTCA and/or stenting causes not only shedding of debris but also even more prominently shedding of potent biofactors such as tissue factor (TF) into the coronary circulation, leading to microvascular thrombosis and no-reflow. Increase in coronary sinus plasma concentration of TF was reported to plateau 4–24 h after PTCA, followed by an increase in coronary sinus concentration of thrombin–antithrombin III complex, a specific and sensitive marker of thrombin generation.109 In the atherosclerotic plaque, TF activity resides with phosphatidylserine-containing shed membrane microparticles, which stem from apoptotic cells, mainly macrophages and leukocytes. Once released, these microparticles sustain potent procoagulant potential, and bearing cell surface marker, they may disseminate procoagulant activity.110 Elevated levels of shed membrane microparticles in the peripheral circulation were confirmed in patients with ACS, found to originate from endothelial cells, and to lead to impairment of endothelial function, including endothelium-dependent vasorelaxation.111,112 Activated platelets are another source of microparticles (‘platelet dust’), which promote binding of platelets to the subendothelium and foster leukocyte–endothelial cell and leukocyte–leukocyte binding.112 As another marker of platelet activation, elevated levels of platelet–monocyte aggregates prior to PCI were shown to increase the incidence of post-procedural cTnI elevation by a factor of 2.5.113 Post-PCI, a higher and more prolonged increase in the transcardiac gradient of the platelet surface activation markers can be seen immediately after stenting compared with PTCA, followed by a more prominent increase in the neutrophil surface activation markers at 24–48 h after stenting.114 Compared with PTCA, DCA and RA are also associated with more extensive platelet activation, relating to a reduction in myocardial tissue level perfusion of significance in the pathogenesis of PMI.115,116 This latter aspect was highlighted by Gibson et al.,117 who were able to demonstrate that patients with impaired TIMI myocardial perfusion grade (TMPG) had a 10-fold higher incidence of post-procedural CKMB elevation (41.2 vs. 4.2%, P<0.01). This association held true even when patients with side-branch occlusion were excluded and was noted in the presence of normal epicardial perfusion status. A subsequent study extended these findings to post-procedural cTnI elevation in patients with NSTEMI-ACS undergoing PTCA and stenting.118 Overall, a decline in TMPG seems to be associated with more extensive PMI and is at least in part due to thrombotic events and platelet activation.119,120
Neurohormonal activation and modulation of vascular and myocardial functions
In addition to microvascular plugging by plaque debris, thrombus, platelets, and neutrophils, myocardial microcirculatory impairment can occur due to neurohormonal mechanisms. In fact, it was noted in a rabbit hind limb model that the marked increase in distal vascular resistance and the marked decrease in distal blood flow after plaque rupture were due to severe vasoconstriction of the microvasculature rather than distal microvascular embolization.121 Indeed, abnormal vasoconstriction on the epicardial and microcirculatory level is a well-known phenomenon in the precipitation and maintenance of ischaemia in UAP.122 Coronary vasoconstriction is also frequently seen after PTCA in patients with and without acute ischaemia, which was related to abnormal autoregulation and abnormal vasoreactivity to any given stimulus, including the release of vasoactive factors from activated platelets.123 In this context, serotonin release and vasoconstriction of the distal epicardial bed seems to be even more pronounced after stenting.124 Even though, compared with serotonin, a less prominent factor for vasoconstriction after plaque rupture in the experimental setting, endothelin-1 (ET-1) remains of consideration given its potency, increased generation in the setting of platelet activation and ischaemia, and proven release from plaque deposits secondary to procedure-related mechanical stretch and pressure.125,126 Of further note, ET-1 can prolong acute coronary vasoconstriction effects mediated by 
-adrenergic stimulation. Vice versa, cardiac myocytes release a factor that stimulates vascular ET-1 production upon -1 receptor stimulation, outlining the significance of cross-talk with the sympathetic nervous system.127,128 Likely, the factor involved is angiotensin II, adding further dimension to the potentiating influence of the renin–angiotensin system upon the activity and vasoconstricting effect of the sympathetic nervous system.129–131
The significance of the sympathetic nervous system is highlighted by the fact that post-procedural vasoconstriction can be seen not only in the culprit vessel but also in other coronary vascular beds and even in the forearm circulation.132,133 These findings point to a thoracic-spinal cardio-cardiac sympatho-excitatory reflex, which can be triggered by stretching of the coronary artery wall and even by myocardial ischaemia.134 Indeed, non-selective 1- and 2- as well as selective 1-adrenoreceptor blockade with urapidil reversed vasoconstriction proximal and distal to the target lesion following PTCA, stenting, and rotational atherectomy.132,135,136 In patients with pre-procedural reduction of coronary flow reserve (CFR) <2.5, post-procedural but pre-adenosine urapidil injection also prevented the prolonged increase in baseline average peak velocity (APV), accounting for a lack of post-procedural increase in CFR.134 As shown by Hori et al.,137 microembolization can increase coronary blood flow, attributable to the release of adenosine from the myocardium scattered around the embolized regions, and it may well be that urapidil inhibited this adenosine release. Contrary to the above, in patients with normal CFR before intervention and reduced CFR 15 min after stenting, at a time when vasoconstriction can be seen along with diffuse left ventricular dysfunction, either 1- or 2-adrenoreceptor inhibition on top of adenosine potentiated the effect of adenosine, leading to CFR normalization.134 Dupouy et al.138 confirmed two patterns of post-procedural coronary flow velocity reserve (CFVR) impairment: the first related to a significantly higher baseline APV value within 10±2 min after the last balloon inflation and normalized within 24 h after PCI and the second related to a lack of increase in hyperemic APV in the setting of a normal post-procedural baseline APV and persisted over the next 24 h. The first pattern was mainly seen in patients undergoing stenting and the latter in patients undergoing PTCA, potentially indicating differences in embolization. In addition, patients with baseline impairment of the myocardial microcirculation seem to be more prone to develop post-procedural CFVR impairment, potentially relating to a lack of functional reserve to any additional insult. This remained true for the first study to attribute post-procedural cTnT and CK elevation to the group of patients, which experienced no normalization of absolute and relative CFVR due to an increase in basal APV and were noted to have a higher plaque burden prior to intervention.139 The link between post-procedural persistence of CFVR impairment and PMI was subsequently confirmed in patients undergoing PTCA experimentally.140,141
Experimental coronary injection of micropheres leads to areas of hypoperfusion with myocardial ischaemia, which are surrounded by areas of hyperperfusion secondary to adenosine-mediated vasodilation. The net effect is preserved regional myocardial blood flow yet with depressed regional myocardial contractile function, so-called perfusion-contraction mismatch as conceptualized by Heusch et al.142 TNF was identified as a key mediator, highlighted by the finding that methylprednisolone, given 30 min before or after microembolization, prevented both myocardial TNF increase and progressive contractile dysfunction, independent of leukocyte infiltration and infarct size.143,144 Worth mentioning is the fact that infarct size doubles to 5.5% of the total area at risk, whereas ventricular dysfunction resolves within the first 6 days after the embolization.144 Arras et al.145 reported on an infarct area of 17% of the total area at risk, which may relate to differences in the volume of injected micropheres and in the technique of infarction size quantification. In the clinical setting, particles released from atherosclerotic plaques are more diverse, on average 2–5x larger in size, and notably not biologically inert, which may or may not include immunogenic potential. At least in UAP, an immunological component to the inflammatory response of the myocardial microcirculation in left ventricular biopsy specimens can be seen.146
Oxidative stress
Acute myocardial increase in TNF expression may be an NF
B-related phenomenon, potentially the consequence of increased oxidative stress, even secondary to ischaemia. Indeed, increase in the concentration of isoprostanes in coronary sinus blood samples was reported after PTCA but not after diagnostic angiography.147 These findings indicate that angioplasty is associated with potentially ischaemia-related increase in oxidative stress and/or release of isoprostanes from the culprit lesion with a potentially enhancing effect on platelet adhesion and coronary vasoconstriction. Increased production of free radical species in the setting of myocardial ischaemia can lead to the formation of ischaemia-modified albumin (IMA), which may serve as a better marker of reactive oxygen species (ROS) production under these circumstances. Indeed, Sinha et al.148 were able to demonstrate increase in isoprotane concentration in peripheral plasma samples within 30 min after PCI in only 9 out of 19 patients but increase in IMA in 18 out of 19 patients who developed chest pain and ECG changes during angioplasty. Of note, cTnT concentrations remained <0.1 ng/mL (<0.05 ng/mL) in all (15) patients within 12 h of PCI.148 Hence, many patients develop myocardial ischaemia and increased endogenous oxidative stress during PCI but only a few progress to myocardial injury for reasons yet to be defined.
Inflammation
Inflammation remains the cardinal response to injury, and increase in neutrophil activation in coronary sinus samples is well described and even seems to be more prominent with new device intervention and following intervention of AHA/ACC type C lesions.114,149,150 The increase in serum CRP concentration follows the increase in serum IL-6 concentration by 12–36 h, reaching its peak value by 24 h after the procedure.151,152 Of note, Bonz et al.152 found the increase in the serum concentrations of both IL-6 and CRP to be more pronounced in patients with concomitant post-procedural cTnT elevation. In contrast, Gaspardone et al.153 excluded patients with any post-procedural cardiac marker elevation and yet found CRP elevation of >0.5 mg/dL within 48 h after PCI in 99% of patients. For this reason, it remains uncertain whether the nidus of injury and resultant inflammation is at the site of the intervention and/or within the myocardium.
|
Clinical presentation and significance of PMI |
|---|
Acute presentation of PMI
As highlighted by the initial reports on this topic, PMI remains clinically silent in the majority of patients. The next most common presentation of PMI is post-procedural chest pain, which was viewed more as an indicator of epicardial artery stretching in former years.154 Epicardial vasospasm remained of further consideration, especially as Versaci et al.155 showed intense coronary vasoconstriction with reproduction of post-procedural chest pain by intracoronary ergonovine administration and pronounced vasodilation with resolution of symptoms by intracoronary nitroglycerine injection in ergonovine-sensitive patients in the absence of post-procedural cardiac marker elevation. On the contrary, Kurz et al.156 were unable to show any difference in the incidence of post-procedural chest pain between patients randomly assigned to either nitroglycerine or saline for 12 h after stenting and any significant association between post-procedural chest pain and post-procedural cardiac marker elevation. Subsequent studies did, however, indicate an association between post-procedural chest pain and PMI, and showed that the incidence of post-procedural cardiac marker elevation is about 2.5–4x higher among patients with chest pain after PCI.157 A subanalysis from the EPISTENT trial, moreover, showed that the 30-day event rate of MI was about 7x higher in patients with post-procedural chest pain (1.4 vs. 10.0%) and exceptionally high if post-procedural chest pain occurred within the first 4 h (12x higher incidence) or in association with ST-T wave changes (22x higher incidence).158 These findings echoed in part the initial findings of Mansour et al.159 showing that the incidence of post-procedural CK elevation was 2x higher among those patients with post-procedural chest pain and 10x higher if associated with ECG changes. It was limited to this patient population that abnormalities on repeat angiography were found, and 75% of the patients requiring repeat intervention had ST-segment depression or elevation
1 mm. In about two-thirds of the patients with post-procedural chest pain, a procedural complication can be clearly identified.157 In the remaining third, it may relate to subtle microembolization and vessel wall stretch and injury, which can also lead to vasoconstriction via the aforementioned thoracic cardio-cardiac reflex pathway. Changes in the autonomous tone and ischaemia may also cause cardiac arrhythmias. Although reported to occur within a window from 4 to 32 h after experimental myocardial microvascular embolization,144 ventricular tachycardia remains less well defined in the clinical context of PMI as is hypotension. Recently, however, there has been increasing mentioning of a correlation of in-hospital mortality with post-procedural cardiac marker elevation even without any correlation with an underlying mechanism (Table 1).
Long-term implications of PMI
Following the pioneering findings of the CAVEAT-I trial, a number of studies demonstrated a correlation between post-procedural cardiac enzyme elevation and future major adverse cardiac events, holding true even for patients with chronic kidney disease and UAP (Table 1).87,160,161 Overall, the prognostic implications seem to relate mainly to cardiac mortality, potentially even related to a higher incidence of sudden cardiac deaths, and the prognostic yield seems to be more established with CKMB than with cTnT/cTnI.162–164 Among those studies listed in Table 1 with available follow-up time of at least 6 months, 18 out of 32 studies using CKMB as a cardiac marker indicated increased long-term cardiac mortality (ORs 1.1–5) compared with only three out of 28 studies using cTnT/cTnI as a cardiac marker, which seems to have greater association with subsequent AMI, recurrent chest pain, and revascularization (Table 1). Of note, one study indicated reduced target vessel revascularization in patients with PMI, presumptively due to the presence of necrotic and hence less symptomatic myocardium.165 Another study considered post-procedural cardiac marker elevation even as a ‘trade off phenomenon’ of optimal stent expansion and also indicated an overall lower 1-year mortality in patients with a stent expansion ratio of >100% although this group had a higher rate of post-procedural CKMB elevation, which would argue against an association between PMI and late mortality.166 However, a comparative analysis indicated that the relative increase in 6-month mortality with each increase in post-procedural peak CKMB level is similar for spontaneous and PCI-related myonecrosis.167 Likewise, a comprehensive meta-analysis of seven studies, including 23 230 patients altogether, showed a 1.5, 1.8, and 3.1 times higher long-term (6–34 months) mortality risk for patients with post-procedural CKMB elevation 1–3x ULN, 3–5x ULN, and >5x ULN, respectively.168 Finally, the recently published results of the prospective ‘CKMB and PCI study’ indicate a linear relationship between post-procedural CKMB elevation and 2-year mortality, but fail to confirm a prognostic merit for post-procedural cTnI elevation.68 On the basis of these findings and in line with the majority of data presented in Table 1, one may conclude that long-term prognosis relates to the degree of PMI with a certain threshold level. Randomization of patients with similar degrees of CAD and anticipated CAD progession to medical therapy with or without PCI, as done in the ongoing COURAGE trial, might be a revealing way to substantiate the incremental effect of PMI on long-term outcome and to ease the ongoing dispute on this matter.169 In line with current guidelines, it seems reasonable to consider post-procedural CKMB elevations >3x ULN clinically relevant.5,170
|
Management of acute PMI |
|---|
Although patients deemed unlikely to develop PMI by early ischaemic risk predictors or clinical risk models may be discharged as early as 4 h after absolutely uneventful elective PCI of a native coronary artery, current guidelines suggest monitoring of all patients with serial cardiac biomarker assessment at 6–12 and 24 h and ECGs immediately after PCI and in the setting of post-procedural chest pain.171–174 As outlined by Mansour et al.,159 ST-T changes in the setting of post-procedural chest pain are 100% sensitive but only 66% specific for abnormal post-PCI angiography findings and eventually only slightly more than half of these abnormalities would be subjected to repeat intervention. In contrast to primary lesions and dissections with and without abrupt closure, side branch occlusion and distal embolization are less amendable to repeat intervention and may resolve over time.175 For no-reflow situations, stabilization of haemodynamics and oxygenation as well as treatment of chest pain and vagal reactions are important elements with additional intracoronary vasodilator therapy as outlined by practice recommendations.176 ‘Rescue’ use of GP IIb/IIIa inhibitors has become increasingly en vogue but available non-randomized clinical data do not provide convincing evidence for the merit of this practice.177,178 Overall, most patients with acute PMI will not undergo repeat angiography and require conservative management per standard practice for myocardial infarction.174 Pharmacological interference with the myocardial apoptotic process itself may become a exciting option in the future. Current clinical recommendations also favour one or more days of additional monitoring and more cautious resumption of daily activities especially in the setting of higher levels of cardiac biomarker elevation.174 Given their proven benefit, discharge medications should include aspirin, likely additional clopidogrel, beta-blocker, HMG-CoA reductase inhibitors, and ACE-inhibitors.179–184
|
Prevention of PMI |
|---|
The most important intervention in PMI management is prevention and a number of procedural and pharmacological approaches have already been suggested (Table 3).
|
Mechanical approaches
Direct stenting
Based on the rationale that stenting without pre-dilation leads to less manipulation and wall injury and may even trap some plaque components, so-called ‘direct’ stenting was pursued as a strategy to reduce PMI. Although initial non-randomized register-based studies did not find a significant difference, Nageh et al. found a three times lower incidence of mild cTnI elevation (0.2–0.9 ng/mL) after direct stenting of native coronary artery lesions and Leborgne et al. reported on a 1.7 times lower incidence of CKMB elevation >4x ULN after direct stenting of SVG lesions.240–242 Among 13 randomized trials with retrievable information on procedure-related MI rate, only the DIRECTO noted a significantly lower incidence of mild (1–2x ULN) post-procedural CK elevation (Table 3). Hence, at present there is no firm evidence to support direct stenting for PMI prevention.
Distal protection devices
Interestingly enough, the first feasibility study on the PercuSurge distal balloon occlusion system in SVG intervention showed that the amount of retrieved particular matter was less with direct stenting than with conventional stenting, which was attributable to a greater release of plaque material after initial balloon pre-dilatation.104 The FilterWire EX feasibility study highlighted the learning curve to be experienced with the use of these devices and the difficulties remaining with the intervention of lesions with high plaque and/or thrombus volume.243 Nevertheless, in the first randomized trial, the incidence of no-reflow after SVG intervention was 66% lower in the distal balloon occlusion arm along with a 40% reduction in 30-day myocardial infarction rate and a trend towards lower 30-day mortality (1.0 vs. 2.3%, P=0.17).194 With no significant differences in efficacy between balloon and filter devices reported so far, it can be concluded that distal protection devices can reduce but not fully prevent PMI.195
Pre-conditioning
Laskey200 assessed the merit of mechanical ischaemic pre-conditioning and noted a 70% relative reduction in the incidence of post-procedural CKMB elevation. In as far as pharmacological pre-conditioning can convey protective effects remains to be seen.244
Pharmacological approaches
Antiplatelet agents
Given the significance of platelet activation and arterial thrombosis in the pathogenesis of PMI, antiplatelet therapy remains essential and aspirin the baseline standard. Its significance in the peri-procedural period is highlighted by the fact that aspirin resistance, noticeable in about 20%, increases the risk of post-procedural cardiac marker elevation nearly three-fold.245 Whether the difference in PMI outcome is solely due to a suboptimal level of antiplatelet therapy remains to be determined. Aspirin does, for instance, limit NFB activation, inflammation, and endothelial dysfunction, which may exert an additional benefit in PCI.246 Furthermore, it remains to be determined to which extent any of these cardioprotective properties may contribute to a long-term benefit (44% relative survival benefit at 1 year with aspirin use at the time of PCI).247 Certainly, platelet activation can be ongoing before PCI despite aspirin treatment, favouring synergistic antiplatelet therapy. Steinhubl et al.248 highlighted this theory and the significance of the duration of pre-treatment with ticlopidine in patients undergoing stenting. Compared with patients started on ticlopidine on the day of the procedure, the incidence of PMI, defined as a post-procedural CK elevation>1x ULN with CKMB 4%, was 48 and 82%, respectively, lower among patients started on ticlopidine 1–2 days and 3 days before the procedure, relating to the fact that maximal platelet inhibition is not achieved before 3 days of therapy. Similar results were obtained in another non-randomized study with a 70% reduction in the incidence of 12-h post-stenting cTnT elevation when ticlopidine was added to aspirin 3 days before the procedure rather than on the day of the procedure.249 A similar degree of reduction of 12-h post-stenting cTnT elevation was found for clopidogrel after a loading dose of 300 mg on the day of the procedure, challenging the need for an at least 6-h pre-treatment period, the effectiveness of which being outlined in different trials.202–204 Similarly, a recently published randomized trial failed to show any difference in the incidence of post-procedural elevation of either CKMB or cTnI elevation relating to the initiation of clopidogrel treatment 3 days prior to or on the day of stenting.205 The larger percentage of patients with stable angina was discussed as one potential factor for the observed differences. Additionally, there may have been a difference in the relative contribution of non-responders. Ultimately, up to 30% of patients may have insufficient platelet inhibition 24 h after a loading dose of 300 mg of clopidogrel, which may be reduced to 11% by a 2x higher loading dose.251,252 In addition, a loading dose of 600 mg achieves plateau levels of inhibition in 2 h compared with 6 h with the 300 mg dose and provides additional platelet inhibition even in patients already on clopidogrel.251 In the randomized ARMYDA-2 trial, a 600-mg loading dose of clopidogrel resulted in a 46 and 41% reduction of post-procedural CKMB and cTnI elevation compared with a standard 4–8-h pre-procedural loading dose of 300 mg.206 The benefits of more potent early clopidogrel therapy in terms of platelet aggregation inhibition and platelet–leukocyte interactions may even negate the need for standard prophylactic platelet glycoprotein IIb/IIIa receptor (GP IIb/IIIa) inhibitor therapy at least in elective interventions (Table 3).213,214,253,254 In the lower risk population of the TOPSTAR trial, additional peri-procedural tirofiban therapy reduced the number of biphasic post-procedural cTnI elevations by 66% but not the absolute increase in cTnI and peak post-procedural CK/CKMB serum concentration.222 On the contrary, in the high-risk population of the ADVANCE trial, addition of high-dose tirofiban bolus/infusion therapy after at least 48 and 6 h pre-treatment with ticlopidine or clopidogrel, respectively, resulted in a significant 33% reduction in the incidence of post-procedural cTnI elevation and a 70 and 60% reduction in the absolute post-procedural peak levels of cTnI and CKMB (Table 3).223 Likewise, non-randomized single-centre data from a 75% UAP patient population and post hoc analyses from the TARGET trial noted significant reductions in the event rate of larger-sized in-hospital and 30-day MIs from 12.5 to 3.4% and from 6.8 to 4.2%, respectively, especially if clopidogrel was started more than 6 h before the procedure.255,256 The acuity of the presenting clinical syndrome also had the greatest impact on the relative superiority of abciximab over tirofiban in the TARGET trial in the context of the merit of prophylactic GP IIb/IIIa inhibition on the reduction of procedure-related ischaemic complications (Table 3).225 Cross-reactivity of abciximab with other integrin receptors above and beyond even eptifibatide's potential, resulting in more potent anti-inflammatory effects, was considered to account for the clinical difference as outlined in an elegant subanalysis of the EPIC trial.257 Furthermore, abciximab was shown to reduce platelet–monocytes aggregate levels prior to PCI and leukocyte–platelet interaction after ischaemia and to improve microvascular flow with cardioprotective effects in experimental models of ischaemia and reperfusion.114,258–260 Although other GPIIb/IIIa inhibitors may improve endothelium-dependent vasofunction as well, abciximab was noted to preserve coronary blood flow response to acetylcholine after stenting.261,262 The aforementioned effects may account for the 39% and 20% relative risk reduction in post-procedural CKMB elevation>3xULN with abciximab use in patients without and with obvious angiographic complications, respectively.263 EPISTENT subanalysis data, furthermore, attributed a significant 29% relative risk reduction of angiographic complications to the use of abciximab.263 Noteworthy are also the TEAM pilot study results indicating a lower incidence of angiographic complications with abciximab than with tirofiban and eptifibatide, which, however, did not translate into a difference in post-procedural cardiac marker elevation.226 In high-risk conditions, >90% of platelet aggregation inhibition may be the key to success, whereas in low-risk situations it may be sufficient to increase the degree of platelet aggregation from 25% with aspirin to 50% with clopidogrel loading.214
Anticoagulants
Only one of the studies listed in Table 1 indicated that the incidence of post-procedural cTnI elevation is higher in patients without pre-procedural heparin use, and randomized trials were unable to show a reduction of PMI by pre-procedural full anticoagulation.81,227 Low-molecular-weight heparin seems to confer no protective benefit over unfractionated heparin.228,229 The same holds true for direct thrombin inhibitors such as bivalirudin, which, however, is not inferior to the combination of heparin with planned GP IIb/IIIa inhibition, either, irrespective of clopidogrel pre-treatment as outlined in the REPLACE-2 trial.233,264 The main merit of replacing heparin with bivalirudin consists in the lower haemorrhagic risk, but this benefit seems to be lost when used in conjunction with GP IIb/IIIa inhibitors as outlined in the REPLACE-1 pilot study.232 Furthermore, direct thrombin inhibitors are an essential alternative in patients with heparin-induced thrombocytopenia. Current guidelines summarize the recommendations for the use of anticoagulants in PCI, all of which are essential and none seemingly superior to the use of unfractionated heparin in terms of peri-procedural cardioprotection.265
HMG-CoA reductase inhibitors
Considering the pathophysiological elements of PMI and the non-lipid-lowering effects of HMG-CoA reductase inhibitors (‘statins’), a first-line analysis indicated that pre-procedural statin therapy reduces the incidence of larger-sized, non-Q-wave PMIs, and improves long-term outcome in line with two other retrospective single-centre studies.266–268 Most recently, this strategy was confirmed in two randomized trials (Table 3).235,236 Activation of the PI3K/Akt signalling pathway seems to be a crucial element in the cardioprotective effects of statins, leading to increased levels of activity of ecto-5'-nucleotidase and hence adenosine generation as well as fostered eNOS activity and hence NO tissue bioavailability, which is further enhanced by the antioxidant effects of statins.269–271 Synergistically, adenosine and NO lead to decreased platelet activation and aggregation, decreased leukocyte activation and infiltration, and decreased vasoconstriction and contractile dysfunction. Of note, in animal models, prolonged (1–2 weeks) treatment courses of statins led to waning of the acute cardioprotective benefit of statins in association with increasing levels of phosphatase and tensin homolog deleted on chromosome 10 (PTEN), the phosphatase that reverses the effects of PI3K.272 In view of these findings, the clinical benefit seen in patients who have been on long-term statin therapy raises again the question of plaque consolidation as a contributing mechanism to PMI prevention.273
Beta-blocker
Sharma et al.274 were the first to report that oral beta-blocker therapy prior to PCI reduces the incidence of minor (1–3x ULN) but not major (>3xULN) post-procedural CKMB elevation by 45% along with a 45% reduction in persistent and recurrent post-procedural chest pain and a 40% reduction in ST-T changes on post-procedural 12-lead ECGs. Of note, the incidence of slow flow was nearly 2x higher, the incidence of vasospasm nearly 3x higher, and the in-hospital congestive heart failure event rate 40% higher in patients on oral beta-blocker therapy before PCI, which may relate to the exaggeration of -adrenergic effects. Nevertheless, oral beta-blocker therapy before PCI was associated with lower long-term mortality, surprisingly mainly among patients without post-procedural CKMB elevation.274 Subsequent registry-based studies from the Cleveland Clinic confirmed that patients who were on oral beta-blocker therapy at the time of PCI had better long-term survival but not a lower PMI rate.247,275,276 Of note, in an experimental coronary artery embolization model, even prolonged IV therapy with metoprolol failed to achieve limitation of MI size over a 24-h period.277 Independent of prior systemic beta-blocker therapy, propanolol, injected distally to the target lesion, was noted to reduce post-procedural CKMB and cTnT elevation by 53 and 61%, respectively.237 Hence, acute regional non-selective beta-blocker therapy may be given for its acute cardioprotective effects and oral beta-blocker therapy for its long-term benefit.
Miscellaneous agents
Calcium channel blocker and ACE-inhibitors/angiotensin receptor blockers are the two ‘mainstream’ cardiovascular medications, which have not been shown to reduce PMI event rate so far. As listed in Table 3, alternative approaches have been tested with varying benefits. Still, there are a number of agents, which have proven benefit in experimental models but have not been transitioned into clinical practice.278 It remains to be seen in the years ahead which of these may give an additional benefit.
|
Summary |
|---|
The past two decades opened the chapter on PMI with summary points as outlined in Table 4. Certainly, the chapter is not closed with the hope remaining for future years to add to our understanding and consensus on the entity of PMI.
|
|
Acknowledgements |
|---|
This work was supported at least in part by the Mayo Foundation and the Kellen Foundation. My gratitude to Professor Erbel and Professor Heusch for their mentoring role in this field.
Conflict of interest: none declared.
|
References |
|---|
- Grüntzig A. Transluminal dilatation of coronary artery stenosis. Lancet 1978;I:263.
- Detre K, Holubkov R, Kelsey S, Cowley M, Kent K, Williams D, Myler R, Faxon D, Holmes D Jr, Bourassa M, Block P, Gosselin A, Bentivoglio L, Leatherman L, Dorros G, King S III, Galichia J, All-Bassam M, Leon M, Robertson T, Passamani E. Percutaneous transluminal coronary angioplasty in 1985–1986 and 1977–1981. The National Heart, Lung, and Blood Institute Registry. N Engl J Med 1988;318:265–270.[Abstract]
- Togni M, Balmer F, Pfiffner D, Maier W, Zeiher AM, Meier B; Working Group, Interventional Cardiology and Coronary Pathophysiology, European Society of Cardiology. Percutaneous coronary interventions in Europe 1992–2001. Eur Heart J 2004;25:1208–1213.
[Abstract/Free Full Text] - Oh JK, Shub C, Ilstrup DM, Reeder GS. Creatine kinase release after successful percutaneous transluminal coronary angioplasty. Am Heart J 1985;109:1225–1231.[CrossRef][ISI][Medline]
- Califf RM, Abdelmeguid AE, Kuntz RE, Popma JJ, Davidson CJ, Cohen EA, Kleiman NS, Mahaffey KW, Topol EJ, Pepine CJ, Lipicky RJ, Granger CB, Harrington RA, Tardiff BE, Crenshaw BS, Bauman RP, Zuckerman BD, Chaitman BR, Bittl JA, Ohman EM. Myonecrosis after revascularization procedures. J Am Coll Cardiol 1998;31:241–251.
[Abstract/Free Full Text] - Katoh H, Shimada T, Sano K, Inoue S, Kitamura J, Sakane T, Murakami Y, Ishibashi Y, Karino K, Masuda J. Troponin T in the coronary sinus and percutaneous transluminal coronary angioplasty related myocardial injury. Clin Exp Pharmacol Physiol 2000;27:14–17.[CrossRef][ISI][Medline]
- Heyndrickx GR, Amano J, Kenna T, Fallon JT, Patrick TA, Manders WT, Rogers GG, Rosendorff C, Vatner SF. Creatine kinase release not associated with myocardial necrosis after short periods of coronary artery occlusion in conscious baboons. J Am Coll Cardiol 1985;6:1299–1303.[Abstract]
- Feng YJ, Chen C, Fallon JT, Lai T, Chen L, Knibbs DR, Waters DD, Wu AH. Comparison of cardiac troponin I, creatine kinase-MB, and myoglobin for detection of acute ischemic myocardial injury in a swine model. Am J Clin Pathol 1998;110:70–77.[ISI][Medline]
- Johansen O, Brekke M, Stromme JH, Valen V, Seljeflot I, Skjaeggestad O, Arnesen H. Myocardial damage during percutaneous transluminal coronary angioplasty as evidenced by troponin T measurements. Eur Heart J 1998;19:112–117.
[Abstract/Free Full Text] - Remppis A, Ehlermann P, Giannitsis E, Greten T, Most P, Muller-Bardorff M, Katus HA. Cardiac troponin T levels at 96 hours reflect myocardial infarct size: a pathoanatomical study. Cardiology 2000;93:249–253.[CrossRef][ISI][Medline]
- Ricciardi MJ, Wu E, Davidson CJ, Choi KM, Klocke FJ, Bonow RO, Judd RM, Kim RJ. Visualization of discrete microinfarction after percutaneous coronary intervention associated with mild creatine kinase-MB elevation. Circulation 2001;103:2780–2783.
[Abstract/Free Full Text] - Selvanayagam JB, Porto I, Channon K, Petersen SE, Francis JM, Neubauer S, Banning AP. Troponin elevation after percutaneous coronary intervention directly represents the extent of irreversible myocardial injury: insights from cardiovascular magnetic resonance imaging. Circulation 2005;111:1027–1032.
[Abstract/Free Full Text] - Alpert JS, Thygesen K, Antman E, Bassand JP. Myocardial infarction redefined—a consensus document of the Joint European Society of Cardiology/American College of Cardiology Committee for the redefinition of myocardial infarction. J Am Coll Cardiol 2000;36:959–969.
[Free Full Text] - Abdelmeguid AE, Ellis SG, Sapp SK, Whitlow PL, Topol EJ. Defining the appropriate threshold of creatine kinase elevation after percutaneous coronary interventions. Am Heart J 1996;131:1097–1105.[CrossRef][ISI][Medline]
- Abdelmeguid AE, Topol EJ, Whitlow PL, Sapp SK, Ellis SG. Significance of mild transient release of creatine kinase-MB fraction after percutaneous coronary interventions. Circulation 1996;94:1528–1536.
[Abstract/Free Full Text] - Kong TQ, Davidson CJ, Meyers SN, Tauke JT, Parker MA, Bonow RO. Prognostic implication of creatine kinase elevation following elective coronary artery interventions. JAMA 1997;277:461–466.[Abstract]
- Ghazzal Z, Ashfaq S, Morris DC, Douglas JS, Marshall JJ, King SB III, Weintraub WS. Prognostic implication of creatine kinase release after elective percutaneous coronary intervention in the pre-IIb/IIIa antagonist era. Am Heart J 2003;145:1006–1012.[CrossRef][ISI][Medline]
- Kanaparti PK, Brown DL. Relation between coronary atherosclerotic plaque burden and cardiac enzyme elevation following percutaneous coronary intervention. Am J Cardiol 2000;86:619–622.[CrossRef][ISI][Medline]
- Liu MW, Douglas JS Jr, Lembo NJ, King SB III. Angiographic predictors of a rise in serum creatine kinase (distal embolization) after balloon angioplasty of saphenous vein coronary artery bypass grafts. Am J Cardiol 1993;72:514–517.[CrossRef][ISI][Medline]
- Spadaro JJ, Ludbrook PA, Tiefenbrunn AJ, Kurnik PB, Jaffe AS. Paucity of subtle myocardial injury after angioplasty delineated with MB CK. Catheter Cardiovasc Diagn 1986;12:230–234.[ISI][Medlin