Review

Percutaneous transluminal coronary angioplasty

The lack of clear benefit from late thrombolytic therapy was thought to result from difficulties in lyzing mature thrombi.²¹ As such, the development of mechanical reperfusion with PTCA should have enabled improved testing of the late open-artery hypothesis. Despite considerable promise, low initial success rates and high rates of reocclusion in initial PTCA trials meant that the benefit seen with late mechanical reperfusion was only mild.^{4, 5}

Pizzetti and colleagues demonstrated that, in patients with an acute anterior MI, late PTCA of lesions in the left anterior descending artery (performed within 18 days [mean 15 1 days] of the MI) resulted in improved left ventricular ejection fraction (LVEF) and a reduced degree of left ventricular dilation at 6 months post-MI.⁵ Horie and colleagues performed a similar study, randomizing 83 patients with an anterior STEMI to PTCA or no PTCA performed more than 24 h after symptom onset.⁴ Left ventricular volumes were significantly smaller at 6-month follow-up and cardiac events significantly less frequent during a mean follow-up period of 50 24 months in the PTCA group than in the non-PTCA group. The data from these two studies were taken as evidence that late mechanical reperfusion with PTCA provided long-term beneficial effects in patients with anterior STEMI (Table 1).^{4, 5}

In an effort to identify patient populations that would benefit from late mechanical reperfusion, the DANish trial in Acute Myocardial Infarction-1 (DANAMI-1) study evaluated the utility of a late invasive strategy (2–10 weeks after a first acute MI) in patients with mild symptoms or inducible ischemia who had initially received thrombolytic treatment.²² The invasively treated group experienced reduced incidences of reinfarction and angina compared with the medically managed group.²² A more recent trial, the Swiss Interventional Study on Silent Ischemia Type II (SWISSI II), focused solely on patients with silent, inducible ischemia during exercise stress testing who had documented one-vessel or two-vessel coronary artery disease late after MI. The study found that mechanical reperfusion within 3 months of the index MI reduced major cardiac events by 6.3% per year compared with medical therapy alone.²³ Although neither of these trials specifically evaluated patients with STEMI, the results suggest that revascularization could be beneficial after MI in patients with signs of continued ischemia.

Coronary stenting

In the Stenting in Chronic Coronary Occlusion (SICCO) study, the adoption of stent technology in PCI minimized the incidence of restenosis when treating totally occluded vessels as compared with PTCA.²⁴ In this trial, stenting of complete coronary artery occlusions more than 2 weeks after MI was associated with improved long-term survival, a finding that was attributed to decreased target-vessel restenosis. Results have not been consistent across trials, however (Table 1). In the Open Artery Trial (TOAT), PCI with stenting, performed 3 days to 6 weeks after anterior MI, was associated with worsened LVEF and a 44% increase in adverse event risk compared with medical therapy.⁷ Given the conflicting data, the large, randomized DEsobstruction COronaire en Post-Infarctus (DECOPI) trial was designed specifically to evaluate the clinical benefit of reperfusion by PTCA, 2–15 days after MI.⁶ Unfortunately, because of enrollment difficulties, the analysis was underpowered and failed to demonstrate a difference in mortality between the PTCA and medical therapy groups; however, a 5% benefit in LVEF was demonstrated at 6 months in the patients treated with PTCA, 80% of whom received at least one stent.

Two recent, parallel trials hoped to end the controversy surrounding late reperfusion in survivors of STEMI. The Occluded Artery Trial (OAT) and the Total Occlusion Study of Canada (TOSCA)-2 trial evaluated the effects of late PCI after STEMI on clinical events and left ventricular size and function, respectively.^{8, 9} Both trials randomized patients with an occluded IRA to PCI with optimum medical management or to optimum medical management alone, 3–28 days after STEMI. In OAT, the combined primary end point of death from any cause, nonfatal reinfarction or diagnosis of NYHA class IV heart failure, occurred in 17.2% of the PCI group and in 15.6% of the medical therapy group (P = 0.18).⁹ When components of the primary end point were evaluated individually, there was a trend towards higher reinfarction rates in the PCI group.⁹ In TOSCA-2, the increase in LVEF was similar in both treatment groups (4.2 8.9% with medical management plus PCI, and 3.5 8.2% with medical management alone; P = 0.47); however, patients with a patent IRA had greater increases in LVEF than did those with an occluded artery, regardless of treatment (absolute difference 3.0%; P = 0.003).⁸ In these large, randomized trials, which had high adherence to current ACC/AHA guideline recommendations,²⁵ reperfusion by use of PCI with stenting 3–28 days after STEMI was not associated with a mortality benefit⁹ or improvement in LVEF⁸ compared with medical therapy alone. These findings seem to refute the late open-artery hypothesis.

Concurrent medical management

Given the numerous contradictions within the literature regarding the utility of late reperfusion, it is beneficial to analyze the data in greater detail. Current ACC/AHA guidelines for the management of patients with STEMI recommend treatment, unless otherwise contraindicated, with aspirin, -blockers and angiotensin-converting enzyme (ACE) inhibitors, in the acute setting, in addition to oxygen and nitrates.²⁵ Unfractionated or low-molecular-weight heparin, and/or glycoprotein IIb/IIIa inhibitors, can be administered alongside reperfusion. Moreover, when bare-metal coronary stents are used, clopidogrel should be started and continued for at least 1 month; with drug-eluting stents this period should be extended to at least 1 year.^{25, 26} Data also support the use of statins in the acute phase after MI.²⁷ Recent trials have adhered to the above recommendations; however, earlier trials frequently did not. As these therapies can alter the course of coronary artery disease, improving morbidity and mortality after STEMI (Table 2), the different outcomes observed in trials that have evaluated reperfusion after STEMI might reflect differences in medical management. Current medical regimens could either negate potential benefits of late reperfusion by altering disease progression through mechanisms similar to those of reperfusion, or simply improve patient outcomes such that invasive strategies have less impact.

Table 2 Effects of medical therapies.

Full tableFigures & Tables indexDownload Power Point slide (254K)

-Blocker therapy

-Blockers are associated with numerous benefits during the post-MI period.^{28, 29, 30, 31} In ISIS-1, patients with an acute MI treated with intravenous atenolol demonstrated a 14% risk reduction in 7-day mortality compared with controls.³⁰ This mortality benefit was a result of limitations in infarct size and reductions in the rates of reinfarction, arrhythmia and cardiac arrest, due to -blockade.^{28, 29, 30, 31} As these mechanisms are thought to be responsible for many of the theoretical benefits described in the late open-artery hypothesis,¹⁹ intensive use of -blockers after STEMI might make demonstrating the benefit from a late-opened IRA more difficult. This argument is supported by data demonstrating that patients with an occluded IRA treated with -blockers had long-term survival similar to that of revascularized patients, presumably because of preventative effects of both therapies on sudden cardiac death.³² Additionally, only studies in which patients have had low adherence to -blocker therapy have demonstrated a mortality benefit with late mechanical reperfusion;^{4, 22, 23} studies with high -blocker use failed to demonstrate any such advantage.^{6, 7, 9} Conversely, in the 263 patients not given -blockers in OAT, the PCI and medical therapy groups had similar event rates.³³ OAT was not powered to study this subgroup of patients, however, as the trial did not detect the well-established differences in outcomes between patients who did and did not receive -blockers.³³ The potential mortality benefit gained from opening an IRA late after STEMI might, therefore, be greater in patients who are not treated with -blockers than in those receiving -blockade; however, additional data are needed to determine optimum therapies in this subset of post-STEMI patients.³⁴

Renin–angiotensin-system blockade

ACE inhibitor therapy after MI is associated with a 1% increase in survival within only 1 month of treatment;^{35, 36} similar effects have been observed with angiotensin II receptor blockers (ARBs).³⁷ Studies evaluating late mechanical reperfusion that reported high concomitant ACE inhibitor or ARB use (58–100% of patients) failed to demonstrate any risk reduction.^{5, 6, 7, 8, 9} As with -blockers, it seems that inhibition of the renin–angiotensin system (RAS) might attenuate the risk reduction attributed to late reperfusion.

In addition to reducing cardiac events after MI, inhibition of the RAS has favorable effects on cardiac remodeling after MI. The Survival and Ventricular Enlargement (SAVE) trial demonstrated that when treated with captopril, patients with an LVEF below 40% were protected from the development of severe heart failure and subsequent hospitalization.³⁶ Additionally, ACE inhibitor therapy slows or reverses left ventricular dilation, reduces filling pressures and enhances exercise tolerance.³⁸ Improvements in left ventricular volumes and ejection fraction gained from late reperfusion are not negated with inhibition of the RAS; late mechanical reperfusion improves left ventricular dimensions and ejection fraction as compared with conservative management, despite high use of ACE inhibitors and/or ARBs.^{5, 6, 8} More data are needed to determine whether these mild effects on left ventricular volumes hold long-term clinical significance.

Glycoprotein IIb/IIIa inhibitors

Treatment with glycoprotein IIb/IIIa inhibitors might benefit patients with STEMI undergoing reperfusion therapy with PCI.^{25, 39} Data from the Abciximab Before Direct Angioplasty and Stenting in Myocardial Infarction Regarding Acute and Long-term Follow-up (ADMIRAL) trial suggest that abciximab administration before PCI in patients with STEMI improves LVEF, increases IRA patency before and after PCI, and reduces the incidence of death, reinfarction and subsequent target-vessel revascularization.³⁹ These benefits are the result of inhibition of thrombolytic-mediated platelet activation and of platelet-thrombus embolization into the microcirculation.⁴⁰

Glycoprotein IIb/IIIa inhibitors have potential added benefits in patients undergoing late reperfusion after STEMI. The no-reflow phenomenon, defined as poor coronary perfusion despite the absence of macroscopic coronary occlusion, is caused by distal embolization of thrombus and plaque, which clog the microcirculation; this phenomenon occurs commonly in patients with totally occluded arteries.⁴¹ Glycoprotein IIb/IIIa inhibitors limit platelet aggregation and interaction with the microvasculature, thereby limiting the no-reflow phenomenon and improving microvascular perfusion.⁴² In OAT and TOSCA-2, a high proportion of subjects (>70%) were treated with glycoprotein IIb/IIIa inhibitors and, as such, the rates of procedure-related reinfarction were remarkably low, at 0.6% for both.^{8, 9}

Statins

Statin therapy reduces the risk of further cardiovascular events considerably when administered early after acute MI.^{24, 43} The potential mechanisms by which statins exert this effect include decreased thrombogenicity,⁴⁴ reduced inflammation,⁴⁵ stabilized vulnerable plaques,⁴⁶ improved endothelial function⁴⁷ and regression of atherosclerotic lesions.⁴⁸ In patients undergoing PCI after MI, chronic pretreatment with statins inhibits the no-reflow phenomenon.⁴⁹ After STEMI, statin therapy can prevent mortality from nonsustained ventricular tachycardia.⁵⁰ Interestingly, only the study by Steg and colleagues demonstrated benefit of late reperfusion concomitant with high-dose statin use.⁶ Clearly, statin therapy lessens postinfarct mortality; however, as many trials do not sufficiently report statin usage, it is unclear whether statin therapy attenuates or negates the effects of late mechanical reperfusion.

Left ventricular ejection fraction

LVEF is a useful predictive marker of post-STEMI outcomes. In the prethrombolytic era, cardiac mortality gradually increased as LVEF decreased below 40%.⁵¹ Additionally, patients with left ventricular systolic dysfunction who had received thrombolytic therapy have been shown to have increased mortality, with a dramatic decline in survival in those with a LVEF of less than 30%.⁵² OAT included patients with a LVEF of less than 50%, but only 21% of patients had a LVEF of less than 40%;⁹ consequently, results of the trial might not apply to patients with severely reduced LVEF, and additional information is needed to determine whether these patients would benefit from late mechanical reperfusion after STEMI.

Myocardial viability

Myocardium involved in a STEMI is not necessarily damaged irreversibly. Areas of myocardium perfused primarily by the IRA could be spared by spontaneous or therapy-induced reperfusion occurring before complete myocardial necrosis or in regions where collateral circulation provides some perfusion. After STEMI, these salvaged areas can become stunned because of transient ischemia, and if a low-flow state persists, dysfunction can continue as a result of myocardial hibernation.⁵³ Importantly, many studies have demonstrated that reperfusion of such viable myocardium improves left ventricular dysfunction, symptoms and prognosis,^{54, 55, 56} and that revascularization in the absence of viability does not.⁵³ Viability has been assessed with a number of techniques, including thallium-201 or technetium-99m single-photon emission CT (SPECT), 2-[¹⁸F]fluoro-2-deoxyglucose-PET, dobutamine echocardiography and dobutamine stress cardiac MRI, all of which have comparable sensitivity and specificity.⁵³ Alternatively, contrast-enhanced MRI provides additional information on the transmural extent of scarring as assessed by delayed hyperenhancement—a parameter that also corresponds to the degree of improvement after revascularization.⁵⁷ Data in this field are limited, however, for patients presenting after STEMI. The Viability-Guided Angioplasty after Acute Myocardial Infarction (VIAMI) trial provides important insight into the utility of assessing viability soon after STEMI.⁵⁸ This study evaluated patients with STEMI who did not undergo primary or rescue PCI. Those with viability, demonstrated on low-dose dobutamine echocardiography performed 48–72 h after STEMI, were randomized to PCI with stenting or to medical management.⁵⁸ Stenting of the IRA was associated with a significant reduction in the ischemic event rate. Additionally, patients without viability were found to have a low incidence of recurrent ischemic events, with 94.7% of patients surviving 6 months without experiencing an adverse event. Thus, viability testing is an accurate means of identifying patients at high risk of future cardiovascular events. Patients with persistent myocardial viability constitute a higher risk population because they possess myocardium that is at risk of suffering a recurrent ischemic event, whereas in those without viability the infarcted tissue no longer carries a risk of reinfarction or recurrent ischemia. These data support the use of viability testing in patients early after thrombolytic therapy and for late presenters who have not undergone reperfusion. Revascularization should be considered if considerable viability is demonstrated.⁵⁸

Stress testing

Exercise stress testing has numerous important roles in the assessment of patients after STEMI. ACC/AHA guidelines emphasize that exercise testing after STEMI is useful in assessing exercise capacity, identifying persistent ischemia, risk stratification for future cardiac events, evaluating the effectiveness of a medical regimen and helping to reassure patients of their functional capacity.²⁵ Recent data further support the use of stress testing in patients who are stable after STEMI. As previously mentioned, the DANAMI-I and SWISSI II trials both demonstrated beneficial effects of PTCA performed late after MI in patients with persistent ischemia on stress testing.^{22, 23} The Adenosine Sestamibi SPECT Post-Infarction Evaluation (INSPIRE) trial revealed that adenosine technetium-99m sestamibi SPECT was effective in monitoring post-infarct ischemia.⁵⁹ Furthermore, the INSPIRE study demonstrated that intensive medical management was as effective as mechanical reperfusion in reversing adenosine-induced ischemia in stable patients in the post-infarct phase, and that cardiovascular events and mortality were identical in both treatment groups.⁵⁹ Together, these data indicate that patients who are stable after STEMI can be managed medically as long as ischemia can be suppressed effectively. Presumably, those patients in whom ischemia cannot be suppressed medically could benefit from late reperfusion, although there are no specific data addressing this issue. Moreover, as recent trials excluded patients with severe ischemia on stress testing and with left main or three-vessel coronary artery disease,^{8, 9} stress testing with imaging is a viable means of identifying these high-risk patients who could still benefit from late reperfusion.⁶⁰ Consequently, stress testing could be beneficial in late-presenting patients following STEMI who do not undergo coronary angiography, in order to identify myocardium at risk of ischemia despite medical management. As with routine stress testing, the modality and protocol to be used will depend on patient characteristics. Patients able to exercise should undergo a submaximal exercise test if performed within 3–5 days after STEMI and a symptom-limited test if performed more than 5 days after STEMI. Myocardial imaging is recommended in all patients with baseline electrocardiographic abnormalities that hinder interpretation and in those in whom evaluation of infarct size or LVEF is warranted.²⁵

Serum biomarkers

Creatine kinase and cardiac troponins are biomarkers commonly used to detect myocardial injury, estimate infarct size and risk stratify patients with MI.⁶¹ Focus has shifted to include newer biomarkers that, in addition to potentially detecting disease in otherwise healthy individuals, could have an important role in risk stratifying stable patients after an MI—C-reactive protein (CRP) and brain natiuretic peptide (BNP) could be effective in both of these patient populations.

C-reactive protein

The role of inflammation in atherosclerotic cardiovascular disease is well established. As such, serum markers of inflammation have been evaluated for their ability to predict cardiovascular events. CRP has emerged as the leading biomarker, in part because of the availability of a standardized high-sensitivity CRP assay that is both consistent and reproducible.⁶² After STEMI, CRP seems to predict mortality and major cardiovascular events; however, controversy exists regarding the ideal time for testing.^{63, 64, 65, 66, 67} CRP levels peak 48–72 h following an MI; therefore, some investigators advocate, and have demonstrated the predictive value of, testing CRP levels approximately 48 h after admission, theorizing that CRP level on admission might only reflect chronic vascular inflammation, and not inflammation caused by the MI.^{63, 64} Dimitrijevic and colleagues showed that admission CRP levels are not predictive of adverse outcomes;⁶⁴ however, other investigators have demonstrated that CRP levels on admission predict not only survival, but also the response to reperfusion therapies.^{65, 66} Perhaps these conflicting results are the result of differences in the progression of MI in the different populations studied. As the predictive value of CRP is related to its ability to estimate the degree of myocardial necrosis, patients in whom coronary occlusion has a stuttering course could have already mounted an inflammatory response upon presentation and, therefore, possess elevated CRP levels. By contrast, those who have sudden vessel occlusion might present earlier in the course of their infarction when CRP levels remain low. Steg and colleagues demonstrated that CRP levels before hospital discharge are a poor predictor of cardiovascular events after MI and suggest that inflammation at this time point could be confounded by PCI-triggered inflammation.⁶⁷ Taken together, these data suggest that CRP holds great prognostic value and could be beneficial in patients presenting late after STEMI who have not undergone PCI.

Considerable data support the ability of statins to lower CRP levels after MI, independent of effects on serum LDL cholesterol.^{43, 45} Less-substantial data demonstrate that aspirin therapy might also lower CRP levels.⁶⁸ Given the ubiquitous nature of aspirin and statins in the management of STEMI, however, it is not clear what additional therapeutic intervention would help lower risk in post-STEMI patients with increased CRP levels. Interestingly, the peroxisome proliferator-activated receptor gamma agonist pioglitazone lowers CRP levels in both patients with and those without diabetes.⁶⁹ Whether this reduction translates into improved clinical outcomes in coronary patients has not yet been established. Directly inhibiting CRP is an emerging therapeutic option that remains in its infancy.⁷⁰

Brain natiuretic peptide

BNP is a neurohormone released by the ventricular myocardium in response to dilation and pressure overload. BNP and the N-terminal fragments of its prohormone, N-terminal pro-BNP (NT-proBNP), have been associated with adverse outcomes after MI.^{71, 72} In the A to Z Trial, elevations in BNP and failure of BNP levels to decline were associated with higher incidence of combined death and new heart failure in both patients with non-STEMI and those with STEMI.⁷¹ A similar study showed that BNP elevation a few days after onset of ischemic symptoms predicted 10-month mortality regardless of subtype of MI.⁷² The prognostic value of BNP and NT-proBNP is thought to result from the hormone's unique characteristic of rising in response to left ventricular dysfunction and dilation. In fact, in patients with non-STEMI, a sudden drop in NT-proBNP could indicate response to clinical therapy, establishing a potential role for serial testing that could drive clinical management;⁷³ however, similar data do not exist for STEMI. NT-proBNP levels might also correlate with residual ischemia after STEMI, a finding that potentially supports more-aggressive management in patients with persistently elevated NT-proBNP levels after STEMI.⁷⁴ Further data are needed to determine the optimum role of BNP and NT-proBNP testing, and to identify therapeutic options for post-STEMI patients with elevated BNP levels.