Noninvasive assessment of reperfusion therapy: slow progress towards a worthwhile goal

ST-segment resolution 

Numerous studies have documented associations between the degree of resolution of ST-segment elevation (ST resolution) and patency and flow in the IRA. These studies show that patients with “complete” (≥70%) ST resolution have a 90% to 95% probability of a patent IRA.10, 11 Recently, studies comparing single-lead with multi-lead ST-segment measurements have found that predictive ability is at least as good with the less complex single-lead measurements.12, 13 Regardless of the methodology, ST resolution poorly discriminates Thrombolysis In Myocardial Infarction (TIMI) grade 2 from TIMI grade 3 flow, and up to 50% of patients with persistent ST elevation actually have a patent IRA at the time of catheterization.10, 11 In many of these patients, the “false positive” ECG diagnosis of failed epicardial reperfusion actually indicates failed myocardial/microvascular reperfusion.4, 14 Indeed, patients with normal epicardial blood flow, but persistent ST elevation, remain at high risk for death, heart failure, and arrhythmias.11, 12, 15 Although the ECG integrates both epicardial and myocardial reperfusion and has an established role in the assessment of patients after reperfusion therapy, a number of factors may limit its interpretation. First, many patients have ECGs in which ST resolution cannot be determined, due to paced rhythm, bundle branch block or severe baseline artifact.11 Second, several clinical factors may confound the interpretation of ST resolution, such as infarct location and time to therapy.16

Cardiac biomarkers 

In the early streptokinase experience described above, reperfusion of an occluded IRA was accompanied by an abrupt increase in serum creatinine kinase (CK) activity followed by an early peak in CK and CK-MB levels, findings that were attributed to “washout” of enzymes from injured cells at the time of restoration of blood flow.5, 6 Because time-to-peak CK or CK-MB can only be determined retrospectively, this variable has not helped to identify patients in a time-sensitive manner for rescue PCI. As a result, investigators have focused on the rate of rise in cardiac enzymes over the first few hours after reperfusion therapy, a time period when rescue intervention is likely to be beneficial.

Myoglobin is a small cytosolic protein that is released and cleared rapidly after myocardial necrosis. A rapid increase in the serum concentration of myoglobin over the first 1 to 2 hours after reperfusion therapy has been associated with restoration of IRA patency; for this purpose, myoglobin has performed better than CK-MB and troponins, probably due to its more favorable kinetic profile.17, 18, 19 A number of factors have limited the clinical application of these findings. As with ST resolution, absence of myoglobin washout appears to overestimate the likelihood of an occluded IRA and cannot accurately distinguish TIMI 2 from TIMI 3 flow.8, 19 Because the washout phenomenon is still poorly understood, it is not known whether both epicardial and microvascular perfusion must be restored in order for rapid biomarker release to occur from cardiac myocytes. In addition, the impact of cyclic changes in coronary blood flow on myoglobin release has not been adequately studied.

Combinations of noninvasive markers 

The limitations of each of the individual noninvasive reperfusion makers has led investigators to combine these markers in an attempt to provide a more accurate noninvasive assessment of reperfusion. Used together, these markers appear to have more value than when used individually, both for assessing status of the IRA and for prognosis.8, 20 In this issue of the Journal, investigators that have made numerous prior contributions in this area adds another piece to our understanding of reperfusion markers.21 In a small group of patients treated with streptokinase for ST-elevation MI, Jurlander et al performed continuous ECG monitoring and frequent measurements of myoglobin. They categorized patients as those with concordant evidence of reperfusion (ST resolution ≥50% and myoglobin increase ≥2.4-fold from baseline to 2 hours) and those with either no evidence of reperfusion or discordant information from the 2 noninvasive markers. Myocardial salvage was inferred by comparing the post-treatment ECG Sylvestor Score with the pretreatment Aldrich Score (a surrogate for the risk area of the infarct).21

The group with concordant evidence for reperfusion was described as having a “mirror-lake” relationship between ST elevation and myoglobin levels, which was markedly attenuated in patients with no or discordant evidence for reperfusion. The authors found that the group with concordant evidence of reperfusion had significantly greater salvage than the group with no or discordant evidence of reperfusion.21 The use of salvage as the end point, albeit with an indirect ECG measurement, is a noteworthy feature of the current study and distinguishes it from prior studies focusing on epicardial blood flow. Unfortunately, the small size of the study did not allow exploration of the groups of most interest—those who had divergent findings with respect to ST resolution and myoglobin. Additional insight into the relative weight of these 2 factors in predicting successful epicardial and myocardial reperfusion, as well as the pathophysiologic basis for their discordance, is likely to be valuable in directing the clinical response. Specifically, an evidence base is needed to guide possible interventions for those patients who have complete ST resolution but do not have myoglobin washout, and the corollary group with persistent ST elevation despite rapid myoglobin release. Future studies will need to be large, and will need to include multiple outcome measures that reflect epicardial and microvascular perfusion, infarct size/salvage, and clinical outcomes.

Conclusions 

Accurate noninvasive assessment of reperfusion therapy remains an unmet need in cardiovascular medicine. Recent work has focused on integration of existing markers, such as ST resolution and myoglobin release, each of which has intrinsic limitations. Although these markers appear to perform better together than they do individually, it is likely that they will remain relatively crude instruments even when optimally used. Furthermore, the logistical challenges of performing multiple measurements of myoglobin in “real-time” will likely limit its use to selected centers. Breakthrough will require novel approaches, such as real-time, percutanteous photometric assessment of myoglobin levels, the identification of more accurate biomarkers, or the incorporation of high quality bedside myocardial perfusion imaging.