ST-Segment Recovery and Outcome After Primary Percutaneous Coronary Intervention for ST-Elevation Myocardial Infarction
Insights From the Assessment of Pexelizumab in Acute Myocardial Infarction (APEX-AMI) Trial
Background— Primary percutaneous coronary angioplasty is an effective and widely adopted treatment for acute myocardial infarction. A simple method of determining prognosis after primary percutaneous coronary intervention (PCI) would facilitate appropriate care and expedite hospital discharge. Thus, we determined the prognostic importance of various measures of ST-segment–elevation recovery after primary PCI in a large, contemporary cohort of patients with ST-elevation myocardial infarction.
Methods and Results— We analyzed ECG data describing the magnitude and extent of ST-segment elevation and deviation before and early after (ie, 30 minutes) primary PCI in the study cohort of 4866 subjects with electrocardiographically high-risk ST-elevation myocardial infarction enrolled in the Assessment of PEXelizumab in Acute Myocardial Infarction (APEX-AMI) trial. Associations among 6 methods for calculating ST-segment recovery, biomarker estimates of infarct size (ie, peak creatine kinase, creatine kinase-MB, and troponin I and T), and prespecified clinical outcomes (ie, rates of 90-day death and 90-day death, heart failure, or shock) were examined. All ST-segment–recovery methods provided strong prognostic information regarding clinical outcomes. A simple ST-segment–recovery method of residual ST-segment elevation measurement in the most affected lead on the post-PCI ECG performed as well as complex methods that required comparison of pre- and post-PCI ECGs or calculation of summed ST-segment deviation in multiple leads (ie, worst-lead residual ST elevation: adjusted hazard ratio for 90-day death rate [reference <1 mm]: 1 to <2 mm, 1.23 [95% CI 0.74 to 2.03]; ≥2 mm, 2.22 [95% CI 1.35 to 3.65], corrected c-index=0.832; 90-day death/congestive heart failure/shock [reference <1 mm]: 1 to <2 mm, 1.55 [95% CI 1.06 to 2.26]; ≥2 mm, 2.33 [95% CI 1.59 to 3.41], corrected c-index=0.802). Biomarker estimates of infarct size declined in association with enhanced ST-segment recovery.
Conclusions— An ECG performed early after primary PCI is a simple, widely available, inexpensive, and powerful prognostic tool applicable to patients with ST-elevation myocardial infarction.
Received January 18, 2008; accepted July 15, 2008.
Bedside interpretation of ECG ST-segment abnormalities plays a fundamental role in assessment and decision making for patients with acute myocardial infarction (MI). Beyond determining infarct location and candidacy for acute reperfusion therapy, the extent of ST-segment elevation or deviation (sum of elevation and depression) provides important prognostic information.1 Early recovery of ST-segment deviation, in particular, has been shown to portend lower short- and long-term risk of death, recurrent ischemia, reinfarction, and congestive heart failure (CHF).2 This relationship is known to be robust and constitutes the basis for guideline recommendations promoting reassessment of ST segments 90 minutes after initiation of therapy and the possible need for rescue percutaneous coronary intervention (PCI).3,4
Editorial p 1312
Clinical Perspective p 1346
A similar relationship between ST-segment recovery and prognosis after primary PCI has been described5–8; however, these reports are retrospective, are lacking in standardized timing of post-PCI ECGs, and examine modest-sized cohorts that preceded routine use of stents, glycoprotein inhibitors, or thienopyridines. Furthermore, published reports have used differing methods for quantifying and categorizing ST-segment recovery.9 The resulting uncertainty regarding the magnitude and nature of the relationship between early ST-recovery and outcome after primary PCI likely accounts, at least in part, for the lack of routine or guideline-recommended evaluation of ST-segment recovery after primary PCI.
In a prespecified ECG study, using core laboratory analysis of protocol-specified postprocedural ECGs, we evaluated the relationship between early ST-segment recovery, enzymatic estimates of infarct size, and clinical outcomes in a large cohort of patients with ST-elevation MI (STEMI) undergoing primary PCI within the Assessment of PEXelizumab in Acute Myocardial Infarction (APEX-AMI) trial. We also compared the performance of prior methods describing ST recovery overall and in subsets with inferior versus anterior MI, to guide the practical application of the present analysis.
The APEX-AMI trial was a randomized, double-blind, placebo-controlled trial of intravenous pexelizumab administered immediately before primary PCI for electrocardiographically high-risk STEMI patients. Specific entry criteria for the APEX-AMI trial have been described previously.10 Briefly, patients were ≥18 years old, with symptom onset <6 hours, and they had qualifying ECGs that fulfilled any of the following 3 criteria: (1) ≥2 mm of ST-segment elevation (ST-E) in 2 anterior or lateral leads; (2) ≥2 mm of ST-E in 2 inferior leads coupled with ST depression in 2 contiguous anterior leads for a total ST-segment deviation ≥8 mm; or (3) new left bundle-branch block with at least 1 mm of concordant ST-E. Prospectively identified end points included 90-day mortality and the composite of death, CHF, or cardiogenic shock at 90 days. Enrollment began on July 13, 2004, and ended on May 11, 2006, which resulted in a final population of 5745 patients from 296 participating hospitals in 17 countries. No treatment difference attributable to pexelizumab was observed.11
The main protocol specified that a standard 12-lead ECG be recorded at baseline and ≈30 minutes after PCI. For the present study, only subjects (n=4866) undergoing primary PCI who had evaluable ST segments recorded on a 12-lead ECG at both time points were included. We used core laboratory ST-segment measurements exclusively (baseline and 30 minutes after PCI) and included patients regardless of whether core analysis confirmed ECG trial eligibility.12 PCI success was defined as final TIMI (Thrombolysis In Myocardial Infarction) flow grade 2 or 3 (site-reported) in the infarct-related artery, but PCI success was not required for inclusion in the present analysis.
ECGs were evaluated centrally at the ECG core laboratories (Canadian VIGOUR Centre, Edmonton, Canada; Duke Clinical Research Institute, Durham, NC) without knowledge of treatment assignment, procedural results, or clinical outcomes. ST-E was measured at the J point with magnified calipers to the nearest 0.05 mV. ST-E sums (ΣST-E) were calculated as follows: for anterior infarction, the sum of ST-E in V1 to V6, I, and aVL; for inferior infarction, the sum of ST-E in leads II, III, aVF, V5, and V6. ST-segment depression was also measured at the J point by similar methods. ST-segment deviation sums (ΣST-D) were calculated by adding the sum of ST-segment depression measured in reciprocal leads to ΣST-E. Leads II, III, and aVF were considered potential reciprocal leads for anterior infarctions, and leads V1 through V4 were considered potential reciprocal leads for inferior infarctions.
Six methods for calculating and categorizing ST-segment recovery, each mirroring a method used in prior literature, were prespecified and applied to raw ECG ST-segment data. These methods were as follows: (1) Single-lead ST elevation (ST-E) recovery (percent reduction in ST-E from baseline to post-PCI ECG in the lead with maximum baseline ST-E) analyzed in 2 categories (≥50%, <50%)9; (2) single-lead ST-E recovery, as above, but in 3 categories (≥70%, 30% to <70%, <30%)9; (3) ΣST-E recovery (percent reduction in ΣST-E from baseline to post-PCI ECG) analyzed in 2 categories (≥50%, <50%)1; (4) ΣST-E recovery, as above, but in 3 categories (≥70%, 30% to <70%, <30%)1; (5) ΣST-D recovery (percent reduction in ΣST-D from baseline to post-PCI ECG) analyzed in 3 categories (≥70%, 30% to <70%, <30%)2; and (6) worst-lead residual ST-E (the absolute magnitude of residual ST-E in the most affected lead on the post-PCI ECG, without reference to the baseline ECG) analyzed in 3 categories (<1 mm, 1 to <2 mm, and ≥2 mm).9
The primary end points of the present study were 90-day rates of death and the composite of death, centrally adjudicated cardiogenic shock, or CHF.10 Briefly, cardiogenic shock was defined as hypotension (<90 mm Hg systolic blood pressure) that lasted for at least 1 hour, was not responsive to fluid resuscitation and/or heart rate correction, was believed to be secondary to cardiac dysfunction, and was associated with hypoperfusion. CHF included new or worsening CHF that began or persisted >24 hours after randomization or rehospitalization for CHF.11
As secondary end points, associations between biomarkers of myocardial necrosis (ie, peak samples of creatine kinase [CK; μg/L], CK-MB [μg/L], troponin I [μg/L], or troponin T [μg/L]) and the above-mentioned 6 categories of ST-segment recovery were examined. Peak samples of these biomarkers were collected and processed according to local protocols. Samples for at least 1 biomarker were available for 95.1% (n=4626) of the study population (missing data: peak CK, n=1090; CK-MB, n=2917; troponin I, n=2242; and troponin T, n=3597).
Patient characteristics are reported as percentages for categorical variables and medians with 25th and 75th percentiles for continuous variables. Comparisons within the ST-segment recovery measures were made with the χ2 test for categorical variables and Mann-Whitney or Kruskal-Wallis nonparametric test for continuous variables where appropriate. Patients were excluded from analysis if ECG features that prevented ST-segment analysis were present, if baseline or post-PCI ECGs were missing, or if primary PCI was not performed.
In the analysis of nonfatal end points (ie, CHF or cardiogenic shock and their composite with death), patients were excluded if a nonfatal end point occurred before the post-PCI ECG in a landmark analysis.13 Kaplan–Meier survival estimates were used to compare time to the first occurrence of the composite of death, CHF, or shock between the groups at 90 days, and pairwise differences were tested via the log-rank test. The adjusted associations between ST-segment recovery measures and 90-day clinical outcomes were examined by development of Cox proportional hazards regression models. These models considered characteristics at randomization (ie, demographics, medical history, blood pressure, heart rate, Killip class, and ECG) and others accrued through post-PCI assessment (ie, complications, angiographic and 30-minute post-PCI ECGs) as covariates.14 Each ST-recovery measure was tested separately within the model. Adjusted hazard ratios and corresponding 95% CIs are presented for the 6 measures of ST-segment recovery; full results of the models are presented in the Appendix in the Data Supplement.
For each 90-day end point, the model likelihood ratio χ2 statistic (LRχ2) was calculated for each adjusted model of ST-recovery measure. Among the 6 models, increased strength or fit of the model would be associated with a lower LRχ2 statistic. The amount of prognostic information contributed by the ST-recovery measure was assessed by the adequacy index, which compares the predictive ability of a factor relative to that in the entire set of predictors (ie, A=LRF/LR, where F is the factor of interest and A is the adequacy index); the higher the adequacy index, the more prognostic information can be attributed to that factor.15 The adequacy of the other 5 ST-recovery measures was compared against the current clinical standard, ΣST-E resolution (50% cut point). Finally, the discriminatory power for each model was calculated and corrected for overoptimism.16 The associations between the 6 measures of ST-segment recovery and peak biomarkers were analyzed descriptively. All hypotheses were determined a priori, and as such, no adjustments were made for multiple comparisons (ie, all tests were 2-sided, with a 5% level of significance).
The authors had full access to the data and take full responsibility for its integrity. All authors have read and agreed to the manuscript as written.
Of 5745 patients enrolled in APEX-AMI, 4866 (84.7%) underwent primary PCI and had paired baseline and early post-PCI ECGs suitable for ST-segment analysis; these patients constitute the study cohort. Reasons for exclusion included missing baseline (n=12) or post-PCI (n=116) ECG, noninterpretable ST segments (ie, left bundle-branch block, ventricular pacing or rhythm) on baseline (n=146) or post-PCI (n=216) ECG, failure to meet conventional guideline-based ST-segment MI entry criteria (n=108), and no primary PCI performed (n=281). Table 1 provides key baseline characteristics and patient outcomes for the overall trial, the study cohort, and those excluded. Irrespective of cohort and as previously reported,11 there was a high usage of evidence-based therapy, which included glycoprotein IIb/IIIa inhibitors and stents. Excluded subjects were older, less frequently current smokers, less likely to be in Killip class I on admission, and more likely to be female and to have a prior cardiac history, impaired renal function, and anterior MI location. Rates of 90-day death and the composite of death/CHF/shock were also higher in excluded patients than in those included in the present study.
Protocol-specified post-PCI ECGs were performed a median (25th, 75th percentile) of 32 minutes (23 and 52 minutes) after PCI completion and 4.5 hours (3.6 and 5.8 hours) after symptom onset. In this population, epicardial flow in the infarct-related coronary artery after PCI was TIMI flow grade 3 in 4445 (91.3%), grade 2 in 298 (6.1%), grade 1 in 33 (0.7%), and grade 0 in 59 (1.2%), which resulted in a PCI success rate of 4743 (97.4%). Overall, median (25th, 75th percentile) ΣST-E, ΣST-D, and worst-lead residual ST-E on post-PCI ECGs were 3.0 (1.5, 6.0), 4.0 (1.5, 7.0) and 1.0 (0.5, 2.0), respectively.
All 6 methods for calculating and categorizing ST-segment recovery were strongly associated with 90-day composite outcome and with each of its components (Table 2). Limiting this analysis to those with successful PCI defined as TIMI flow grade 2/3 (Table 3) or as TIMI flow grade 3 (data not shown) did not appreciably weaken these relationships. In all cases and for all specified outcomes, measures that used 3 categories of ST-segment recovery (ie, ≥70%, 30% to <70%, and <30% or <1 mm, 1 to <2 mm, and ≥2 mm) identified a continuum of risk that included both lower- and higher-risk subgroups than defined by the corresponding 2-category measures (ie, ≥50% and <50%). Notably, single-lead methods performed at least as well as methods based on summation of ST-segment elevation or deviation from multiple leads, and single-lead ST-E recovery identified the highest-risk subgroup overall (90-day death rate 8.0%). Moreover, the worst-lead residual ST-E method produced a wide spectrum of risk separation that identified the lowest-risk subgroup overall (90-day death rate 1.7%) and demonstrated a >3-fold higher mortality rate in its high-risk subgroup (6.2%); this was also the largest high-risk subgroup of any of the methods.
After multivariable adjustment, ST-segment recovery remained predictive of both the 90-day death rate (Figure 1A) and the composite of death/CHF/shock (Figure 1B), regardless of which of the 6 measures were used. The model-fit statistics were all statistically significant for the 90-day outcomes across the 6 measures of ST-segment recovery. The ability of the models to separate subjects’ outcomes, ie, discriminatory power or predictive accuracy, was excellent among the models (c-index >0.83 for mortality and >0.79 for the composite), with a minimal range in values among the 6 ST-recovery measures. The amount of prognostic information (ie, adequacy index) individually contributed by the 5 ST-recovery measures relative to ΣST-E resolution (50% cut point), the current clinical standard, was also comparable (Table 4).
Kaplan–Meier curves for 2 of the measures (single-lead ST-E recovery [Figure 2] and worst-lead residual ST-E [Figure 3]) according to infarct location are presented here. In general, the 3-category method based on percent recovery (from baseline) performed less well with respect to identifying low-risk anterior MI subgroups (Figure 2B; ≥70% versus 30% to 70%: adjusted hazard ratio 0.68, 95% CI 0.41 to 1.31). However, the lower-risk anterior MI subgroups still carried a >8% event rate for 90-day death, CHF, or shock. Worst-lead residual ST-E performed substantially better by identifying a low-risk anterior MI subgroup with a 90-day composite event rate of just 3.8% (Figure 3B; <1 versus 1 to <2 mm: adjusted hazard ratio 0.52, 95% CI 0.31 to 0.90). Conversely, worst-lead residual ST-E demonstrated poorer risk distinction for subjects with inferior MI who had <1 mm versus those with 1 to <2 mm of residual ST-E (Figure 3C; <1 versus 1 to <2 mm: adjusted hazard ratio 0.92, 95% CI 0.52 to 1.62), although both of these subgroups had low absolute risk.
In Table 5, the peak measures of cardiac markers according to the 6 different ST-segment recovery measures are shown. In every instance, there was a significant inverse relationship between the degree of ST-segment recovery and the magnitude of biomarker elevation. Notably, however, the most striking gradient was observed in patients who were categorized by the worst-lead residual ST-E measure.
The present data from the largest prospectively evaluated trial of primary PCI demonstrate the power of the 12-lead ECG performed early after primary PCI for predicting death, heart failure, or shock up to 90 days after STEMI. This finding was largely independent of the method used to calculate ST-segment recovery, existed for each of the component end points, and remained strong even when the analysis was limited to those with angiographically successful PCI. Most patients undergoing primary PCI have angiographically successful procedures and on average enjoy a favorable prognosis.11,17 These overall results notwithstanding, there remain high-risk subsets of patients within such cohorts. The present data indicate that identification of patients at continued high risk despite angiographically successful PCI is possible very early after primary PCI, before other metrics such as peak biomarker values or ejection fraction may be available. Given the simplicity and low cost of ECGs, the data provide a compelling argument for the inclusion of early ST-segment recovery as a routine early metric in patients undergoing primary PCI.
Other groups have previously reported an association between ST-segment recovery and mortality after primary PCI. van't Hof and colleagues5 reported an adjusted relative risk of 6.4 (95% CI 2.7 to 15.3) for cardiovascular death at 3.1±1.9 years in 60 patients with <30% ST-segment recovery versus 204 patients with complete recovery. Claeys et al7 found persistent ST-E (ΣST ≥50% of baseline) to be associated with a 3.4-fold relative risk of major adverse cardiovascular events at 1 year in an analysis of 91 patients. Matetzky et al6 also reported a trend toward excess mortality and heart failure in 28 of 117 patients with successful primary PTCA who demonstrated persistent ST-E (ΣST ≥50% of baseline). These exploratory studies share important limitations, including a retrospective single-center design, nonsystematic ECG collection, absence of core laboratory ECG analysis, and limited statistical power (few fatal events). Moreover, they studied patients from an era that preceded the routine use of glycoprotein IIb/IIIa inhibitors or stents, therapies known to improve indices of myocardial perfusion and to reduce epicardial reocclusion after primary PCI.18,19 In contrast, ≈72% of the present cohort received glycoprotein IIb/IIIa inhibitors, and 96% received stents. Frequent use of these therapies suggests subjects in the present study were at lower risk of early coronary reocclusion than those in prior studies.
More recently, Prasad et al8 examined the relationship of ΣST-segment recovery and outcome in 695 STEMI patients with paired preprocedure and postprocedure ECGs undergoing primary PCI within 12 hours of symptom onset in the Controlled Abciximab and Device Investigation to Lower Late Angioplasty Complications (CADILLAC) trial. Approximately half of the patients studied were treated with stents and abciximab by random assignment. Failure of ST-segment recovery was associated with increased major adverse cardiovascular events but not death. Although this analysis was the first to use an independent ECG core laboratory, the timing of ECGs was not standardized, and tracings performed up to 4 hours after the procedure were permitted.
Although the relationship between ST-segment recovery and outcome in the present study was statistically robust for all methods tested, important distinctions between methods are evident. Current STEMI guidelines emphasize the use of a binary cut point (<50% versus ≥50%) for categorizing recovery of the ST segment after fibrinolytic therapy. This approach is promoted both for assessing prognosis and for determining eligibility for rescue PCI.3 Similar methods were used in studies of primary PCI cohorts by Claeys et al and Matetzky et al as noted above.6,7 Both 2-category methods tested in the present data set (single-lead ST-E recovery and ΣST-E recovery) performed well; however, we found the corresponding 3-category methods, including that originally described by Schroder et al,1 partitioned a gradient of risk that cannot be described by 2 categories alone.
Although each of the four 3-category methods we tested defined subgroups that spanned a wide spectrum of risk, inherent differences exist in the ease of their bedside application. Given its ease of calculation, the worst-lead residual ST-E may be best suited for risk stratification in the clinical setting. Features that add complexity for clinicians include requirements to sum multiple leads, assess and sum ST-segment depression, and compare preprocedure and postprocedure tracings to calculate percent ST-segment recovery. In agreement with others, we found that evaluation of ST-segment resolution from a single lead with maximal ST-E (single-lead ST-E recovery) provided comparable prognostic stratification to that achieved with complex methods, including summed ST-E or summed ST-segment deviation methods.9
We tested 2 methods that use interpretation of just a single ECG lead (single-lead ST-E recovery and worst-lead residual ST-E). Consistent with McLaughlin et al,9 we found that both provided excellent overall separation into 3 prognostic categories. The present study design allowed us to extend their observations to include resolution of the independent risk of death, as well as shock and heart failure end points. Compared with single-lead ST-E recovery, worst-lead residual ST-E is more easily applied at the bedside, because it requires no baseline ECG for comparison, and it would appear, therefore, to be the method of choice.
Differential apportioning of risk in anterior versus inferior infarctions was an unexpected finding that potentially complicates the adoption of a single method by clinicians. Assessment of prognosis for anterior infarctions was better served by worst-lead residual ST-E, because single-lead ST-E recovery failed to resolve low- versus intermediate-risk subgroups (Figures 2B and 3⇑B). The converse was true for inferior infarctions, but the consequences of poor resolution between low versus intermediate risk in the setting of inferior infarction are likely to be mitigated by the low absolute risk of both subgroups.
Categorization of ST-segment recovery generated survival curves with remarkably early separation, which implies important differences in event rates in the days immediately after primary PCI (Figures 2 and 3⇑; overall 7-day death rate 1.4%, 7-day death/CHF/shock rate 4.0%). The ability to accurately predict short-term risk in the immediate postreperfusion period speaks to the potential for early post-PCI ECGs to tailor the length of initial hospitalization according to patient risk. Moreover, accurate and early identification of high-risk patients helps define a target population for novel therapeutic interventions that are potentially applicable before other conventional objective risk markers are known. Here, methods based on the percent of ST-segment recovery performed especially well. The 90-day rates for the composite end point and for death in the high-risk anterior subgroup were 15.3% and 7.3%, respectively.
Previous studies have observed a correlation between persistent ST-E, abnormal myocardial blush grade, the degree of elevation of myocardial biomarkers, and the degree of impairment of left ventricular ejection fraction.20,21 Despite limitations of our methods for standardized collection of myocardial biomarkers, we found an inverse relationship between the extent of biomarker elevation with the category of ST-segment recovery, regardless of the biomarker or ST method used. These data are consistent with the hypothesis that ST recovery, like myocardial blush, implies effective microvascular and tissue reperfusion, with salvage of viable myocardium. Our unique observations regarding impaired ST recovery and the risk of clinically evident left ventricular dysfunction (shock and heart failure) further support this construct.
Patients with baseline or acquired conduction abnormalities or arrhythmias may not have ST segments on ECG suitable for evaluation. Prognostication by post-PCI ECG would not be possible in such cases. Similarly, our method cannot identify the highest-risk patients who deteriorate or die precipitously such that an early post-PCI ECG cannot be obtained. The enrollment criteria of APEX-AMI required high-risk ECG criteria and excluded patients who presented beyond 6 hours or who had isolated low-risk inferior infarction. The present results may not be applicable to other patients with STEMI without these characteristics.
ST-segment analysis performed early after primary PCI for acute MI is a widely available, inexpensive, and powerful prognostic tool irrespective of PCI procedural outcome and final TIMI flow grade in the infarct-related coronary artery. Measurement of residual absolute ST-E from a single (worst) lead categorized into 3 groups (<1 mm, 1 to 2 mm, and ≥2 mm) is at least comparable to more complex ECG-based methods and provides a simple metric easily incorporated into contemporary guidelines and routine practice.
We thank Cynthia Yau, MSc, and Jeff Bakal, PhD, for their statistical assistance and advice and Barbara Litwinowich, RN, BSN, for her project management.
Sources of Funding
This trial was jointly funded by Procter & Gamble Pharmaceuticals and Alexion Pharmaceuticals.
Drs Armstrong, Granger, and Van de Werf received research grants from the trial sponsors. Members of the Steering Committee received honoraria for their participation. Dr Mahaffey has received research grants and consultant fees from Alexion Pharmaceuticals. Dr Todaro is an employee of Procter & Gamble, and Dr Adams is an employee of Alexion.
Schroder R, Wegscheider K, Schroder K, Dissmann R, Meyer-Sabellek W. Extent of early ST-segment elevation resolution: a strong predictor of outcome in patients with acute myocardial infarction and a sensitive measure to compare thrombolytic regimens: a substudy of the International Joint Efficacy Comparison of Thrombolytics (INJECT) trial. J Am Coll Cardiol. 1995; 26: 1657–1664.
Antman EM, Anbe DT, Armstrong PW, Bates ER, Green LA, Hand M, Hochman JS, Krumholz HM, Kushner FG, Lamas GA, Mullany CJ, Ornato JP, Pearle DL, Sloan MA, Smith SC Jr. ACC/AHA guidelines for the management of patients with ST-elevation myocardial infarction: executive summary: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Writing Committee to Revise the 1999 Guidelines for the Management of Patients With Acute Myocardial Infarction). Circulation. 2004; 110: 588–636.
Van de Werf F, Ardissino D, Betriu A, Cokkinos DV, Falk F, Fox KAA, Julian D, Lengyel M, Neumann FJ, Ruzyllo W, Thygesen C, Underwood SR, Vahanian A, Verheugt FW, Wijns W. Management of acute myocardial infarction in patients presenting with ST-segment elevation: the Task Force on the Management of Acute Myocardial Infarction of the European Society of Cardiology. Eur Heart J. 2003; 24: 28–66.
Matetzky S, Novikov M, Gruberg L, Freimark D, Feinberg M, Elian D, Novikov I, Di Segni D, Agranat O, Har-Zahav Y, Rabinowitz B, Kaplinsky E, Hod H. The significance of persistent ST elevation versus early resolution of ST-segment elevation after primary PTCA. J Am Coll Cardiol. 1999; 34: 1932–1938.
Claeys MJ, Bosmans J, Veenstra L, Jorens P, De Raedt H, Vrints CJ. Determinants and prognostic implications of persistent ST-segment elevation after primary angioplasty for acute myocardial infarction: importance of microvascular reperfusion injury on clinical outcome. Circulation. 1999; 99: 1972–1977.
Prasad A, Stone GW, Aymong E, Zimetbaum P, McLaughlin M, Mehran R, Garcia E, Tcheng J, Cox D, Grines C. Impact of ST-segment resolution after primary angioplasty on outcomes after myocardial infarction in elderly patients: an analysis from the CADILLAC trial. Am Heart J. 2004; 147: 669–675.
McLaughlin MG, Stone GW, Aymong E, Gardner G, Mehran R, Lansky AJ, Grines CL, Tcheng JE, Cox DA, Stuckey T, Garcia E, Guagliumi G, Turco M, Josephson ME, Zimetbaum P. Prognostic utility of comparative methods for assessment of ST-segment resolution after primary angioplasty for acute myocardial infarction: the Controlled Abciximab and Device Investigation to Lower Late Angioplasty Complications (CADILLAC) trial. J Am Coll Cardiol. 2004; 44: 1215–1223.
Armstrong PW, Adams PX, Al-Khalidi HR, Hamm C, Holmes D, O'Neill W, Todaro TG, Vahanian A, Van de Werf F, Granger CB; on behalf of the APEX-AMI Steering Committee. Assessment of Pexelizumab in Acute Myocardial Infarction (APEX AMI): a multicenter, randomized, double-blind, parallel-group, placebo-controlled study of pexelizumab in patients with acute myocardial infarction undergoing primary percutaneous coronary intervention. Am Heart J. 2005; 149: 402–407.
Tjandrawidjaja MC, Fu Y, Al-Khalidi H, Todaro TG, Adams P, Van de Werf F, Granger CB, Armstrong PW. Failure of investigator adherence to electrocardiographic entry criteria is frequent and influences clinical outcomes: lessons from APEX-AMI. Eur Heart J. 2007; 28: 2850–2857.
Anderson JR, Cain KC, Celber RD. Analysis of survival by tumor response. J Clin Oncol. 1983; 1: 710–719.
Chang WC, Kaul P, Fu Y, Westerhout CM, Granger CB, Mahaffey KW, Wallentin L, Van de Werf F, Armstrong PW. Forecasting mortality: dynamic assessment of risk in ST-segment elevation acute myocardial infarction. Eur Heart J. 2006; 27: 419–426.
Nallamothu BK, Wang Y, Magid DJ, McNamara RL, Herrin J, Bradley EH, Bates ER, Pollack CV Jr, Krumholz HM; for the National Registry of Myocardial Infarction (NRMI) Investigators. Relation between hospital specialization with primary percutaneous coronary intervention and clinical outcomes in ST-segment elevation myocardial infarction: National Registry of Myocardial Infarction-4 analysis. Circulation. 2006; 113: 222–229.
Montalescot G, Barragan P, Wittenberg O, Ecollan P, Elhadad S, Villain P, Boulenc JM, Morice MC, Maillard L, Pansieri M, Choussat R, Pinton P; for the ADMIRAL Investigators. Platelet glycoprotein IIb/IIIa inhibition with coronary stenting for acute myocardial infarction N Engl J Med. 2001; 344: 1895–1903.
Grines CL, Cox DA, Stone GW, Garcia E, Mattos LA, Giambartolomei A, Brodie BR, Madonna O, Eijgelshoven M, Lansky AJ, O'Neill WW, Morice MC; for The Stent Primary Angioplasty in Myocardial Infarction Study Group. Coronary angioplasty with or without stent implantation for acute myocardial infarction, N Engl J Med. 1999; 341: 1949–1956.
van 't Hof AW, Liem A, Suryapranata H, Hoorntje JC, de Boer MJ, Zijlstra F; Zwolle Myocardial Infarction Study Group. Angiographic assessment of myocardial reperfusion in patients treated with primary angioplasty for acute myocardial infarction: myocardial blush grade. Circulation. 1998; 97: 2302–2306.
De Luca G, van 't Hof AW, de Boer MJ, Ottervanger JP, Hoorntje JCA, Gosselink ATM, Dambrink JKE, Zijlstra F, Suryapranata H. Time-to-treatment significantly affects the extent of ST-segment resolution and myocardial blush in patients with acute myocardial infarction treated by primary angioplasty. Eur Heart J. 2004; 25: 1009–1013.
In nearly 4900 patients from the APEX-AMI trial, 6 measures of ST-segment recovery after percutaneous coronary intervention (PCI) were derived from baseline and/or 30-minute post-PCI ECGs. These measures, from a core laboratory, were examined for their association with biomarker estimates of infarct size and 90-day clinical outcomes that included death and the composite of death, congestive heart failure, or shock. Incomplete ST-segment recovery, irrespective of method, was associated with poorer 90-day survival and clinical outcomes. The simple measure of residual ST-segment elevation in the most affected lead on the post-PCI ECG performed as well as more complex methods that required comparison of pre- and post-PCI ECGs or calculation of summed ST-segment deviation in multiple leads. Thus, single-lead ST-segment recovery is recommended as a simple, routine clinical tool for prognostic assessment in patients with ST-elevation myocardial infarction undergoing PCI. Such an approach may inform care decisions and healthcare resource use in these patients.
The online-only Data Supplement is available with this article at http://circ.ahajournals.org/cgi/content/full/CIRCULATIONAHA.108.767772/DC1.
Clinical trial registration information—URL: http://www.clinicaltrials.gov. Unique identifier: NCT00091637.
Presented in part at the 56th annual Scientific Sessions of the American College of Cardiology, New Orleans, La, March 24–27, 2007, and published in abstract form (JACC. 2007;49[Suppl A]:189A).