Correlation Between Markers of Reperfusion and Mortality in ST-Elevation Myocardial Infarction: A Systematic Review

Abstract: Objective. To correlate early and late mortality with markers of reperfusion in ST-elevation myocardial infarction (STEMI). Background. Early reperfusion improves STEMI outcomes. Reperfusion can be assessed using angiographic (Thrombolysis in Myocardial Infarction [TIMI] flow grade or myocardial blush grade [MBG]) or electrocardiographic markers (ST-segment recovery (STR). Methods. We searched electronic databases for all STEMI randomized clinical studies from the last decade reporting markers of reperfusion and clinical outcome. We used a generalized estimating equation (GEE) model with logistic regression link in order to assess the correlation between each marker of reperfusion and mortality at 30 and 365 days. We also performed random effect meta-analysis for selected studies comparing mortality for specific categories of MBG. Results. We identified 44 studies with 19,955 patients. Final TIMI 3 flow was achieved in 87%, 70% had MBG 2 or 3, and 66% had complete STR. Average 30-day and 1-year mortality was 2.97 ± 2.34% and 4.11 ± 2.52%, respectively. Adjusting (study level) for age, diabetes, chronic kidney disease, infarct location, ejection fraction, and female sex, there was significant correlation between each of the three markers and 1-year mortality (P=.03 for TIMI 3; P=.02 for MBG 2 or 3; and P=.04 for STR). In nearly 6000 patients, there was substantial excess mortality in those with MBG 0/1 compared with MBG 2/3 (relative risk = 2.14 [1.65-2.77] with P<.001 at 30 days; relative risk = 1.49 [1.3-1.7] and P<.001 at 1 year). Conclusion. After correcting for clinical factors known to affect outcome, there was a significant correlation between survival and better reperfusion.

J INVASIVE CARDIOL 2014;26(11):587-595

Key words: STEMI outcomes, meta-analysis, reperfusion

_________________________________________

ST-segment elevation myocardial infarction (STEMI) continues to be a significant public health problem in both industrialized and developing countries. STEMI is associated with significant morbidity and mortality.¹ Rapid and durable restoration of flow in the obstructed infarct artery after the onset of STEMI symptoms is a key determinant of clinical outcomes.^2,3 However, patients with timely and successful restoration of epicardial flow, measured by Thrombolysis in Myocardial Infarction (TIMI) flow grade, may still have poor short-term clinical outcomes. Hence, myocardial blush grade (MBG), Thrombolysis in Myocardial Infarction perfusion grade (TMPG), and ST-segment resolution (STR) have been added to refine our ability to identify optimal reperfusion and predict short-term clinical outcomes.^4-6 Many studies using these markers were limited by small patient populations, and the relationship between these reperfusion markers and longer-term mortality has not been well studied.

Hence, we performed a systematic review to evaluate rates of reperfusion success in randomized clinical trials over the past decade — however achieved — and assessed the relationship between reperfusion markers and mortality in STEMI. We also performed a formal meta-analysis of selected studies, from which comparisons were available according to MBG. 

Methods

Search strategy and study selection. We searched PUBMED, Cochrane, and EMBASE for studies using the following key words: STEMI/acute myocardial infarction; reperfusion; clinical outcome; microvascular dysfunction; ST-segment resolution (STR); Thrombolysis in Myocardial Infarction (TIMI); Thrombolysis in Myocardial Infarction perfusion grade (TMPG); myocardial blush grade (MBG); corrected TIMI frame count (cTFC); percutaneous coronary intervention (PCI); and fibrinolysis.  We limited our search to randomized controlled trials of adult patients published in English. We also searched the references of review articles, editorials, professional societal guidelines, and original studies identified by the electronic search to find other potentially eligible studies. The following criteria were used for study selection: (1) study population consisting of subjects with acute STEMI; (2) information available on angiographic characteristics (any of: MBG, TMPG, STR, or TIMI flow grade) and mortality; and (3) standard definitions were used to define reperfusion variables.^4,6-8 In brief, TIMI flow grade (0-3) assesses the rapidity with which contrast material reaches anatomical landmarks in the infarct artery, in comparison with unaffected arteries. Quantitatively, it translates into TIMI frame counts, corrected (division by 1.73) for the greater length of the left anterior descending artery (cTFC). MBG was assessed using the densitometric method.⁶ MBG 0 denotes no contrast density or persistent blush or staining in the territory supplied by the infarct artery; MBG 1 denotes minimal contrast density; MBG 2 denotes moderate contrast density but less than that obtained during angiography of a contralateral or ipsilateral non-infarct related coronary artery; and MBG 3 denotes normal contrast density (comparable with that obtained during angiography of a contralateral or ipsilateral non-infarct related coronary artery). TMPG uses a dynamic scale: TMPG 0 for no or minimal blush; TMPG 1 for stain present, blush persists on next injection; TMPG 2 for dye strongly persistent at end of washout, but resolved by next injection; and TMPG 3 for normal ground glass appearance of blush, dye mildly persistent at end of washout. Successful myocardial reperfusion was defined as MBG or TMPG grades 2/3. ST-segment elevation was measured to the nearest 0.1 mV 40 msec after the end of the QRS, using the TP segment as baseline. Complete STR was present if ≥50% of the sum of ST elevation present on the baseline electrocardiogram (ECG) had resolved by 60 minutes after PCI.

Data abstraction. Two authors (SS and SB) independently abstracted data from the selected studies. All reported baseline and angiographic characteristics including reperfusion parameters (based on independent adjudication in central laboratory or on investigator interpretation) as well as all-cause mortality were extracted from each of the included studies and tabulated. 

Statistics. Continuous variables are presented as means with standard deviation. Categorical variables are presented as proportions. Each variable summary was weighted for trial size. We used a generalized estimating equation (GEE) model with logistic regression link, which properly accounts for the clustering of data for each trial. That is, each trial arm summary (each row of data) is considered a “cluster” and these clusters are random effects in the model, considered based on their size and the absolute number of events they contributed. Based on the trials with both 30-day and 1-year (yr) data, we built a regression model to predict the probability of death at 1 yr based on 30-day rates. This model was used to impute the 1-yr death rates in the studies missing it. Both models adjusted for age, ejection fraction, sex, infarct location, diabetes, and chronic kidney disease. We performed formal random effect meta-analysis for studies reporting mortality for specific  categories of MBG. Heterogeneity was assessed using a standard chi-square test and the inverse variance method. Significance level was set at <.05. All analyses were performed with STATA SE 9.2 and SAS 9.2.

Results

Our final analysis included 44 studies (reported between 2002 and 2013) with 89 different treatment clusters and 19,955 patients (see Appendix, available at www.invasivecardiology.com). The vast majority of patients had reperfusion with primary PCI. Five of these studies were related to stent implantation, 16 addressed thrombus aspiration or emboli protection devices, and 23 evaluated adjunctive pharmacological interventions, including use of antiplatelet agents and glycoprotein IIb/IIIa inhibitors (Figure 1).

Incidence of successful reperfusion and correlation with outcome. The clinical characteristics, incidence of the various reperfusion markers, and mortality rate for each trial are presented in Tables 1 and 2. The mean age was 61 ± 12 years, 76% of the patients were males, 19% had diabetes mellitus, 7% had chronic kidney disease (defined as estimated creatinine clearance of <60 mL/min), and average left ventricular ejection fraction was 47 ± 10%. The infarct-related artery was the left anterior descending (LAD) in 47%. 

Final TIMI 3 flow in the infarct-related artery was achieved in 87% (based on 77 clusters), 70% had MBG or TMPG 2 or 3 (67 clusters), the corrected TIMI frame count was 21 ± 11 (45 clusters), and 66% had complete STR (46 clusters). The average 30-day mortality was 2.97 ± 2.34% (64 clusters, 374 events) and average 1-year mortality was 4.11 ± 2.52% (35 clusters, 797 events). All studies without 30-day mortality data published longer-term follow-up data.

The correlations between death rate at 30 days and each marker of reperfusion are shown in Figures 2A-2C. Similarly, correlations between mortality at 1 year and markers of reperfusion are shown in Figures 3A-3C. Even after adjusting for study size, there was essentially no univariate correlation between outcome and markers of reperfusion at both time points.

Using the GEE model and adjusting (cluster level) for age, ejection fraction and rates of diabetes, CKD, LAD infarct, and female sex, there was significant correlation between TIMI 3 flow rates and 30-day mortality (P<.001) and between each of the three markers and 1-year mortality (P=.03 for TIMI 3; P=.02 for MBG 2 or 3; and P=.04 for STR) (Table 3). In a sensitivity analysis using 3 categories of STR (<30%, 30%-70%, and >70%), the results were consistent with those we reported above (data not shown). We performed a sensitivity analysis excluding the sole fibrinolysis-only trial (AMI-SK) and the results remained essentially unchanged.

Comparison of outcomes between patients with optimal and suboptimal reperfusion. A formal meta-analysis was feasible for 3 studies (CADILLAC, HORIZONS-AMI, and INFUSE-AMI; 5020 patients), which compared 30-day mortality in patients with MBG 0/1 vs MBG 2/3. Substantial excess mortality existed in those with MBG 0/1 compared with MBG 2/3 (relative risk = 2.14 [1.65-2.77]; P<.001 (P heterogeneity=.77). Similarly, meta-analysis of 4 studies (CADILLAC, HORIZONS-AMI, APEX-AMI, and INFUSE-AMI; 5959 patients), which compared mortality up to 12 months in patients with MBG 0/1 vs MBG 2/3, revealed a substantially higher mortality in those with MBG 0/1 (relative risk = 1.49 [1.3-1.7]; P<.001 (P heterogeneity=.66) (Figures 4 and 5).     

Discussion

To our knowledge, this is the only systematic review of the correlation between markers of reperfusion and mortality in patients receiving reperfusion for STEMI. Three important findings merit attention. First, predominantly mechanical reperfusion achieves high rates of epicardial (87%) and myocardial (70%) reperfusion in a wide range of STEMI patients enrolled in randomized clinical trials. Two-thirds had also achieved substantial ST-segment resolution. Second, there was weak correlation between any of the reperfusion parameters and short- or long-term mortality in the overall STEMI population, even after weighing for study size. Only after adjusting for important determinants of outcome though, there was significant correlation between each of the reperfusion parameters and 1-year survival. Third, the significance of achieving adequate myocardial reperfusion was confirmed in a more limited number of patients (nearly 6000) with direct comparison of outcome according to reperfusion result assessed by MBG. These data may inform the design of randomized clinical trials in which surrogate endpoints, such as TIMI flow, MBG, or STR, may be considered acceptable in view of their above-described correlation with mortality. Such an approach may reduce the size of the cohort studied and enable more frequent iterations and testing of novel therapies. 

In comparison to TIMI flow, both STR and MBG are better predictors of clinical outcomes.^4-6 Sorajja et al compared the predictive value for low MBG and persistent ST-segment elevation (STE) for mortality in a cohort of 456 AMI patients undergoing mechanical reperfusion therapy. Both MBG 0/1 and persistent STE were independent predictors for mortality at a mean follow-up of 12 months. Combining MBG and persistent STE increased the predictive power for mortality compared to each individual parameter.⁹ 

The observation that patients with a patent infarct artery may still experience poor outcome led to better understanding of the importance of restoration of microcirculatory function.¹⁰ Because of ubiquitous distal embolization of plaque and thrombus from the infarct lesion as well as the inherent destruction of capillaries and the ensuing edema, there is frequent dysfunction of the microcirculation in the infarct bed. This can be temporary or permanent. MBG and STR present complementary information on the integrity of the microvasculature and the surrounding myocardium. The former pertains to patency of small arterioles, while the latter depicts the restoration of electrical stability in the myocardial cells. There is concordance in only ~70% of patients between these two parameters, ie, favorable MBG is accompanied by significant STR. In at least one detailed analysis, MBG correlated better with long-term mortality than STR.¹¹ Ultimately, the exact mechanism underpinning the relationship between successful reperfusion and better survival is not entirely clear. Most likely, outcome beyond the first 30 days is influenced by infarct size. In some studies, better MBG is associated with smaller infarcts.¹²

It is notable in our data (Figures 2A-2C and 3A-3C) that because reperfusion was achieved predominantly with primary PCI in randomized clinical trials, the rate of success was quite high and homogenous, limiting the ability to detect large differences between clusters of patients. Therefore, it is not surprising than only by using GEE and correcting for rates of variables known to affect outcome (age, sex, infarct location, diabetes, ejection fraction, and chronic kidney disease) were we able to demonstrate significant correlation between outcome and markers of reperfusion. 

We also performed a meta-analysis of studies with available data on outcome according to final MBG achieved with primary PCI. Suboptimal MBG (grade 0 or 1) showed significant association with higher 30-day and 1-year mortality. The pathophysiological explanation of the association between improved reperfusion and better survival is likely related to myocardial salvage. Preservation of left ventricular function, lower arrhythmia burden, and reduced sudden cardiac death are the plausible reasons explaining this association.^13-20 It is important, though, to underscore that these associations do not demonstrate causality and may simply reflect the baseline characteristics and total ischemic time of the population analyzed. 

Study limitations. We recognize important limitations in our methodology, particularly as it relates to lack of patient-level data. Although we weighted the regression model for study size and adjusted for incidence of factors known to affect outcome, we could not completely overcome the absence of important individual patient data, such as delay to reperfusion. The confirmation of our results obtained from the meta-analysis of mortality according to final MBG in ~30% of the entire cohort is somewhat reassuring. Furthermore, we appreciate the fact that patients enrolled in randomized clinical trials may not accurately represent the universe of patients with STEMI treated in routine clinical practice, particularly as it relates to heterogeneity in time to presentation, previous myocardial infarction, other important comorbidities and concomitant medical therapy. Finally, we acknowledge that assessment of reperfusion using angiographic and electrocardiographic markers is fraught with difficulties related to lack of standardized methodology. 

Conclusion

Despite these important limitations, we conclude that predominantly mechanical reperfusion in a variety of clinical trials achieves high levels of success, as measured by TIMI flow, myocardial blush grade, and ST-segment resolution. The study-level rate of successful reperfusion does not correlate with outcome. After correcting for study-level incidence of clinical factors known to affect outcome, there is a significant correlation between survival and better reperfusion, as assessed by each of these indices. 

References

1. O’Gara PT, Kushner FG, Ascheim DD, et al. 2013 ACCF/AHA guideline for the management of ST-elevation myocardial infarction: executive summary: a report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines. Circulation. 2013;127(4):529-555.
2. Mehta RH, Ou FS, Peterson ED, et al. Clinical significance of post-procedural TIMI flow in patients with cardiogenic shock undergoing primary percutaneous coronary intervention. JACC Cardiovasc Interv. 2009;2(1):56-64.
3. Keeley EC, Boura JA, Grines CL. Primary angioplasty versus intravenous thrombolytic therapy for acute myocardial infarction: a quantitative review of 23 randomised trials. Lancet. 2003;361(9351):13-20.
4. van ‘t Hof AW, Liem A, de Boer MJ, et al. Clinical value of 12-lead electrocardiogram after successful reperfusion therapy for acute myocardial infarction. Zwolle myocardial infarction study group. Lancet. 1997;350(9078):615-619.
5. Gibson CM, Cannon CP, Murphy SA, et al. Relationship of TIMI myocardial perfusion grade to mortality after administration of thrombolytic drugs. Circulation. 2000;101(2):125-130.
6. Henriques JP, Zijlstra F, van‘t Hof AW, et al. Angiographic assessment of reperfusion in acute myocardial infarction by myocardial blush grade. Circulation. 2003;107(16):2115-2119.
7. Gibson CM, Cannon CP, Murphy SA, Marble SJ, Barron HV, Braunwald E; TIMI Study Group. Relationship of the TIMI myocardial perfusion grades, flow grades, frame count, and percutaneous coronary intervention to long-term outcomes after thrombolytic administration in acute myocardial infarction. Circulation. 2002;105(16):1909-1913.
8. TIMI Study Group. The Thrombolysis In Myocardial Infarction (TIMI) trial. Phase I findings. N Engl J Med. 1985;312(14):932-936.
9. Sorajja P, Gersh BJ, Costantini C, et al. Combined prognostic utility of ST-segment recovery and myocardial blush after primary percutaneous coronary intervention in acute myocardial infarction. Eur Heart J. 2005;26(7):667-674.
10. Ito H, Tomooka T, Sakai N, et al. Lack of myocardial perfusion immediately after successful thrombolysis. A predictor of poor recovery of left ventricular function in anterior myocardial infarction. Circulation. 1992;85(5):1699-1705.
11. Brener SJ, Dizon JM, Mehran R, et al. Complementary prognostic utility of myocardial blush grade and st-segment resolution after primary percutaneous coronary intervention: Analysis from the horizons-ami trial. Am Heart J. 2013;166(4):676-683.
12. Brener SJ, Maehara A, Dizon JM, et al. Relationship between myocardial reperfusion, infarct size, and mortality: the INFUSE-AMI (Intracoronary Abciximab and Aspiration Thrombectomy in Patients With Large Anterior Myocardial Infarction) trial. JACC Cardiovasc Interv. 2013;6(7):718-724.
13. Costantini CO, Stone GW, Mehran R, et al. Frequency, correlates, and clinical implications of myocardial perfusion after primary angioplasty and stenting, with and without glycoprotein IIb/IIIa inhibition, in acute myocardial infarction. J Am Coll Cardiol. 2004;44(2):305-312.
14. Lepper W, Sieswerda GT, Vanoverschelde JL, et al. Predictive value of markers of myocardial reperfusion in acute myocardial infarction for follow-up left ventricular function. Am J Cardiol. 2001;88(12):1358-1363.
15. Bellandi F, Leoncini M, Maioli M, et al. Markers of myocardial reperfusion as predictors of left ventricular function recovery in acute myocardial infarction treated with primary angioplasty. Clin Cardiol. 2004;27(12):683-688.
16. Nijveldt R, Beek AM, Hirsch A, et al. Functional recovery after acute myocardial infarction: comparison between angiography, electrocardiography, and cardiovascular magnetic resonance measures of microvascular injury. J Am Coll Cardiol. 2008;52(3):181-189.
17. Gibson CM, Pride YB, Buros JL, et al. Association of impaired thrombolysis in myocardial infarction myocardial perfusion grade with ventricular tachycardia and ventricular fibrillation following fibrinolytic therapy for ST-segment elevation myocardial infarction. J Am Coll Cardiol. 2008;51(5):546-551.
18. Ndrepepa G, Tiroch K, Fusaro M, et al. 5-year prognostic value of  no-reflow phenomenon after percutaneous coronary intervention in patients with acute myocardial infarction. J Am Coll Cardiol. 2010;55(21):2383-2389.
19. Seyfeli E, Abaci A, Kula M, et al. Myocardial blush grade: to evaluate myocardial viability in patients with acute myocardial infarction. Angiology. 2007;58(5):556-560.
20. Angeja BG, Gunda M, Murphy SA, et al. TIMI myocardial perfusion grade and ST segment resolution: association with infarct size as assessed by single photon emission computed tomography imaging. Circulation. 2002;105(3):282-285.

_____________________________________________

From the Division of Cardiology, Department of Medicine, New York Methodist Hospital, Brooklyn, New York.

Disclosure: The authors have completed and returned the ICMJE Form for Disclosure of Potential Conflicts of Interest. The authors report no conflicts of interest regarding the content herein.

Manuscript submitted March 25, 2014, provisional acceptance given April 17, 2014, final version accepted April 30, 2014.

Address for correspondence: Sorin J. Brener MD, FACC, Professor of Medicine - Weill Cornell Medical College, Director, Cardiac Catheterization Laboratory, NY Methodist Hospital, 506 6th street, KP-2, Brooklyn, NY 11215. Email: sjb9005@nyp.org

___________________________________________