Utility of Peak Creatine Kinase-MB Measurements in Predicting Myocardial Infarct Size, Left Ventricular Dysfunction, and Outcome After First Anterior Wall Acute Myocardial Infarction (from the INFUSE-AMI Trial)




Infarct size after ST-segment elevation myocardial infarction (STEMI) is associated with long-term clinical outcomes. However, there is insufficient information correlating creatine kinase-MB (CK-MB) or troponin levels to infarct size and infarct location in first-time occurrence of STEMI. We, therefore, assessed the utility of CK-MB measurements after primary percutaneous coronary intervention of a first anterior STEMI using bivalirudin anticoagulation in patients who were randomized to intralesion abciximab versus no abciximab and to manual thrombus aspiration versus no aspiration. Infarct size (as a percentage of total left ventricular [LV] mass) and LV ejection fraction (LVEF) were evaluated by cardiac magnetic resonance imaging at 30 days and correlated to peak CK-MB. Peak CK-MB (median 240 IU/L; interquartile range 126 to 414) was significantly associated with infarct size and with LVEF (r = 0.67, p <0.001; r = −0.56, p <0.001, respectively). A large infarct size (greater than or equal the median, defined as 17% of total LV mass) and LVEF ≤40% were more common in the highest peak CK-MB tertile group than in the other tertiles (87.6% vs 49.5% vs 9.1%, p <0.001; 43.2% vs 14.0% vs 4.6%, p <0.001, respectively). Peak CK-MB of at least 300 IU/L predicted with moderate accuracy both a large infarct size (area under the curve 0.88) and an LVEF ≤40% (area under the curve 0.78). Furthermore, CK-MB was an independent predictor of 1-year major adverse cardiac events (hazard ratio 1.42 per each additional 100 IU/L [1.20 to 1.67], p <0.001). In conclusion, CK-MB measurement is useful in estimating infarct size and LVEF and in predicting 1-year clinical outcomes after primary percutaneous coronary intervention for first anterior STEMI.


After ST-segment elevation myocardial infarction (STEMI), decreased left ventricular (LV) systolic function is significantly associated with poor clinical outcomes. Creatine kinase-MB (CK-MB) or troponin measurements are often performed in clinical practice, and the peak values have been shown to reflect infarct size and clinical outcome. Cardiac magnetic resonance imaging (cMRI) produces images with high spatial and temporal resolution. Combined with late gadolinium enhancement, cMRI permits accurate and reproducible noninvasive assessment of infarct size and LV function. Although some studies have reported the relation between cardiac biomarkers, such as CK-MB and troponins, and infarct size by cMRI, there is little evidence about patients with first anterior STEMI and large territories at risk but who underwent reperfusion. Therefore, we used data from the Intracoronary Abciximab and Aspiration Thrombectomy in Patients With Large Anterior Myocardial Infarction (INFUSE-AMI) trial to determine the utility of CK-MB measurements in predicting infarct size and clinical outcomes.


Methods


The design and primary results of the INFUSE-AMI trial have been previously described in detail. Briefly, INFUSE-AMI was an open-label, multicenter, 2 × 2 factorial, single-blind evaluation of (1) bolus intralesion abciximab delivered through the ClearWay RX catheter (Atrium Medical, Hudson, New Hampshire) versus no abciximab and (2) manual thrombus aspiration using the Export catheter (Medtronic, Santa Rosa, California) versus no thrombus aspiration. In total, 452 patients with anterior MI (infarct lesion in the proximal or mid-left anterior descending artery [LAD]) with visually assessed Thrombolysis In Myocardial Infarction (TIMI) 0 to 2 flow were enrolled. Patients underwent primary percutaneous coronary intervention (PCI) with bivalirudin anticoagulation. The primary end point was infarct size by cMRI as percent of LV mass at 30 days. Markers of reperfusion included ST-segment resolution, TIMI flow, myocardial blush grade, and corrected TIMI frame counts. All electrocardiographic, angiographic, ST-segment, and cMRI data were assessed at independent, blinded core laboratories at the Cardiovascular Research Foundation (New York, New York) using standard methodology and definitions.


In the present study, we included only patients with available CK-MB values. Measurements of CK-MB levels were required by protocol using an immune-inhibition method before the index procedure and every 8 hours after PCI until the peak has been reached. Among the reported CK-MB values from each site after the protocol, we chose the maximum CK-MB value to indicate peak CK-MB.


Cardiac MRI acquisition and analysis were previously described in detail. In brief, all images were acquired using a commercially available 1.5 T MRI scanner. cMRI included cine-cMRI for LV volumes and systolic function and delayed enhancement cMRI for evaluation of infarct size. Delayed enhancement cMRI was performed using a 2-dimensional breath-held segmented inversion-recovery gradient-echo sequence 10 minutes after 0.2 mmol/kg gadolinium contrast injection. All images were analyzed using Food and Drug Administration–approved ReportCARD software with investigational multivendor conversion software (General Electric, Milwaukee, Wisconsin). LV mass (the area between epicardial and endocardial borders) and infarct area (the white area within the black myocardium) were analyzed on the delayed enhancement cMR images. Infarct size was calculated as percent of LV myocardial infarct mass. LV dysfunction was defined as LV ejection fraction (LVEF) <40%.


All patients were monitored during hospitalization, and clinical follow-up was scheduled at 30 days and at 1 year. Major adverse cardiac events (MACE) were defined as the composite of death, reinfarction, new-onset heart failure, or rehospitalization for heart failure. All clinical events were adjudicated by an independent clinical events committee.


Outcomes were examined according to tertiles of peak CK-MB levels. Continuous variables are presented as median with interquartile range (IQR). Baseline characteristics according to tertiles of peak CK-MB were compared using Kruskal-Wallis test, analysis of variance, or chi-square test where appropriate. Correlation coefficients were based on a nonparametric method (Spearman rank). Multiple linear regression analysis was used to determine independent predictors of LVEF at 30 days. The following variables were entered into the regression model: peak CK-MB, age, hypertension, diabetes, proximal LAD, Killip class III on presentation, and final blush grade 0 or 1. Receiver-operator characteristic (ROC) curves were generated for the relation between peak CK-MB and infarct size and LVEF at 30 days. Kaplan-Meier time-to-event estimates for clinical outcomes were compared with the log-rank test. Clinical variables with a p value <0.2 by univariate analysis were included in the multivariable Cox proportional hazards model. A p value <0.05 was considered significant. All data were analyzed using SAS, version 9.2 (SAS Institute Inc., Cary, North Carolina) and JMP, version 9.0 (SAS Institute Inc.).




Results


Of the 452 anterior patients with STEMI enrolled in the INFUSE-AMI trial, peak CK-MB was available in 311 patients (69%). In the other 141 patients, only troponin T, troponin I, or CK were measured. Baseline characteristics and procedural results were similar in patients with and without available peak CK-MB data except for age, hyperlipidemia, baseline TIMI flow, and medication at discharge when compared, without any significant difference in 30-day or 1-year outcomes ( Supplementary Tables 1 to 4 ). The median (IQR) time from symptom onset to peak CK-MB and CK was 10.3 hours (8.0, 12.5). The median (IIQR) peak CK-MB and CK values were 240 IU/L (126, 414) and 2,367 IU/L (1027, 4047), respectively. Clinical and procedural characteristics of patients stratified by tertiles of peak CK-MB are listed in Tables 1 and 2 . Patients in the highest tertile had significantly higher incidence of diabetes mellitus, proximal LAD infarcts, and TIMI 0/1 at baseline and higher white blood cell count on admission.



Table 1

Clinical characteristics of the overall study population and of patients stratified by tertiles of peak creatine kinase-MB





















































































































































































Variable Overall
(N=311)
Creatine kinase-MB (IU/L) p Value
Tertile 1
≤163
(n=103)
Tertile 2
>163 to 343
(n=104)
Tertile 3
≥343
(n=104)
Peak CK-MB (IU/L) 240 [126, 414] 71 [39, 126] 240 [202, 292] 500 [412, 602] <0.001
Time from symptom onset to peak CK-MB (hours) 10.3 [8.0, 12.5] 10.9 [7.8, 16.0] 10.1 [7.9, 12.2] 10.1 [8.0, 11.7] 0.08
Peak CK (U/L) 2367 [1027, 4047] 705 [322, 1111] 2518 [1812, 3500] 4693 [3140, 5939] <0.001
Age (years) 60 [51, 69] 60 [51, 70] 59 [51, 68] 60 [52, 69] 0.93
Men 238 (76.5%) 79 (76.7%) 82 (78.8%) 77 (74.0%) 0.71
Body mass index (kg/m 2 ) 26.9 [24.2, 29.4] 26.6 [24.2, 29.7] 26.3 [24.1, 28.8] 27.4 [24.3, 29.6] 0.35
Hypertension 102 (32.8%) 28 (27.2%) 34 (32.7%) 40 (38.5%) 0.22
Hyperlipidemia 39 (12.6%) 13 (12.6%) 17 (16.5%) 9 (8.7%) 0.23
Diabetes mellitus 32 (10.3%) 18 (17.6%) 5 (4.8%) 9 (8.7%) 0.01
Current/recent smoker 149 (48.4%) 47 (46.1%) 47 (46.1%) 55 (52.9%) 0.53
Angina prior to myocardial infarction 28 (9.1%) 9 (8.7%) 11 (10.8%) 8 (7.8%) 0.75
Medication pre-hospitalization
Beta blockers 27 (8.7%) 7 (6.8%) 11 (10.6%) 9 (8.7%) 0.63
ACE inhibitor or ARB 51 (16.4%) 17 (16.5%) 18 (17.3%) 16 (15.4%) 0.93
Statins 23 (7.4%) 9 (8.7%) 9 (8.7%) 5 (4.8%) 0.47
Symptom onset to first device (minutes) 158 [121, 221] 160 [114, 228] 149 [120, 205] 174 [132, 243] 0.12
Killip class III 3 (1.0%) 0 (0.0%) 0 (0.0%) 3 (2.9%) 0.05
CrCl <60 mL/min 41 (14.2%) 16 (17.0%) 11 (11.5%) 14 (14.3%) 0.55
WBC Count (×10 3 /μL) 11.5 [9.1, 14.4] 9.6 [7.5, 12.3] 11.8 [9.3, 14.4] 12.8 [10.6, 16.0] <0.001
Relative ST segment resolution (%) 72.8 [44.5, 89.3] 81.6 [43.9,95.6] 69.6 [42.0,82.7] 72.3 [54.8,87.2] 0.42
Medication at discharge
Beta blockers 303 (98.4%) 103 (100.0%) 100 (96.2%) 100 (99.0%) 0.08
ACE inhibitor or ARB 285 (92.5%) 93 (90.3%) 98 (94.2%) 94 (93.1%) 0.54
Statins 301 (97.7%) 101 (98.1%) 102 (98.1%) 98 (97.0%) 0.85

Values are median [interquartile range] unless specified.

ACE = angiotensin-converting enzyme; ARB = angiotensin II receptor blocker; CrCl = creatinine clearance; WBC = white blood cell.

Based on sum of positive ST elevation leads.



Table 2

Procedural and angiographic findings in overall study population and of patients stratified by peak creatine kinase-MB





























































































































Variable Overall Creatine kinase-MB (IU/L) p Value
Tertile 1
≤163
Tertile 2
>163 to 343
Tertile 3
≥343
Aspiration performed 161 (51.8%) 52 (50.5%) 50 (48.1%) 59 (56.7%) 0.44
Abciximab administered, intracoronary 156 (50.2%) 56 (54.4%) 54 (51.9%) 46 (44.2%) 0.31
Location of LAD narrowing <0.001
Proximal 207 (66.6%) 55 (53.4%) 70 (67.3%) 82 (78.8%)
Mid 104 (33.4%) 48 (46.6%) 34 (32.7%) 22 (21.2%)
Three vessel coronary disease 40 (12.9%) 13 (12.6%) 15 (14.4%) 12 (11.5%) 0.82
Baseline coronary thrombus 263 (83.8%) 76 (73.1%) 89 (85.6%) 98 (92.5%) <0.001
Baseline TIMI flow
TIMI 0 or 1 214 (68.8%) 52 (50.5%) 75 (72.1%) 87 (83.7%) <0.001
TIMI 2 49 (15.8%) 19 (18.4%) 15 (14.4%) 15 (14.4%) 0.66
TIMI 3 48 (15.4%) 32 (31.1%) 14 (13.5%) 2 (1.9%) <0.001
Final TIMI flow
TIMI 0 or 1 6 (1.9%) 4 (3.9%) 1 (1.0%) 1 (1.0%) 0.21
TIMI 2 19 (6.1%) 5 (4.9%) 6 (5.8%) 8 (7.7%) 0.68
TIMI 3 286 (92.0%) 94 (91.3%) 97 (93.3%) 95 (91.3%) 0.83
Final blush grade 0 or 1 64 (20.6%) 21 (20.4%) 18 (17.3%) 25 (24.0%) 0.49

Values are n (%).

LAD = left anterior descending artery; TIMI = Thrombolysis In Myocardial Infarction.


Among the 311 patients with available CK-MB, cMRI at 2 to 7 days was analyzable in 121 patients; and cMRI at 30 days was analyzable in 258 patients ( Table 3 ). Overall, the median (IQR) cMRI-determined infarct size was 22.1% (12.9, 32.0) at 2 to 7 days and 17.4% (9.0, 24.7) at 30 days. At 30 days after STEMI, patients with higher peak CK-MB values had a larger infarct size and a lower LVEF than patients with lower peak CK-MB values. The frequency of large infarcts (≥17%—the median for the entire study) and of LVEF ≤40% was significantly higher in the highest tertile than in the other groups (87.6% vs 49.5% vs 9.1%, p <0.001; 43.2% vs 14.0% vs 4.6%, p <0.001, respectively; Figure 1 ).



Table 3

Cardiac magnetic resonance imaging results in overall study population and in patients stratified by peak creatine kinase-MB





















































































































































































Variable Overall Creatine kinase-MB (IU/L) p Value
Tertile 1
≤163
Tertile 2
>163 to 343
Tertile 3
≥343
2–7-Day cMRI N=121 n=32 n=42 n=47
Total LV myocardial mass (g) 148.1 [123.1, 183.5] 138.9 [119.6, 156.9] 145.6 [118.8, 174.5] 164.3 [139.8, 199.5] <0.01
Infarct mass (g) 32.3 [18.2, 50.5] 8.2 [0.9, 22.7] 28.7 [21.5, 40.9] 52.3 [37.7, 66.6] <0.001
Infarct size (% of total LV mass) 22.1 [12.9, 32.0] 5.7 [0.6, 16.5] 21.7 [15.3, 27.6] 32.3 [25.7, 39.2] <0.001
LV ejection fraction (%) 47.6 [40.5, 52.5] 52.1 [47.7, 60.8] 48.4 [42.6, 52.4] 41.5 [35.5, 47.6] <0.001
LV stroke volume index (mL/m 2 ) 41.7 [33.8, 47.0] 44.0 [38.2, 50.0] 40.3 [34.4, 47.5] 39.0 [32.5, 45.0] 0.09
LV end diastolic volume index (mL/m 2 ) 88.1 [77.8, 100.5] 84.2 [72.8, 92.8] 87.4 [72.1, 105.3] 95.7 [85.6, 105.6] <0.01
LV end systolic volume index (mL/m 2 ) 47.1 [37.5, 56.6] 37.1 [31.9, 47.4] 44.2 [36.1, 57.7] 53.1 [46.8, 64.6] <0.001
30-Day cMRI N=258 n=85 n=84 n=89
Total LV myocardial mass (g) 131.3 [110.4, 155.0] 126.3 [106.6, 150.5] 128.4 [107.4, 148.6] 141.2 [119.1,1 60.4] <0.01
Infarct mass (g) 21.3 [10.8, 34.2] 6.5 [1.0, 16.1] 21.4 [14.6, 30.3] 35.3 [27.6, 42.0] <0.001
Infarct size (% of total LV mass) 17.4 [9.0, 24.7] 5.0 [1.0, 10.6] 17.6 [12.8, 22.3] 25.0 [20.4, 29.5] <0.001
LV ejection fraction (%) 48.9 [42.1, 55.7] 55.5 [50.5, 60.3] 50.1 [43.5, 56.1] 42.4 [35.1, 46.4] <0.001
LV stroke volume index (mL/m 2 ) 42.3 [37.6, 48.3] 43.5 [38.1, 48.3] 41.8 [38.4, 48.7] 41.9 [35.6, 48.1] 0.38
LV end diastolic volume index (mL/m 2 ) 88.3 [76.0, 104.9] 79.7 [69.2, 90.8] 86.1 [75.5, 103.1] 101.8 [89.5, 117.7] <0.001
LV end systolic volume index (mL/m 2 ) 44.9 [33.3, 59.4] 34.6 [30.1, 42.1] 42.6 [32.8, 57.2] 59.1 [49.3, 75.3] <0.001
Change from 2–7-day to 30-day N=121 n=32 n=42 n=47
Total LV myocardial mass (g) -12.7 [-21.4, -4.4] -9.4 [-19.9, -1.8] -14.2 [-21.0, -8.1] -11.6 [-23.3, -3.3] 0.59
Infarct mass (g) -8.3 [-17.4, -0.5] -1.2 [-5.6, 0.0] -8.9 [-15.3, -0.2] -13.0 [-25.0, -7.3] <0.001
Infarct size (% of total LV mass) -3.9 [-8.4, 0.0] -0.4 [-3.4, 0.0] -4.3 [-7.3, 0.5] -6.6 [-11.4, -3.4] <0.001
LV ejection fraction (%) 2.1 [-2.1, 5.8] 2.1 [-0.9, 6.3] 3.9 [-0.9, 7.9] 0.3 [-3.5, 4.3] 0.05
LV stroke volume index (mL/m 2 ) 2.3 [-1.5, 7.8] 1.5 [-2.6, 7.0] 2.8 [-0.8, 5.6] 3.1 [-0.4, 8.5] 0.47
LV end diastolic volume index (mL/m 2 ) 2.9 [-4.5, 10.9] -1.2 [-6.5, 6.3] -2.2 [-7.8, 6.7] 10.6 [2.6, 17.3] <0.001
LV end systolic volume index (mL/m 2 ) -0.5 [-5.9, 6.7] -1.9 [-6.2, 1.4] -4.8 [-8.5, 3.9] 4.5 [-3.3, 11.0] <0.001

Values are median [interquartile range] or n (%).

CK-MB = creatine kinase-MB; cMRI = cardiac magnetic resonance imaging; LV = left ventricular.



Figure 1


Frequency of large infarct size and LV dysfunction in patients stratified by peak CK-MB tertiles. Frequency of large infarcts (≥17%—the median for the entire study) and of LVEF ≤40% was significantly higher in the highest tertile than in the other groups.


Correlations between peak CK-MB and 30-day infarct size (r = 0.67, p <0.001; Figure 2 ) and LVEF (r = −0.56, p <0.001; Figure 2 ) were robust and statistically significant. For every 100 IU/L increase in peak CK-MB, there was a 2.7% decrease in LVEF. The correlation between infarct size and LVEF was also strong and statistically significant (r = −0.71, p <0.001; Figure 2 ), such that for every 10% increase in infarct size, there was a 7.4% decrease in LVEF. Correlations between peak CK and 30-day infarct size (r = 0.52, p <0.001) and LVEF (r = −0.43, p <0.001) were also statistically significant but more weakly than peak CK-MB.




Figure 2


Scatter plot of the relation between peak CK-MB and cMRI findings. (A) Peak CK-MB and infarct size; (B) peak CK-MB and LV ejection fraction; (C) infarct size and LV ejection fraction.


Multiple linear regression demonstrated that independent predictors of LVEF were peak CK-MB (per 100 IU/L, coefficient −2.62, 95% confidence interval [CI] −3.10 to −2.14, p <0.0001), final blush grade 0 or 1 (coefficient −3.34, 95% CI −5.90 to −0.78, p = 0.01), and age (per 10 years, coefficient −0.92, 95% CI −1.83 to −0.01, p = 0.048, Table 4 ).



Table 4

Linear regression analysis for left ventricular ejection fraction at 30 days





























































Variable Univariable Analysis Multivariable Analysis
Unstandardized Coefficient p Value Unstandardized Coefficient p Value
Peak CK-MB (per 100 IU/L) -2.66 <0.001 -2.62 <0.0001
Final blush grade 0 or 1 -4.21 0.007 -3.34 0.01
Age (per 10 years) -0.81 0.16 -0.92 0.048
Proximal LAD location -3.21 0.02 Not selected
Hypertension -2.36 0.09 Not selected
Diabetes mellitus 3.25 0.14 Not selected
Killip class III -10.38 0.16 Not selected
Initial TIMI flow grade 0 or 1 -7.60 <0.0001 Not selected

Only gold members can continue reading. Log In or Register to continue

Stay updated, free articles. Join our Telegram channel

Nov 30, 2016 | Posted by in CARDIOLOGY | Comments Off on Utility of Peak Creatine Kinase-MB Measurements in Predicting Myocardial Infarct Size, Left Ventricular Dysfunction, and Outcome After First Anterior Wall Acute Myocardial Infarction (from the INFUSE-AMI Trial)

Full access? Get Clinical Tree

Get Clinical Tree app for offline access