Abstract
Background: Non-contrast T1 hypointense infarct cores (ICs) within infarcted myocardium detected using cardiac magnetic resonance imaging (CMR) T1 mapping may help assess the severity of left ventricular (LV) injury. However, because the relationship of ICs with chronic LV reverse remodeling (LVRR) is unknown, this study aimed to clarify it.
Methods and Results: We enrolled patients with reperfused AMI who underwent baseline CMR on day-7 post-primary percutaneous coronary intervention (n=109) and 12-month follow-up CMR (n=94). Correlations between ICs and chronic LVRR (end-systolic volume decrease ≥15% at 12-month follow-up from baseline CMR) were investigated. We detected 52 (47.7%) ICs on baseline CMR by non-contrast-T1 mapping. LVRR was found in 52.1% of patients with reperfused AMI at 12-month follow-up. Patients with ICs demonstrated higher peak creatine kinase levels, higher B-type natriuretic peptide levels at discharge, lower LV ejection fraction at discharge, and lower incidence of LVRR than those without ICs (26.5% vs. 73.3%, P<0.001) at follow-up. Multivariate logistic regression analysis showed that the presence of ICs was an independent and the strongest negative predictor for LVRR at 12-month follow-up (hazard ratio: 0.087, 95% confidence interval: 0.017–0.459, P=0.004). Peak creatine kinase levels, native T1 values at myocardial edema, and myocardial salvaged indices also correlated with ICs.
Conclusions: ICs detected by non-contrast-T1 mapping with 3.0-T CMR were an independent negative predictor of LVRR in patients with reperfused AMI.
Left ventricular (LV) remodeling following acute myocardial infarction (AMI) affects clinical outcomes. LV remodeling consists of reverse remodeling (LVRR), defined as a decreased end-systolic volume (ESV), and adverse remodeling, defined as an increased end-diastolic volume (EDV).1 Early reperfusion via percutaneous coronary intervention (PCI) and subsequent administration of β-blockers and renin-angiotensin system blockers improves the prognosis of patients with AMI complicated with heart failure and adverse LV remodeling.2–7 Conversely, LVRR is associated with a favorable prognosis.8 Therefore, early prediction of chronic cardiac remodeling after coronary reperfusion in patients with AMI is essential for clinical decision-making and may have benefit for the development of cardioprotective drugs, PCI procedures, and device therapy.
Editorial p 1980
Contrast cardiac magnetic resonance imaging (CMR) is a validated tool for an accurate and detailed evaluation of the severity of LV damage and the prognosis of patients with AMI. Patients with moderate to severe renal failure with contrast contraindications have a poor prognosis for cardiac events, and have been excluded from previous contrast CMR studies.9 A recently developed T1-mapping CMR technique permits quantitative evaluation of myocardial tissue characteristics without using contrast agents. A previous animal study validated with histology showed that native T1 mapping, when compared with enhanced CMR, is a feasible alternative for detecting the extent and location of infarcted myocardium.10 Moreover, the detection of hypointense infarct cores (ICs) on T1 imaging is a potential alternative for the evaluation of microvascular obstruction (MVO) and intramural hematoma. In a 1.5-T CMR cohort study, ICs were observed in 56% of patients, and low T1 values within the ICs were independently associated with adverse LV remodeling and heart failure occurrence.11
Only a few clinical cohort studies have been conducted with non-contrast T1 mapping technique and 3.0-T CMR. Additionally, the relationship between LVRR and T1 hypointense ICs on native T1 mapping has been rarely reported.12–18 The primary aim of our study was to use 3.0-T CMR to evaluate the relationship of ICs to LV remodeling during the 1-year follow-up of patients with reperfused AMI. The secondary aims were to identify factors correlating with ICs (e.g., patient background, PCI procedure, and other CMR findings) and to clarify which marker is most predictive for chronic LVRR.
Methods
Study Population
This single-center, prospective, observational cohort study was conducted at the Department of Cardiovascular Medicine of Nara Prefecture Seiwa Medical Center. A total of 189 patients with AMI who underwent emergency PCI from September 2015 to February 2019 were screened for the inclusion and exclusion criteria. The inclusion criteria were: (1) diagnosed with AMI, defined as ischemic symptoms lasting ≥30 min with ST-segment elevation or depression (≥1 mm), elevated cardiac troponin I levels (≥0.1 ng/mL), and an onset-to-admission time <24 h; (2) underwent emergency PCI; and (3) underwent CMR within 14 days of AMI onset. The exclusion criteria were: (1) unable or unwilling to consent or commit to follow-up requirements; (2) standard contraindications to contrast agents or magnetic resonance imaging; and (3) history of MI. Finally, 109 patients were included in the baseline cohort, subdivided into those with and without ICs, and their background and CMR findings were compared. Of them, 14 patients did not undergo the 12-month follow-up CMR, and 1 patient had rapid progression of aortic valve stenosis; thus, 94 patients were eventually evaluated for chronic LV remodeling (Figure 1). The PCI procedure, such as aspiration thrombectomy, direct stenting, and pre- or post-dilatations, was chosen at the physician’s discretion. The definitions of the clinical data are summarized in Supplementary Methods 1.
This study was approved by the Ethics Committee of Nara Prefecture Seiwa Medical Center (approval no. 58) and conducted following the standards outlined in the Declaration of Helsinki. Before commencing the baseline CMR study, all participants provided written informed consent according to the study protocol.
Image Acquisition
Baseline CMR was performed using a 3.0-T scanner (MAGNETOM Skyra, Siemens, Erlangen, Germany) with an 18-element phased-array cardiac surface coil around 6 days after the primary PCI. The CMR protocol, details of dark blood-T2 weighted short tau inversion recovery (STIR), native T1 mapping, extracellular volume (ECV) mapping, cine imaging, and late gadolinium enhancement (LGE) images acquired for the analysis are provided in the Supplementary Methods 2 and Figure 2.
Image Analysis
All images were analyzed using CVI42 software (Circle Cardiovascular Imaging, Calgary, Canada). The decision on the presence or absence of ICs was made by agreement of 2 cardiologists (Y.H. and S.O. with over 6 and 15 years of CMR experience, respectively) who were blinded to the patients’ data. The reproducibility of the result for the presence of ICs was investigated for 30 randomly selected subjects from among the present cases. To calculate the intra-assay variation, a cardiologist (Y.H.) evaluated them again following a period of time from the initial analysis, and to calculate the inter-assay variation another cardiologist (A.S.) with 10 years’ experience with CMR evaluated the presence and absence of ICs independently. The intra- and interobserver variabilities for the detection of ICs were excellent; Kappa value: 0.932, P<0.0001; McNemar-Bowker analysis: P=0.317, and Kappa value: 0.798, P<0.0001; McNemar-Bowker analysis: P=0.564, respectively.
The LV EDV, ESV, ejection fraction (EF), and mass were measured using cine images in the short-axis view. Adverse LV remodeling was defined as an increase in the EDV index ≥20% at follow-up from baseline. LVRR was defined as a decrease in the ESV index ≥15% at follow-up from baseline.19
T1 mapping was performed with breath-hold acquisitions and motion correction. The endocardial and epicardial borders were drawn manually. The myocardial area at risk (AAR) was defined as the region with T1 values >2 standard deviations (SDs) of the normal myocardium, that is, 1,186±24 ms in the mid-LV septal wall. This value was determined using data from 28 healthy volunteers without coronary risk factors, as per international consensus-based recommendations (19 men; median age, 47 [30–75] years). The area of salvaged myocardium was calculated by subtracting the area of infarcted myocardium from the AAR. The infarcted myocardium was defined as the region of contrast-enhanced myocardium with a signal intensity >5 SDs of that of the remote myocardium with good contractility on LGE images acquired 10–15 min after gadolinium injection. The infarcted myocardial area on native T1 mapping was matched using LGE, and its contours were subsequently copied and pasted to the native T1 mapping image while matching the slice positions. MVO was defined as the presence of a hypointense core within the infarcted myocardium on LGE images. Therefore, the MVO area was included in the infarcted myocardium. The calculated area and weight of each myocardial tissue were adjusted by body surface area and expressed in each index.
To identify ICs on native T1 mapping, we referred to the same slices of the T2-weighted STIR and LGE images, matching the lesion morphologies. The hypointense area within salvaged myocardium was excluded from ICs because it potentially represented residual viable myocardium. Infarcted myocardium was distinguished from ICs. To confirm the T1 value for each region of the myocardium, the region of interest was delineated meticulously with adequate margins to separate it from the adjacent myocardium, endocardium, and LV blood pool (Supplementary Figure A). The T1 values for each region of the myocardium with and without ICs are illustrated in Supplementary Figure B,C. Previous pathological studies have reported a correlation between high T1 values and intracellular or extracellular molecular alterations.20–22 Our study showed that remote, edematous, and infarcted myocardium with and without ICs could be distinguished using native T1 values (median: 1,197 vs. 1,336 vs. 1,490 ms, respectively, in patients with ICs and median: 1,156 vs. 1,321 vs. 1,495 ms, respectively, in those without ICs). In contrast, we found no difference in T1 values between ICs and remote myocardium with ICs.
Statistical Analysis
Continuous variables are expressed as mean±SD and median (25th–75th percentile) for normally and non-normally distributed data, respectively. Categorical data are presented as absolute numbers and percentages. The Wilcoxon rank-sum and chi-square tests were used to compare variables between patients with and without ICs, and patients with and without LVRR. The statistical analysis of T1 values between the remote myocardium of patients with AMI with and without ICs and normal myocardium of healthy volunteers was adjusted for age. Univariate and multivariate logistic regression analyses were performed for covariables with P<0.05 in the comparative analyses, to identify factors associated with the occurrence of ICs and LVRR. The forward stepwise selection method was adopted for multivariate logistic regression analysis using appropriate candidate factors (that were previously reported) with P<0.05 in the univariate logistic regression analysis.12–18 Receiver-operating characteristic (ROC) curves were drawn to determine the peak threshold creatine kinase (CK) value predictive of ICs. The intra- and interobserver variabilities for the detection of IC were assessed by Kappa Coefficient and McNemar-Bowker test.
All statistical analyses were performed using JMP®
version 13.0.0 (SAS Institute Inc., Cary, NC, USA). P<0.05 was considered statistically significant.
Results
Clinical Characteristics and PCI Procedures of Patients With and Without ICs
The clinical characteristics, PCI findings, and CMR findings at baseline of the 109 patients with reperfused AMI and at follow-up of the 94 patients are shown in Table 1 and the Supplementary Table. Baseline CMR findings revealed ICs in 52 (47.7%) patients with reperfused AMI. Of these, LGE images revealed MVO in 51 (98.1%) patients. Patients with ICs had a lower incidence of non-ST-segment elevation MI (non-STEMI) and fewer episodes of ischemic preconditioning (P=0.033 and P=0.006, respectively), as well as significantly higher peak CK values, white blood cell counts, and B-type natriuretic peptide (BNP) levels at discharge (P<0.001, P=0.006, and P<0.001, respectively), compared with those without ICs. Patients with ICs had a lower incidence of non-left main trunk (LMT)/left anterior descending artery (LAD) lesions in the culprit artery (P=0.035), initial TIMI flow grade ≥2 (P=0.002), final TIMI flow grade 3 (P=0.004), poor blush grade score (P=0.026), Forrester class 1 (P=0.008), and Killip class 1 (P=0.006), compared with those without ICs. Thrombectomy was more commonly performed in patients with ICs (P=0.002). There was no remarkable difference in the backgrounds of the patients at baseline and at follow up.
Table 1. Clinical Characteristics, PCI and CMR Findings of 109 Patients With Reperfused AMI With and Without ICs
| |
All patients
(n=109) |
With ICs
(n=52) |
Without ICs
(n=57) |
P value |
| Male |
85 (78.0) |
44 (84.6) |
41 (71.9) |
0.164 |
| Age, years |
65.1±12.2 |
63.8±1.7 |
66.3±1.6 |
0.196 |
| Body mass index, kg/m2 |
24.4±3.8 |
24.77±0.52 |
24.12±0.50 |
0.245 |
| Hypertension |
64 (58.7) |
28 (53.8) |
36 (63.2) |
0.338 |
| Diabetes mellitus |
28 (25.7) |
15 (28.8) |
13 (22.8) |
0.516 |
| Dyslipidemia |
79 (72.5) |
39 (75.0) |
40 (70.2) |
0.669 |
| Current smoker |
45 (41.3) |
19 (36.5) |
26 (45.6) |
0.436 |
| ST-elevation MI |
100 (91.7) |
51 (98.1) |
49 (86.0) |
0.033 |
| Episodes of ischemic preconditioning |
45 (41.3) |
14 (26.9) |
31 (54.4) |
0.006 |
| Statin therapy before admission |
11 (10.3) |
3 (5.9) |
8 (14.3) |
0.208 |
| Antiplatelet loading before PCI |
99 (90.8) |
47 (90.4) |
52 (91.2) |
0.879 |
| Laboratory tests |
| Hct, % |
36.5±4.9 |
36.3±0.68 |
36.7±0.65 |
0.730 |
| Serum creatinine on admission, mg/dL |
0.86±0.19 |
0.87±0.03 |
0.85±0.03 |
0.707 |
| LDL-C on admission, mg/dL |
132.2±40.6 |
134.8±40.8 |
129.8±40.6 |
0.685 |
| L/H ratio |
3.0±1.2 |
3.02±0.17 |
2.94±0.17 |
0.847 |
| WBC 48 h after PCI, /μL |
10,925.7±2,913.1 |
11,698.1±3,126.3 |
10,221.1±2,531.1 |
0.006 |
| Monocytes, % |
7.1 (5.7, 8.2) |
6.8 (5.5, 8.2) |
7.2 (5.9, 8.1) |
0.430 |
| CRP, mg/dL |
5.6 (3.3, 8.7) |
6.4 (3.5, 9.4) |
4.8 (2.8, 8.1) |
0.162 |
| Peak CK, U/L |
2,247
(1,157, 4,135) |
4,134.5
(2,575.5, 5,522.5) |
1,232
(695, 1,827) |
<0.001 |
| BNP on admission, pg/mL |
29.7 (10.2, 79.2) |
25.5 (10.1, 69.4) |
33.9 (10.4, 83.0) |
0.872 |
| BNP at discharge, pg/mL |
61.7 (37.3, 151.4) |
108.9 (52.6, 200.4) |
54 (30.5, 94.4) |
<0.001 |
| ECG ST resolution after primary PCI |
| Complete resolution (>70%) |
61 (61.0) |
24 (47.1) |
37 (75.5) |
0.004 |
| Medication at discharge |
| Calperitide after primary PCI |
70 (64.2) |
37 (71.2) |
33 (57.9) |
0.166 |
| Diuretics at discharge |
19 (17.4) |
16 (30.8) |
3 (5.3) |
<0.001 |
| ACEI/ARB at discharge |
107 (98.2) |
52 (100.0) |
55 (96.5) |
0.496 |
| Statin |
103 (94.5) |
47 (90.4) |
56 (98.3) |
0.101 |
| SGLT2 inhibitor at discharge |
4 (3.7) |
4 (7.7) |
0 (0) |
0.046 |
| β-blockers at discharge |
89 (81.7) |
47 (90.4) |
42 (73.7) |
0.028 |
| Carvedilol equivalent at discharge, mg |
5.27±3.63 |
5.66±3.69 |
4.82±3.55 |
0.207 |
| Coronary angiography |
| Culprit artery |
|
|
|
0.117 |
| LMT |
2 (1.8) |
1 (1.9) |
1 (1.8) |
|
| LAD |
53 (48.6) |
31 (59.6) |
22 (38.6) |
|
| LCX |
9 (8.3) |
2 (3.8) |
7 (12.3) |
|
| RCA |
45 (41.3) |
18 (34.6) |
27 (47.4) |
|
| No. of diseased arteries |
1.73±0.80 |
1.83±0.83 |
1.65±0.77 |
0.264 |
| LMT/LAD in culprit artery |
55 (50.5) |
32 (61.5) |
23 (40.4) |
0.035 |
| Initial TIMI flow grade ≥2 |
19 (17.4) |
3 (5.8) |
16 (28.1) |
0.002 |
| Initial Rentrop grade 0 |
69 (63.3) |
32 (61.5) |
37 (64.9) |
0.843 |
| Final TIMI flow grade ≤2 |
22 (20.2) |
17 (32.7) |
5 (8.8) |
0.004 |
| Blush grade ≤1 |
8 (7.3) |
7 (13.5) |
1 (1.8) |
0.026 |
| Forrester classification |
|
|
|
0.008 |
| I |
86 (78.9) |
34 (65.4) |
52 (91.2) |
|
| II |
19 (17.4) |
14 (26.9) |
5 (8.8) |
|
| III |
1 (0.9) |
1 (1.9) |
0 |
|
| IV |
3 (2.8) |
3 (5.8) |
0 |
|
| Killip class 2–4 |
10 (9.2) |
9 (17.3) |
1 (1.8) |
0.006 |
| PCI procedure |
| Onset to balloon time, min |
197.5 (137.3, 311.8) |
177.5 (129.8, 267.8) |
215.5 (143.8, 351.8) |
0.168 |
| Thrombectomy |
81 (74.3) |
46 (88.5) |
35 (61.4) |
0.002 |
| Distal protection |
9 (8.3) |
6 (11.5) |
3 (5.3) |
0.305 |
| Post-balloon dilatation |
57 (52.3) |
25 (48.1) |
32 (56.1) |
0.446 |
| Post-balloon dilatation, mm |
3.14±0.53 |
3.11±0.56 |
3.17±0.51 |
0.626 |
| Stentless PCI |
7 (6.4) |
4 (7.7) |
3 (5.3) |
0.707 |
| Stent diameter, mm |
3.16±0.46 |
3.19±0.43 |
3.14±0.49 |
0.675 |
| Stent length, mm |
21.3±7.4 |
22.3±8.7 |
20.4±5.9 |
0.366 |
| IABP |
33 (30.3) |
20 (38.5) |
13 (22.8) |
0.096 |
| Duration of hospitalization, days |
14.0 (11.0, 16.0) |
15.5 (13.0, 18.0) |
13.0 (10.5, 15) |
<0.001 |
| CMR LV/RV functional findings at baseline |
| LVEF, % |
51.7±12.5 |
46.4±12.2 |
56.5±10.8 |
<0.001 |
| LV end-diastolic volume index, mL/m2 |
67.0±19.1 |
73.2±18.1 |
61.3±18.4 |
0.001 |
| LV end-systolic volume index, mL/m2 |
33.4±15.4 |
40.0±15.0 |
27.4±13.3 |
<0.001 |
| LV mass index, g/m2 |
69.3±13.0 |
71.9±13.1 |
66.9±71.9 |
0.022 |
| T1/ECV mapping and other CMR findings at baseline |
Baseline native T1 values in remote
myocardium, ms |
1,169.9±39.7 |
1,189.6±33.4 |
1,151.9±33.4 |
<0.001 |
Baseline native T1 value in edematous
myocardium, ms |
1,321.8±56.3 |
1,338.9±45.8 |
1,306.2±60.68 |
0.006 |
Baseline native T1 value in infarcted
myocardium, ms |
1,499.7±66.3 |
1,498.6±66.3 |
1,500.7±66.8 |
0.660 |
| Baseline ECV value in remote myocardium, % |
26.96±2.87 |
27.74±2.84 |
26.24±2.72 |
0.005 |
| Microvascular obstruction |
51 (46.8) |
51 (98.1) |
0 (0) |
<0.001 |
| ECV value in remote myocardium, % |
26.96±2.87 |
27.74±2.84 |
26.24±2.72 |
0.005 |
| ECV value in edematous myocardium, % |
34.65±4.45 |
35.6±4.2 |
33.8±4.5 |
0.085 |
| ECV value in infarcted myocardium, % |
49.05±9.97 |
51.0±9.0 |
47.3±10.5 |
0.048 |
| AAR index, mm2/g/m2 |
20.15±9.34 |
25.0±9.2 |
16.4±7.4 |
<0.001 |
| MSA index, mm2/g/m2 |
8.93±6.78 |
7.7±4.8 |
10.7±6.4 |
0.013 |
| Myocardial salvage index, % |
47.5±24.2 |
30.7±16.3 |
62.9±19.5 |
<0.001 |
| Transmural extent of LGE >75% |
73 (67.0) |
52 (100) |
21 (36.8) |
<0.001 |
| Infarct index, mm2/g/m2 |
9.66 (4.35, 16.46) |
16.46 (12.10, 22.57) |
4.95 (3.02, 7.98) |
<0.001 |
Missing data: time from symptom onset to reperfusion, n=5; HbA1c, n=5; LDL-C on admission and LDL-C/HDL-C ratio, n=2; monocyte, n=10; BNP at discharge, n=2; high-potency statin before admission, n=2. Data are mean±SD, median (interquartile range), or n (%). β-blockers included carvedilol or bisoprolol, and the doses of bisoprolol were considered to be equivalent to 4-fold of the doses of carvedilol. Diseased artery was defined as the coronary artery with stenosis >50%. % area of infarct or salvaged myocardium was calculated by the infarct area or MSA (mm2) divided by the total myocardial mass (g)×100 (%). Myocardial salvage index was calculated by the MSA divided by the AAR×100 (%). AAR, area at risk; ACEI, angiotensin-converting enzyme inhibitor; AMI, acute myocardial infarction; ARB, angiotensin receptor blocker; BNP, brain natriuretic peptide; CK, creatine kinase; CMR, cardiac magnetic resonance imaging; CRP, C-reactive protein; ECG, electrocardiogram; ECV, extracellular volume; Hct, hematocrit; IABP, intra-aortic balloon pumping; ICs, infarct cores; LGE, late gadolinium enhancement; L/H, low-density lipoprotein/high-density lipoprotein; LAD, left anterior descending artery; LCX, left circumflex artery; LDL-C, low-density-lipoprotein cholesterol; LMT, left main trunk; LVEDP, left ventricular end-diastolic pressure; LVEF, left ventricular ejection fraction; MI, myocardial infarction; MSA, myocardial salvage area; PCI, percutaneous coronary intervention; RCA, right coronary artery; SD, standard deviation; SGLT2, sodium-glucose cotransporter 2; TIMI, Thrombolysis in Myocardial Infarction; WBC, white blood cell count.
The CMR findings are shown in Table 1. Patients with ICs had significantly larger LV volumes and lower LVEFs than those without ICs (P<0.001 and P<0.001, respectively). Moreover, there was a higher incidence of a transmural extent of LGE >75% (P<0.001), a higher percentage of infarcted myocardium and myocardial AAR (P<0.001 and P<0.001, respectively), and a smaller percentage of salvaged myocardium and myocardial salvage index (MSI) (P=0.013 and P<0.001, respectively) in patients with ICs than in those without ICs. Native T1 values and ECV in the remote myocardium were higher in patients with ICs than in those without ICs at baseline (P<0.001 and P=0.017, respectively), akin to ECV at follow-up CMR (P<0.001) (Supplementary Table).
Predictors of ICs
The absence of ischemic preconditioning, STEMI, LMT/LAD lesions, thrombectomy, poor initial and final TIMI flow grades, higher peak CK values, and high Forrester class were positively correlated with the presence of ICs in the univariate logistic analysis (Figure 3). We performed forward stepwise selection to determine candidate factors for multivariate analysis to predict the presence of ICs using 29 covariates. Of these, 24 covariates showed P<0.05 in the univariate analysis, and 5 covariates (age, sex, C-reactive protein level, blood glucose level on admission, and diabetes) were added, as they were reportedly relevant to adverse clinical outcomes and MVO. Finally, 3 variables (peak CK level, T1 value of myocardial edema, and MSI) were independently correlated with ICs among patients with reperfused AMI in the multivariate analysis (odds ratio [OR]: 2.597, 95% confidence interval [CI]: 1.434–4.703, P<0.001; OR: 1.198, 95% CI: 1.040–1.379, P=0.013; and OR: 0.918, 95% CI: 0.869–0.969, P=0.002, respectively). ROC curve analysis to predict IC occurrence suggested that peak CK >2,124 U/L after reperfusion therapy can be used to stratify patients without ICs (area under the curve, 0.905). The sensitivity, specificity, positive predictive value, negative predictive value, and Youden’s index were 0.885, 0.825, 0.835, 0.878, and 0.709, respectively (Figure 4).
Among the 94 patients who underwent 12-month follow-up CMR, LV adverse and reverse remodeling were observed in 14 (14.9%) and 49 (52.1%) patients, respectively. Their characteristics and PCI procedures are shown in Table 2. Episodes of ischemic preconditioning and the administration of a high-potency statin before admission were more common in patients with LVRR than in those without LVRR (P=0.038 and P=0.027, respectively). The peak CK level was significantly lower and the Killip class was significantly better in patients with reverse remodeling than in those without reverse remodeling (P=0.003 and P=0.026, respectively). The use of diuretics at discharge was lower in patients with reverse remodeling than in those without reverse remodeling (P=0.010). Among 24 patients with diabetes, dipeptidyl peptidase-4 inhibitors were more commonly administered in patients with reverse remodeling than in those without reverse remodeling (P=0.019). Patients with reverse remodeling had a lower incidence of LMT/LAD lesions in the culprit artery, blush grade ≤1, and Killip class ≥2 than those without remodeling (P=0.042, P=0.022, and P=0.026, respectively).
Table 2. Clinical Characteristics, PCI and CMR Findings of 94 Patients With and Without LVRR at Baseline and 12-Month Follow-up
| |
Follow-up
(n=94) |
With LVRR
(n=49) |
Without LVRR
(n=45) |
P value |
| Male |
77 (81.9) |
37 (75.5) |
40 (88.9) |
0.113 |
| Age, years |
65.4±10.8 |
64.8±1.5 |
66.1±1.6 |
0.736 |
| Body mass index, kg/m2 |
24.4±3.5 |
24.7±0.5 |
24.1±3.8 |
0.520 |
| Hypertension |
54 (57.4) |
30 (61.2) |
24 (53.3) |
0.532 |
| Diabetes mellitus |
24 (25.5) |
16 (32.7) |
8 (17.8) |
0.155 |
| Dyslipidemia |
70 (74.5) |
37 (75.5) |
33 (73.3) |
0.818 |
| Current smoker |
38 (40.4) |
24 (49.0) |
15 (36.8) |
0.095 |
| ST-elevation MI |
86 (91.5) |
44 (51.2) |
42 (48.8) |
0.716 |
| Episodes of ischemic preconditioning |
40 (42.6) |
26 (65.0) |
14 (35.0) |
0.038 |
| Statin before admission |
6 (6.5) |
6 (12.5) |
0 (0) |
0.027 |
| Antiplatelet loading before PCI |
85 (90.4) |
46 (93.9) |
39 (86.7) |
0.303 |
| Laboratory findings |
| Hematocrit, % |
36.7±5.0 |
36.8±4.6 |
36.5±5.4 |
0.803 |
| Serum creatinine, mg/dL |
0.87±0.20 |
0.84±0.17 |
0.91±0.22 |
0.110 |
| Blood glucose level on admission, mg/dL |
141 (119, 178) |
144 (118, 186) |
141 (124, 172) |
0.835 |
| HbA1c, % |
6.2±1.1 |
6.4±1.2 |
6.0±1.0 |
0.373 |
| LDL-C on admission, mg/dL |
132.0±41.2 |
133.0±42.4 |
131.0±39.9 |
0.941 |
| Peak WBC within 48 h after PCI, /μL |
10,918.1±2,556.4 |
10,702.0±2,682.9 |
11,153.3±2,410.8 |
0.172 |
| Monocytes, % |
7.1 (5.8, 8.2) |
7.6 (5.9, 8.4) |
6.3 (5.6, 7.7) |
0.057 |
| CRP, mg/dL |
5.7 (3.4, 8.6) |
5.8 (3.1, 9.9) |
5.6 (3.5, 7.9) |
0.714 |
| Peak CK, U/L |
2,186 (1,129, 4,134) |
1,589 (912, 2,874) |
2,978 (1,536, 4,931) |
0.003 |
| BNP on admission, pg/mL |
25.6 (10.1, 80.1) |
33.3 (10.2, 83.7) |
25.4 (9.3, 74.1) |
0.575 |
| BNP at discharge, pg/mL |
58.1 (36.2, 127.3) |
54.8 (37.1, 85.3) |
61.1 (33.6, 199.4) |
0.190 |
| BNP at follow-up, pg/mL |
29.4 (16.1, 56.5) |
25.2 (14.7, 42.4) |
44.2 (17.4, 74.1) |
0.006 |
| ECG ST resolution after primary PCI |
| Complete resolution (>70%) |
54 (62.8) |
29 (65.9) |
25 (59.5) |
0.540 |
| Medications following primary PCI |
| Statins after primary PCI |
88 (93.6) |
48 (98.0) |
40 (88.9) |
0.101 |
| β-blockers |
78 (83.0) |
40 (81.6) |
38 (84.4) |
0.788 |
| Equivalent doses with carvedilol, mg |
| At discharge |
5.26±3.73 |
5.41±3.72 |
5.10±3.75 |
0.601 |
| At follow-up |
6.89±4.90 |
7.25±5.17 |
6.51±4.60 |
0.529 |
| Carperitide after primary PCI |
61 (64.9) |
28 (57.1) |
33 (73.3) |
0.131 |
| Loop diuretics at discharge |
15 (16.0) |
3 (6.1) |
12 (26.7) |
0.010 |
| ACEI/ARB at discharge |
94 (100) |
49 (100) |
45 (100) |
1.000 |
| SGLT2 inhibitor at discharge |
3 (3.2) |
2 (4.1) |
1 (2.2) |
1.000 |
| Duration of hospitalization, days |
14 (11, 16) |
13 (11, 15) |
15 (12, 17) |
0.060 |
| Angiographic findings |
| LMT/LAD of culprit artery |
48 (51.1) |
20 (41.7) |
28 (58.3) |
0.042 |
| Number of diseased arteries |
1.7±0.8 |
1.67±0.11 |
1.78±0.12 |
0.715 |
| Initial TIMI ≥2 |
18 (19.1) |
12 (24.5) |
6 (13.3) |
0.198 |
| Final TIMI ≤2 |
19 (20.2) |
7 (14.3) |
12 (26.7) |
0.198 |
| Blush grade ≤1 |
5 (5.3) |
0 (0) |
5 (11.1) |
0.022 |
| Initial Rentrop grade 0 |
59 (62.8) |
29 (49.2) |
30 (50.9) |
0.524 |
| LVEDP, mmHg |
18.2±6.5 |
17.4±5.4 |
19.1±7.7 |
0.332 |
| Forrester class >II |
21 (22.3) |
8 (16.3) |
13 (28.9) |
0.215 |
| Killip class >2 |
8 (8.5) |
1 (2.0) |
7 (15.6) |
0.026 |
| PCI procedure |
| Time from symptom onset to reperfusion, min |
215 (142, 320) |
215 (145, 345) |
216 (131, 302) |
0.580 |
| Thrombectomy |
70 (74.5) |
32 (45.7) |
38 (54.3) |
0.057 |
| Post-balloon dilatation |
51 (54.3) |
25 (51.0) |
26 (57.8) |
0.540 |
| Post-balloon dilatation, mm |
3.16±0.55 |
3.19±0.46 |
3.13±0.63 |
0.969 |
| Stentless PCI |
6 (6.4) |
3 (50.0) |
3 (50.0) |
1.000 |
| Distal protection |
6 (6.4) |
3 (6.1) |
3 (6.7) |
1.000 |
| IABP |
30 (31.9) |
13 (26.5) |
17 (37.8) |
0.274 |
| CMR LV/RV findings |
| Baseline LVEF (%) |
51.9±12.1 |
52.7±11.1 |
51.0±13.1 |
0.568 |
| Baseline LV end-diastolic volume index, mL/m2 |
66.7±19.2 |
68.3±17.0 |
64.9±21.4 |
0.515 |
| Baseline LV end-systolic volume index, mL/m2 |
32.9±14.7 |
32.5±12.6 |
33.4±16.6 |
0.997 |
| Baseline LV mass index, g/m2 |
69.2±12.9 |
68.4±13.4 |
70.0±12.4 |
0.429 |
| Follow-up LVEF, % |
55.90±12.20 |
62.26±9.53 |
48.97±14.57 |
<0.001 |
| Follow-up LV end-diastolic volume index, mL/m2 |
64.77±19.44 |
56.80±15.57 |
73.44±22.93 |
<0.001 |
| Follow-up LV end-systolic volume index, mL/m2 |
30.44±18.72 |
22.00±9.50 |
39.64±21.82 |
<0.001 |
| Follow-up LV mass index, g/m2 |
64.01±11.21 |
63.46±10.43 |
64.61±12.01 |
0.714 |
| Adverse LV remodeling |
14 (14.9) |
0 (0) |
14 (31.1) |
<0.001 |
| CMR T1/ECV mapping and other findings at baseline and follow-up |
| Native T1 values at remote myocardium, ms |
1,170.2±37.5 |
1,167.8±33.9 |
1,172.8±41.0 |
0.222 |
| Native T1 value at myocardial edema, ms |
1,500.6±67.0 |
1,498.8±69.4 |
1,502.6±64.2 |
0.759 |
| Native T1 value at infarct myocardium, ms |
1,319.5±56.8 |
1,317.5±56.1 |
1,321.7±57.6 |
0.596 |
| Native T1 value at IC, ms |
1,175.0
(1,075.8, 1,261.4) |
1,179.1
(1,031.2, 1,269.1) |
1,170.9
(1,031.2, 1,333.1) |
0.961 |
| Microvascular obstruction on LGE image |
45 (47.9) |
13 (26.5) |
32 (71.1) |
<0.001 |
| Hypointense IC on native T1 map (IC) |
46 (48.9) |
13 (26.5) |
33 (73.3) |
<0.001 |
| Baseline ECV value at remote myocardium, % |
226.9±2.8 |
26.9±2.7 |
26.8±2.9 |
0.747 |
| Baseline ECV value at myocardial edema, % |
49.0±10.0 |
47.6±9.8 |
50.4±10.0 |
0.222 |
| Baseline ECV value at infarct myocardium, % |
34.5±4.7 |
34.5±4.8 |
34.6±4.5 |
0.985 |
| Baseline AAR index, mm2/g/m2 |
20.3±9.1 |
18.5±8.3 |
22.3±9.9 |
0.061 |
| Baseline MSA index, mm2/g/m2 |
9.0±5.9 |
9.5±6.3 |
8.4±5.4 |
0.489 |
| Baseline myocardial salvage index, % |
46.1±23.9 |
51.6±22.8 |
40.2±23.8 |
0.015 |
| Baseline transmural extent of LGE >75% |
62 (66.0) |
26 (53.1) |
36 (80.0) |
0.009 |
| Baseline infarct index, mm2/g/m2 |
10.2 (4.8, 16.8) |
7.3 (4.0, 11.8) |
12.6 (8.1, 19.6) |
0.002 |
| Follow-up ECV value in remote myocardium, % |
26.12±3.74 |
25.46±3.27 |
26.71±4.11 |
0.375 |
| Follow-up ECV value in myocardial edema, % |
47.68±13.43 |
42.06±8.82 |
52.74±16.49 |
0.002 |
| Follow-up ECV value in infarct myocardium, % |
31.74±7.24 |
31.61±6.42 |
31.86±7.90 |
0.713 |
Missing data: body mass index, n=12; time from symptom onset to reperfusion, n=5; HbA1c, n=4; LDL-C on admission and LDL-C/HDL-C ratio, n=2; monocyte, n=9; BNP at discharge, n=1; Statins before admission, n=2. LV end-diastolic volume, n=17; baseline ECV data, n=1. LVRR, left ventricular reverse remodeling. Other abbreviations as in Table 1.
Patients with LVRR showed significantly lower incidences of ICs, MVO, and transmural extent of LGE >75%, as well as a lower infarct index and higher MSI at baseline, than those without remodeling (P<0.001, P<0.001, P=0.009, P=0.002, and P=0.015, respectively). Additionally, significant improvements in both LV and right ventricular function and size were observed in patients with LVRR compared with those without remodeling. Patients with LVRR showed a significantly lower follow-up ECV in myocardial edema (P=0.002); of the 14 patients (14.9%) with adverse LV remodeling, none had LVRR (Table 2).
ICs and LVRR
Previous episodes of ischemic preconditioning, peak CK values, peak monocyte counts, BNP level at discharge, LMT/LAD lesions, thrombectomy, Killip class, administration of diuretics, the presence of ICs, LGE transmurality, MSI, and the percent area of infarcted myocardium were significantly related to chronic LVRR in the univariate analysis (Figure 5). We performed forward stepwise selection to determine candidate factors for the multivariate analysis to predict LVRR using 17 covariates. Of these, 13 covariates had P<0.05 in the univariate analysis, to which we added 4 covariates (age, sex, initial TIMI coronary flow, and T1 value in edematous myocardium) because they were reportedly relevant to adverse clinical outcomes. The latter 2 strongly correlated with the occurrence of ICs in this study. Consequently, multivariate logistic regression analysis revealed that the presence of ICs and LMT or LAD lesions were independent negative predictors of LVRR (OR: 0.180, 95% CI: 0.040–0.770, P<0.001; and OR: 0.311, 95% CI: 0.104–0.923, P=0.035, respectively), and the peak monocyte count within 48 h after reperfusion was an independent positive predictor (OR: 1.352, 95% CI: 1.025–1.784, P=0.033).
Discussion
We investigated the significance of hypointense ICs on non-contrast myocardial T1 mapping using 3.0-T CMR to predict the occurrence of chronic LV remodeling among patients with reperfused AMI. The main findings of this study were: (1) ICs detected in 47.7% of patients with reperfused AMI; (2) close association of the presence of ICs with AMI-related (e.g., peak CK level, Killip class, Forrester class, etc.) severity indices of LV damage, such as exacerbation of acute LV function and dilatation; (3) a higher peak CK value, lower MSI, and higher T1 value at myocardial edema by native T1 mapping as independent predictors of ICs; and 4) the presence of ICs as the strongest independent negative predictor of chronic LVRR.
Detection and Significance of ICs
Our study demonstrated that ICs were detected using non-contrast T1 mapping in 47.7% of patients with reperfused AMI. T1 mapping has the following advantage: reduced patient breath-holding and scan times, resulting in a lower frequency of uninterpretable images compared with conventional analyses using T2* and spin-echo imaging.14 All patients were analyzed by native T1 mapping; however, some patients who underwent T2-weighted or ECV imaging had uninterpretable images due to motion artifacts. Interestingly, the T1 values of ICs were more widely distributed, with a median of 1,181 ms (1,095–1,260 ms), than those of other myocardial tissues. Low T1 values in the widely distributed ICs may reflect the presence of hemorrhage within the MVO because iron deposits after myocardial hemorrhage shorten T1 values. Conversely, the higher water content due to inflammation prolongs T1 values.
Carrick et al11 conducted a CMR study involving 267 patients with reperfused AMI using a 1.5-T scanner. They suggested that the lower the T1 value in the ICs, the higher the risk of adverse remodeling, all-cause death, or first hospitalization for heart failure post-discharge. They also implied that it was feasible to identify ICs using non-contrast T1 mapping; moreover, its prognostic value was comparable to that of MVO revealed via LGE imaging.11 Notably, ICs detected by 3.0-T CMR native T1 mapping had better conformity with MVO detected by contrast-enhanced CMR images compared with ICs detected using 1.5-T CMR. Our 3.0-T CMR had better sensitivity, specificity, and accuracy than 1.5-T CMR in MVO detection (sensitivity: 0.981 vs. 0.856; specificity: 1.000 vs. 0.938; accuracy: 0.991 vs. 0.892).
In this study, patients with ICs had higher peak CK levels and smaller areas of salvaged myocardium. The presence of ICs reflected more severe AMI, even after primary PCI. As a result, patients with ICs showed lower EFs, larger LV dilatations, and higher BNP levels at discharge, as well as a lower incidence of chronic LVRR than those without ICs. The early detection of ICs is essential to stratify patients with poor prognoses.
T1 Hypointense ICs and MVO
In our study using 3.0-T CMR, MVO was detected in 47.2% of patients with reperfused AMI, and 98.1% of patients with T1 hypointense ICs had MVO. A previous study using 1.5-T CMR showed that 145 (50.3%) of 288 patients with reperfused AMI had MVO, and 137 (85.6%) of 160 patients with ICs had MVO.11 Our study showed a better concordance between IC and MVO occurrence, probably because of the higher resolution of 3.0-T CMR on T1 mapping and the timing of CMR.
MVO size and hemorrhage quantity markedly change over time after reperfusion. Fernandez et al reported that the quantity of myocardial hemorrhage increased from day 4 to day 7.23 In the study by Carrick et al,11 CMR images were obtained 2 days after reperfusion therapy; thus, MVO size and myocardial hemorrhage quantity might have been underestimated in their results.
ICs and LVRR
It is essential to predict the occurrence of chronic LVRR because reverse remodeling correlates with favorable long-term clinical outcomes in patients with AMI.24 In our study, several discriminators were discovered that predict such occurrence in patients with reperfused AMI. Among them, the presence of ICs was the strongest negative predictor for LVRR at the 12-month follow-up. In addition, patients without ICs showed smaller infarct sizes and larger areas of salvaged myocardium after reperfusion therapy. Therefore, LV dysfunction and LV enlargement rarely occurred during the follow-up of patients without ICs. Conversely, we found that the peak CK level, which is a well-established predictor of death and myocardial recovery,25 was not a predictor of LVRR. One possible reason is that elevated CK levels ordinarily reflect the spread and extent of “damaged” myocardium but not necessarily the extent of severe transmural infarction or the presence of microvascular injury. Using contrast-enhanced CMR, we demonstrated that patients with chronic LVRR had less transmural myocardial damage and MVO. Therefore, the simple, objective and noninvasive detection of ICs via native T1 mapping may help predict the improvement of acute LV dysfunction and enlargement in the chronic phase of AMI for all patients with reperfused AMI, including those with renal dysfunction and contrast allergy (Figure 6).
The association between ICs and clinical adverse events in this cohort may not show in this study because the number of patients with major adverse clinical outcomes, which are a composite of cardiac death, MI, and unplanned rehospitalization for heart failure, was extremely small (n=3, 3.2% in patients with ICs vs. none in those without), though there was a worse trend of the incidence of major adverse clinical outcomes in patients with ICs in the Kaplan-Meier curve statistically analyzed by the log rank test (P=0.07).
Study Limitations
Our study had several limitations. First, it was a prospective cohort single-center study with a small sample size, so several analyses might have been statistically underpowered. The incidence of clinical adverse events could not be discussed in this report due to the small number of cases. Second, the strategies for PCI and medication administration after coronary reperfusion depended on the physician. Third, although native T1 mapping was assessed without contrast enhancement, the infarcted area, salvaged area, AAR, and ECV map were evaluated using contrast-enhanced CMR. Thus, patients with renal dysfunction were excluded and such patients may show different results.
Conclusions
Non-contrast T1 mapping detected hypointense ICs in 47.7% of patients with reperfused AMI, and the presence of T1 hypointense ICs was the most valuable predictor of LVRR after 12 months of follow-up.
Acknowledgments
We thank the patients who participated in this study and the staff of Seiwa Medical Center.
Conflicts of Interest / Funding
None.
IRB Information
The Ethics Committee of Nara Prefecture Seiwa Medical Center (approval no. 58).
Data Availability
✓ The deidentified participant data will not be shared.
Supplementary Files
Please find supplementary file(s);
https://doi.org/10.1253/circj.CJ-22-0479
References
- 1.
Huttin O, Mandry D, Eschalier R, Zhang L, Micard E, Odille F, et al. Cardiac remodeling following reperfused acute myocardial infarction is linked to the concomitant evolution of vascular function as assessed by cardiovascular magnetic resonance. J Cardiovasc Magn Reson 2017; 19: 2.
- 2.
Zijlstra F, Hoorntje JC, de Boer MJ, Reiffers S, Miedema K, Ottervanger JP, et al. Long-term benefit of primary angioplasty as compared with thrombolytic therapy for acute myocardial infarction. N Engl J Med 1999; 341: 1413–1419.
- 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: 13–20.
- 4.
Bahit MC, Kochar A, Granger CB. Post-myocardial infarction heart failure. JACC Heart Fail 2018; 6: 179–186.
- 5.
Watanabe H, Ozasa N, Morimoto T, Shiomi H, Bingyuan B, Suwa S, et al; CAPITAL-RCT investigators. Long-term use of carvedilol in patients with ST-segment elevation myocardial infarction treated with primary percutaneous coronary intervention. PLoS One 2018; 13: e0199347.
- 6.
Bolognese L, Neskovic AN, Parodi G, Cerisano G, Buonamici P, Santoro GM, et al. Left ventricular remodeling after primary coronary angioplasty: Patterns of left ventricular dilation and long-term prognostic implications. Circulation 2002; 106: 2351–2357.
- 7.
Hamirani YS, Wong A, Kramer CM, Salerno M. Effect of microvascular obstruction and intramyocardial hemorrhage by CMR on LV remodeling and outcomes after myocardial infarction: A systematic review and meta-analysis. JACC Cardiovasc Imaging 2014; 7: 940–952.
- 8.
Hellawell JL, Margulies KB. Myocardial reverse remodeling. Cardiovasc Ther 2012; 30: 172–181.
- 9.
Wright RS, Reeder GS, Herzog CA, Albright RC, Williams BA, Dvorak DL, et al. Acute myocardial infarction and renal dysfunction: A high-risk combination. Ann Intern Med 2002; 137: 563–570.
- 10.
Cui C, Wang S, Lu M, Duan X, Wang H, Jia L, et al. Detection of recent myocardial infarction using native T1 mapping in a swine model: A validation study. Sci Rep 2018; 8: 7391.
- 11.
Carrick D, Haig C, Rauhalammi S, Ahmed N, Mordi I, McEntegart M, et al. Prognostic significance of infarct core pathology revealed by quantitative non-contrast in comparison with contrast cardiac magnetic resonance imaging in reperfused ST-elevation myocardial infarction survivors. Eur Heart J 2016; 37: 1044–1059.
- 12.
Orn S, Manhenke C, Anand IS, Squire I, Nagel E, Edvardsen T, et al. Effect of left ventricular scar size, location, and transmurality on left ventricular remodeling with healed myocardial infarction. Am J Cardiol 2007; 99: 1109–1114.
- 13.
Bhatt AS, Ambrosy AP, Velazquez EJ. Adverse remodeling and reverse remodeling after myocardial infarction. Curr Cardiol Rep 2017; 19: 71.
- 14.
Mehta RH, Harjai KJ, Cox D, Stone GW, Brodie B, Boura J, et al. Clinical and angiographic correlates and outcomes of suboptimal coronary flow in patients with acute myocardial infarction undergoing primary percutaneous coronary intervention. J Am Coll Cardiol 2003; 42: 1739–1746.
- 15.
Hendriks T, Hartman MHT, Vlaar PJJ, Prakken NHJ, van der Ende YMY, Lexis CPH, et al. Predictors of left ventricular remodeling after ST-elevation myocardial infarction. Int J Cardiovasc Imaging 2017; 33: 1415–1423.
- 16.
Cohn JN, Ferrari R, Sharpe N. Cardiac remodeling: Concepts and clinical implications: A consensus paper from an international forum on cardiac remodeling. Behalf of an International Forum on Cardiac Remodeling. J Am Coll Cardiol 2000; 35: 569–582.
- 17.
Hadamitzky M, Langhans B, Hausleiter J, Sonne C, Byrne RA, Mehilli J, et al. Prognostic value of late gadolinium enhancement in cardiovascular magnetic resonance imaging after acute ST-elevation myocardial infarction in comparison with single-photon emission tomography using Tc99 m-Sestamibi. Eur Heart J Cardiovasc Imaging 2014; 15: 216–225.
- 18.
Masci PG, Ganame J, Strata E, Desmet W, Aquaro GD, Dymarkowski S, et al. Myocardial salvage by CMR correlates with LV remodeling and early ST-segment resolution in acute myocardial infarction. JACC Cardiovasc Imaging 2010; 3: 45–51.
- 19.
Bulluck H, Go YY, Crimi G, Ludman AJ, Rosmini S, Abdel-Gadir A, et al. Defining left ventricular remodeling following acute ST-segment elevation myocardial infarction using cardiovascular magnetic resonance. J Cardiovasc Magn Reson 2017; 19: 26.
- 20.
de Meester de Ravenstein C, Bouzin C, Lazam S, Boulif J, Amzulescu M, Melchior J, et al. Histological validation of measurement of diffuse interstitial myocardial fibrosis by myocardial extravascular volume fraction from Modified Look-Locker imaging (MOLLI) T1 mapping at 3 T. J Cardiovasc Magn Reson 2015; 17: 48.
- 21.
Zeng M, Zhang N, He Y, Wen Z, Wang Z, Zhao Y, et al. Histological validation of cardiac magnetic resonance T1 mapping for detecting diffuse myocardial fibrosis in diabetic rabbits. J Magn Reson Imaging 2016; 44: 1179–1185.
- 22.
Nakamori S, Dohi K, Ishida M, Goto Y, Imanaka-Yoshida K, Omori T, et al. Native T1 mapping and extracellular volume mapping for the assessment of diffuse myocardial fibrosis in dilated cardiomyopathy. JACC Cardiovasc Imaging 2018; 11: 48–59.
- 23.
Fernández-Jiménez R, Barreiro-Pérez M, Martin-García A, Sánchez-González J, Agüero J, Galán-Arriola C, et al. Dynamic edematous response of the human heart to myocardial infarction: Implications for assessing myocardial area at risk and salvage. Circulation 2017; 136: 1288–1300.
- 24.
Spinelli L, Morisco C, Assante di Panzillo E, Izzo R, Trimarco B. Reverse left ventricular remodeling after acute myocardial infarction: The prognostic impact of left ventricular global torsion. Int J Cardiovasc Imaging 2013; 29: 787–795.
- 25.
Halkin A, Stone GW, Grines CL, Cox DA, Rutherford BD, Esente P, et al. Prognostic implications of creatine kinase elevation after primary percutaneous coronary intervention for acute myocardial infarction. J Am Coll Cardiol 2006; 47: 951–961.
