ST-Segment Resolution as a Marker for Myocardial Scar in ST-Segment Elevation Myocardial Infarction

Objective: To investigate the relationship between ST-segment resolution (STR) and myocardial scar thickness after percutaneous coronary intervention (PCI) in patients with ST-segment elevation myocardial infarction (STEMI). Methods: Forty-two STEMI patients with single-branch coronary artery stenosis or occlusion were enrolled. ST-segment elevations were measured at emergency admission and at 24 h after PCI. Late gadolinium-enhanced cardiac magnetic resonance imaging (CMR-LGE) was performed 7 days after PCI to evaluate myocardial scars. Statistical analyses were performed to assess the utility of STR to predict the development of transmural (>75%) or non-transmural (<75%) myocardial scars. Results: The sensitivity and specicity of STR for predicting transmural scars were 96% and 88%, respectively, at an STR cut-off value of 40.15%. The area under the curve was 0.92. Multivariate logistic proportional hazards regression analysis disclosed that patients with STR<40.15% had a 112.95-fold higher probability of developing transmural scars compared with patients with STR ≥ 40.15%. STR percentage was negatively correlated with myocardial scar thickness (β=-0.838, P<0.001) and size (β=-0.714, P<0.001). Conclusion: STR<40.15% at 24 h after PCI may provide meaningful diagnostic nformation regarding the extent of myocardial scarication in STEMI patients. are as mean ± SD, median (interquartile ranges), or number (%).


Introduction
Transmural myocardial scars of the left ventricle that complicate ST-segment elevation myocardial infarction (STEMI) predispose to heart failure and cardiac death. Therefore, early identi cation and analysis of myocardial scars are particularly important [1]. The size and tissue heterogeneity of healing scars identi ed by late gadolinium-enhanced cardiac magnetic resonance imaging (CMR-LGE) are independent predictors of arrhythmia and sudden cardiac death [2][3][4][5][6][7]. CMR-LGE is the gold standard for the diagnosis of myocardial scar and assessment of myocardial salvage [8][9][10][11], but is expensive, timeconsuming, un t for wide population studies, and generally contraindicated in patients with cardiac implants. These disadvantages have restricted its clinical application.
Delayed ST-segment resolution (STR) is prevalent during major adverse cardiovascular events, and is predictive of arrhythmia, heart failure, and 30-day mortality [12]. We hypothesized that analysis of STR could represent a convenient, widely accessible, and inexpensive diagnostic method for patients who cannot tolerate CMR-LGE. Consequently, the aim of our study was to establish whether poor STR, as well as CMR-LGE ndings, can detect myocardial scari cation in the early post-infarction period in STEMI patients.

Study design and population
Forty-two consecutive patients with STEMI treated with coronary angiography and percutaneous coronary intervention (PCI) within 12 h of the onset of pain were enrolled between March 2017 and October 2017. Inclusion criteria were (1) single-branch coronary artery stenosis or occlusion, and (2) restoration of coronary perfusion to TIMI ow grade 3 after PCI. Exclusion criteria were a prior history of the acute coronary syndrome; coronary revascularization; severe chronic kidney disease; intracardiac pacing leads or other implants precluding CMR-LGE; hemodynamic instability; or known claustrophobia.
The study was performed at the First A liated Hospital of Chongqing Medical University, China.
Demographic and clinical characteristics including ECG STR 24 h after PCI were recorded. CMR-LGE was performed 7 days after PCI. This study was conducted in accordance with the Declaration of Helsinki.
The research protocol was approved by the locally appointed Ethics Committee, and written informed consent was obtained from all study participants.

CMR-LGE protocols
Patients were examined in the supine position using a 1.5-T imaging unit (Signa In nity Twinspeed, General Electric Healthcare, USA) equipped with master gradients (30 mT/m peak gradients; 150 mT/m/ms slew rate) and a 5-element cardiac phased-array receiver coil. Images were obtained using electrocardiographic gating and expiratory breath holds. A dose of 0.2 mmol/kg of body weight of gadopentetate dimeglumine (Magnevist; Bayer Schering Health Care, Cambridge, UK) was administered intravenously at a rate of 5 ml/s with a power injector. Ten minutes after contrast agent injection, a Look-Locker sequence was performed to obtain the most appropriate inversion time to nullify the signal intensity of normal myocardium. The left ventricular short-axis imaging layer was 8 mm thick and 0 mm apart. The left ventricular 2-chamber and 4-chamber scanning imaging layers were 5 mm thick and 0 mm apart. This was immediately followed by the acquisition of LGE images, with an inversion recovery prepared T1-weighted gradient-echo sequence (4.9/1.9; ip angle, 15 degrees; turbo eld-echo factor, 30; spatial resolution, 1.35×1.35×10 mm). Late gadolinium enhancement was interpreted as present or absent by the consensus of two CMR-trained physicians, and was considered present only if con rmed on both short-axis and matching long-axis myocardial locations.
First, we marked the segments of the myocardial scar with the bull's eye segmental comparison (17segment model) and compared them with the results of coronary angiography. Second, we searched for the thickest myocardial scar layer-by-layer on the short axis imaging and calculated the percentage of the thickness of the myocardial scar, which was de ned as a transmural myocardial scar when the percentage was > 75%. Finally, we recorded the area of myocardial scar layer-by-layer on the short axis imaging and multiplied the thickness of the layer to determine the total volume of myocardial scar, and to calculate the percentage of myocardial scar volume.

ECG protocols
Standard 18-lead ECGs were obtained at emergency admission and 24 h after coronary angiography and PCI. The TP segment was used as the isoelectric line in the ST-segment measurement. The ST segment was measured 20 ms after the J point. The summed ST-segment elevation was measured by adding the ST amplitude in all leads with ST-elevation at emergency admission and 24 h after PCI [13]. STR percentage was calculated as the initial sum of ST-segment elevation minus the sum of ST-segment elevation on the second ECG, divided by the initial sum of ST-segment elevation.

Statistical analysis
Basic descriptive statistics were used. CMR-LGE location data were described on a patient-by-patient basis (Supplemental Fig. 1). Statistical analyses were performed to assess the clinical utility of using of STR to predict myocardial scari cation. Myocardial scars were assessed using two measures: (1) scar thickness and (2) scar size. Receiver operating characteristic (ROC) curve analysis, logistic regression analysis, and multivariate logistic proportional-hazards regression estimated transmural myocardial scar.
Scar thickness and size were estimated using linear regression and Student's t-test. All statistical analyses were performed using SPSS v.22.0 (IBM, Armonk, NY, USA). P-value < 0.05 was considered statistically signi cant.

Results
Location, size, thickness of the myocardial scar determined by CMR-LGE Myocardial scars were diagnosed in 41 of 42 STEMI patients (96.7%) by using CMR-LGE. A patient-bypatient visual analysis of scar tissue location in the STEMI group, with bull's eye segmental comparison of CMR-LGE ndings is shown in Supplemental Figure 1. In all patients, the anatomic locations of scars de ned by CMR-LGE corresponded to the distributions of the culprit vessels treated with primary angioplasty. For example, in a patient with angiographically proven left anterior descending coronary artery occlusion, CMR-LGE indicated scari cation of the basal and middle segments of the left anterior ventricular wall. In another patient with right coronary artery occlusion, CMR-LGE disclosed a scar that involved the entire inferior wall of the left ventricle and the middle and apical segments of the posterior interventricular septum (Figure 1). However, scar size and thickness were unrelated to the degree of coronary artery occlusion. Stenoses in all culprit arteries exceeded 90%; nonetheless, there were signi cant inter-patient differences in scar size and thickness (P<0.001).
Determination of STR cut-off value A transmural scar was de ned as a myocardial lesion extending >75% of the wall thickness. All myocardial scars were classi ed as either non-transmural (0-75%) or transmural (76-100%) according to CMR-LGE results. The relationship of the ST-segment resolution percentage to transmural scari cation was identi ed by the ROC curve. The ROC curve analysis demonstrated a sensitivity of 96% and a speci city of 88% to predict transmural myocardial scari cation following STEMI at an STR cut-off value of 40.15%. The area under the curve was 0.92 ( Figure 2).

Discussion
Our study showed that ECG, as a convenient and non-invasive technique, can monitor the occurrence of transmural myocardial brosis after acute myocardial infarction. In our study, it is possible to detect myocardial sar by CMR-LGE at 7 days after PCI, even when PCI is performed within 12 h of the onset of pain. Many previous studies have focused on the correlation between STR in ECG and poor prognosis in patients with MI, but no cutoff value and clinical markers have been formed. Our study attempted to quantify the predictive value of STR for transmural myocardial scar after STEMI using a small sample.In our study population, there were no statistically signi cant differences in gender, age, BMI, smoking history, hypertension history, diabetes history, hyperlipidemia history, culprit artery, number of stent implantation, hemogram, HbA1c, ALT, CCU time and the use of drugs including IIa/IIIb inhibitor, Aspirin, Ticagrelor/Clopidogrel, Statins, β-blocker, ACEI/ARB, Nitrates, Diuretics. However, compared with the STR ≥ 40.15% group, the incidence of transmural myocardial brosis was higher (P < 0.001) and the size of myocardial scar was larger (P = 0.001) in the STR < 40.15% group, the difference was statistically signi cant (Table 1).
It is signi cant to quantify STR and apply such a convenient and non-invasive technique to the clinical monitoring of transmural brosis in myocardial infarction. Although CMR-LGE is the gold standard for the diagnosis of myocardial brosis, patients with cardiac scar often cannot tolerate CMR-LGE and other methods are needed to diagnose transmural scar. STR is a useful predictor of the nal infarct size, left ventricular function, and clinical outcome after early reperfusion [14][15][16]. Nonetheless, the predictive value of STR is still controversial [17,18]. Rakowski et al. showed that STR < 70% is a marker of large infarct size [18]. However, the transmural depth of an infarction is more consequential than its size. To the best of our knowledge, there are no data to date on the relationship between STR and the transmural thickness of infarcted myocardium. Therefore, our study was focused on de ning a speci c cut-off value of STR for the diagnosis of transmural scars. We found that the predicted critical value of STR of a transmural myocardial scar after STEMI was 40.15%, with a sensitivity of 96% and a speci city of 88% ( Figure. 2). Poor STR was related to thicker and larger scars ( Figure. 3). Previous studies have reported relationships between STR and reduced myocardial perfusion and between early STR and myocardial rescue [19]. In addition, STR following PCI and restoration of perfusion to TIMI ow grade 3 was correlated with collateral circulation [20]. STR < 50% may be associated with worse left ventricular function and increased mortality [11,21]. These results are consistent with our ndings, but our study found that patients with STR > 40.15% had thinner and smaller myocardial scar.
STR is signi cant in monitoring the prognosis and treatment of transmural myocardial brosis after myocardial infarction. Transmural myocardial scars are caused by ischemic injury followed by brosis of necrotic tissue [1,[22][23][24]. Poor perfusion, limited myocardial salvage, and microvascular disease promote scari cation. The severity and localization of ventricular wall injuries are also in uenced by the length of coronary artery stenosis and the degree of collateral circulation [25][26][27]. The extent of myocardial brosis is an important determinant of prognosis. In STEMI, transmural MF usually leads to irreversible ventricular remodeling and heart failure, and is also the pathological basis for arrhythmias, and is generally considered to be associated with sudden cardiac death[28-31]. Nguyen et al. [32] found that the severity of myocardial brosis was signi cantly correlated with the frequency of ventricular arrhythmias (r = 0.83, P < 0.01). Our study showed that the STR < 40.15% group had a higher incidence of transmural myocardial brosis, and no differences in the incidence of arrhythmias and readmission rate were found between the two groups, which may be related to the small sample size (Supplemental Fig. 2). Scientists' understanding of the mechanisms and consequences of cardiac brosis has only improved greatly in recent years, with the improvement of non-invasive techniques to better track its development [33]. It is signi cant to explore the effect of STR on the prognosis of transmural myocardial brosis after myocardial infarction in a larger sample and a longer follow-up time.
The physiology of STR is related to the restoration of myocardial perfusion. Following PCI of epicardial coronary arteries, microvascular spasm and embolism may lead to persistent coronary microvascular dysfunction (CMD) and subsequently cause myocardial and especially endocardial ischemia. In the setting of CMD, extracellular potassium ion clearance is decreased, thus prolonging repolarization and delaying STR. Poor STR re ects microvascular and left ventricular dysfunction [17,34,35], and is thereby an important biomarker of CMD after PCI in STEMI patients. Inadequate perfusion due to CMD is the proximate cause of transmural myocardial injury; consequently, assessment of the severity of ischemia by monitoring dynamic ST-segment changes is of the utmost importance. In our study, it was found that compared with the group with STR ≥ 40.15%, the pain to balloon time was signi cantly prolonged (P = 0.001), the left ventricular ejection fraction was signi cantly reduced (P < 0.001), and the troponin and BNP were higher (P < 0.001) in the group with STR < 40.15%. The delay of opening time of effective coronary blood ow in STR < 40.15% group may be an important reason for the aggravation of myocardial transmural injury and the formation of transmural myocardial brosis. Shortening the pain to balloon time may avoid or reduce the occurrence of poor STR, thus affecting the clinical prognosis of patients. There are several limitations of this study. First, this was a cross-sectional study with a relatively small number of patients. Second, it was di cult to recruit STEMI patients who were willing or able to undergo CMR-LGE 7 days after PCI. Third, because this study was limited to STEMI patients, crossvalidation analysis is needed to determine whether delayed STR can be used to predict myocardial scari cation in non-STEMI patients. Fourth, to verify the clinical effects of STR, longer clinical follow-up is needed, especially for the monitoring and follow-up of malignant arrhythmias. We found that STR correlated with myocardial scar thickness following STEMI. To the best of our knowledge, this is the rst study to con rm that STR < 40.15% after PCI can provide important prognostic information regarding myocardial brosis in STEMI patients. These results suggest that STR may represent a safe, readily accessible, easily administered, inexpensive diagnostic modality in the management of STEMI patients for whom CMR-LGE is contraindicated.

Declarations Ethics approval
This study was registered in clinicaltrials.gov (Approval No. NCT04586582). The study was conducted in accordance with the "Declaration of Helsinki" and approved by the ethics committee of the rst a liated hospital of Chongqing medical university (approval No.2019-057). We followed ethical guidelines and obtained informed consent from the participants.

Con icts of interest
The authors declare that they have no con icts of interest.

Availability of data and material
The datasets used or analysed during the current study are available from the corresponding author on reasonable request.

Funding
The study was supported by