Simplified Selvester QRS Score as an Infarct Size Parameter in STEMI Patients Undergoing Pharmacoinvasive or Primary Percutaneous Coronary Intervention

Background: Cardiovascular disease especially acute myocardial infarction (AMI) is one of the highest causes of mortality worldwide. The majority of AMI comes in the form of an ST-elevation myocardial infarct (STEMI) that requires timely diagnosis and revascularization management to restore myocardial circulation. The simple method to estimate infarct size is by using a simplified Selvester QRS Score to electrocardiogram records, which is a tested method that has a good correlation with the gold standard, namely cardiac magnetic resonance imaging. Objective: To investigate the difference in infarct size with simplified Selvester QRS score parameters between STEMI patients undergoing pharmacoinvasive compared to primary PCI. Methods: Eighty-two STEMI patients, 41 of pharmacoinvasive and 41 of primary PCI were scored with simplified Selvester QRS score from electrocardiogram recording. Patient data are retroactively taken from Sardjito Cardiovascular Intensive Care (SCIENCE) registry. Scoring simplified Selvester QRS Score was done by two experienced cardiologists blinded to the patient procedure, and results were then measured for interobserver agreement with the Bland-Altman test. A comparison of QRS Scores in pharmacoinvasive and primary PCI groups was done with an independent sample T-test followed by a multivariable linear regression test. Results: The means of simplified Selvester QRS score in the pharmacoinvasive and primary PCI group is 7.240±3.015 and 8.900±4.188, p=0.043, respectively. Independent sample T-test shows a significant difference in the simplified Selvester QRS score in the pharmacoinvasive and primary PCI groups. The multivariable analysis shows that variables other than the revascularization method independently influence QRS score are onset, anterior segment ST elevation, and ST-segment elevation in more > than 3 leads in the electrocardiogram. Conclusion: There is a significant difference in infarct size measured by simplified Selvester QRS score between STEMI patients undergoing pharmacoinvasive method compared to primary PCI procedure, which is lower in the pharmacoinvasive group.


M
anagement of STEMI patients requires timely and accurate diagnosis and immediate revascularization which aims to minimize myocardial cell damage and complications. 1,2In general, it is known that the short and long-term prognosis of STEMI patients is directly related to the size of left ventricular infarction. 3,4herefore, a reliable method is required to estimate the extent of myocardial infarction. 5There are studies about the evaluation of several methods aimed at estimating the size of myocardial infarction. 6,7The gold standard of this measurement is the use of Contrast-Enhanced Magnetic Resonance Imaging, as this method provides precise in vivo estimation of infarct area in a noninvasive manner. 8However, this examination is difficult to perform because of the limited number of facilities and resources and high cost as well. 9Another method that is simpler and easier to use at a lower cost is the application of a scoring system by electrocardiogram recording. 10,11ur hospital, Dr. Sardjito Hospital is a tertiary center in the region that can carry out revascularization measures for STEMI by using primary PCI and pharmacoinvasive methods.Primary PCI is an attempt to revascularize the coronary artery without the administration of fibrinolysis drugs but with coronary intervention using stents, balloons, or other recommended devices.The pharmacoinvasive method is an attempt to revascularize the coronary artery by giving full doses of fibrinolytic agents, both specific and nonspecific, and followed by angiography and PCI within 2 to 24 hours, namely immediate rescue PCI if fibrinolysis fails or angiography evaluation and PCI as appropriate for successful fibrinolysis.Our previous study found that the average time for performing primary PCI was 200 minutes, which was longer than the target of less than 120 minutes. 12This delay was caused by many factors, both technical and non-technical constraints. 12There has never been a study comparing the myocardial infarct area in STEMI patients undergoing revascularization with fibrinolysis and primary PCI in our hospital.The study aims to investigate the difference in myocardial infarct area between STEMI patients who received the pharmacoinvasive method and who underwent primary PCI by assessing post-intervention electrocardiogram recordings using the Selvester QRS scoring system.The Selvester QRS score is a scoring system developed by Selvester et al.(1985) using 12-lead electrocardiogram recordings to estimate the extent of myocardial infarct size. 5It consists of 54 criteria with 32 points based on the criteria for Q and R wave duration, R and S wave amplitude, R/Q and R/S ratio, and QRS notch in 10 of the 12 electrocardiogram leads. 5

Methods
This research was an observational study.The study design was cross-sectional.The data taken were secondary data of STEMI patients who were treated at RSUP Dr. Sardjito and had been registered in the Sardjito Cardiovascular Intensive Care Registry (SCIENCE registry).The subjects of this study were STEMI patients who were treated at the intensive cardiovascular care unit (ICCU) of Dr. Sardjito Hospital that were registered in the SCIENCE registry starting in July 2019.The sampling was carried out consecutively from the subjects who met the research criteria until the number of subjects was met.The inclusion criteria were (1) patients with male and female sexes, (2) 18 years old or older, (3) patients diagnosed with STEMI with onset < 24 hours, (4) patients underwent revascularization by primary PCI or fibrinolysis and (5) patients were treated at the ICCU of Dr. Sardjito hospital and (6) patients registered in the SCIENCE registry.The exclusion criteria were (1) patients who underwent fibrinolysis and performed angiography in > 24 hours, (2) patients who underwent fibrinolysis but not receiving fulldose fibrinolytic agents, (3) patients with a history of coronary artery bypass surgery, (4) patients with a history of previous infarction, (5) patients with a history of pacemaker insertion, (6) patients with confounding factors from the electrocardiogram recording (bundle branch block, fascicular block, ventricular hypertrophy, and preexcitation), (7) an unreadable quality of electrocardiogram recording, and (8) patients with congenital heart disease and coronary anomalies.
The electrocardiogram recording used was carried out with a standard 12-lead surface electrocardiography, with a voltage configuration of 10mm/mv and a speed of 25mm/s.The electrocardiogram recording was done after the primary PCI procedure in the acute phase, which was 30-60 minutes after the primary PCI procedure was completed, then the QRS score was calculated using the Simplified Selvester QRS Scoring System method which consisted of 37 criteria and produced a maximum of 29 points.The simplified Selvester QRS score is a simplification of the Selvester QRS scoring system in which qualitative criteria such as the notch and absolute amplitude of the QRS wave deflection are not evaluated, resulting in 37 criteria being applied to 10 ECG leads, resulting in 29 points.The electrocardiogram assessment was carried out by two investigators who were blinded to the patient's procedure.The investigators were experienced cardiologists, who had been educated on how to assess the Selvester Simplified QRS Score.Furthermore, the Bland-Altman test was also carried out to find the interobserver agreement of the numerical calculation results.
The basic characteristic data were presented in frequency and percentage for categorical data and the form of mean and standard deviation for numerical data.The normality of the data will be analyzed by the Shapiro-Wilk test.The comparative analysis for numerical data was performed by an unpaired T-test if the data variance was normal and the Mann-Whitney test if the data variance was not normal.Confounding factor analysis was performed using bivariate analysis followed by multivariate analysis.Bivariate analysis was performed with Pearson's or Spearman's if the variables were numerical.The analysis used an unpaired t-test or Mann-Whitney test if the variables were categorical.The multivariate analysis was performed with multiple linear regression.A p-value <0.05 was set as a statistical significance.

Figure 1. The flow of research subjects enrollment and sample selection
From the results of the Bland-Altman test, the observers obtained a mean difference of -0.889 with 95% CI -2.2447 to 0.4669 with p-value = 0.169, no statistical difference in mean was observed.Based on these data, the interobserver conformity had a good match and there was no difference in assessing the Simplified Selvester QRS Score.

Result
The STEMI patients who met the inclusion and exclusion criteria were retrieved consecutively from the SCIENCE registry.From the registry, there were 460 STEMI patients.A total of 91 patients were not included in the study because the onset was more than 24 hours or unknown.From the remaining 367 patients, 58 patients were excluded because they had a history of previous myocardial infarction.After excluding patients whose electrocardiogram recordings could not be interpreted, had confounders, and did not meet the definition (figure 1), consecutive sampling was carried out until the number of subjects was met.Eventually, 41 subjects were enrolled in the pharmacoinvasive group and 41 patients in the primary PCI group.The basic characteristics of subjects were depicted in Table 1.
From the results of the Bland-Altman test, the Value in means±standard deviation or median (interquartile range) for numerical data observers obtained a mean difference of -0.889 with 95% CI -2.2447 to 0.4669 with p-value = 0.169, no statistical difference in mean was observed.Based on these data, the interobserver conformity had a good match and there was no difference in assessing the Simplified Selvester QRS Score.
In this study, the Simplified Selvester QRS Score was tested for data distribution in the pharmacoinvasive and the primary PCI group using the Shapiro Wilk test.The Simplified QRS Score data in both the pharmacoinvasive group (p= 0.055) and the primary PCI group (p= 0.057) indicated a normal distribution.
There was a statistically significant difference in the myocardial infarct size as measured by the Simplified Selvester QRS Score parameter between STEMI patients who received the pharmacoinvasive method and primary PCI (mean score: 7.240±3.015vs. 8.900±4.188,p=0.043).The myocardial infarct size in the pharmacoinvasive group was significantly smaller than in the primary PCI group, with a mean difference of 1.66 (Table 2).
Bivariate analysis showed that body mass index (p=0.44),location of ST-elevation (p=0.02),number of leads with ST-segment elevation on electrocardiogram (p=0.026), and culprit artery (0.026) were significantly associated with Simplified Selvester QRS Score, in addition to the method of revascularization.Table 3 and Table 4 show the bivariate analysis for confounding variables and their correlation with the Simplified Selvester QRS Score.
The variables included in the multivariate analysis were variables that have a p-value <0.25 from the bivariate analysis.These variables were body mass index, onset (p=0.221),hypertension medication (p=0.205),diabetes mellitus (p=0.203),location of segment elevation ST in anterior or inferior, number of leads with ST-segment elevation on electrocardiogram, and culprit artery in left coronary (LCA) or right coronary artery (RCA), along with revascularization method.The multivariate analysis showed that onset (p=0.032),location of segment elevation ST in anterior (p=0.044), and several leads with ST-segment elevation on diagnostic ECG >3 (p=0.026) were independently associated with Simplified Selvester QRS Score.The method of revascularization, namely pharmacoinvasive and primary PCI, did not independently associated with the Simplified Selvester QRS Score (p=0.543)(Table 5).

Discussion
This study aimed to compare the simplified Selvester QRS score between STEMI patients who underwent the pharmacoinvasive method and patients who underwent primary PCI.This study showed that the mean simplified QRS Selvester score in the pharmacoinvasive group was significantly smaller than in the primary PCI group.Other than with the revascularization method, the simplified Selvester QRS score significantly correlated with body mass index, location of the ST segment elevation in the anterior and inferior, the number of ST-segment elevations in the electrocardiogram, and the location of the lesions in the LCA or RCA.Further multivariate analysis showed that the method of revascularization, namely the pharmacoinvasive method and primary PCI, did not independently affect the simplified Selvester QRS score.Other confounding variables which influenced the simplified Selvester QRS score were the location of the ST-elevation in the anterior, the onset of STEMI, and the number of ST-segment elevation >3 segments in the electrocardiogram.
The difference in simplified Selvester QRS score between groups receiving pharmacoinvasive versus primary PCI may be influenced by the timing of the electrocardiogram taking, namely in the acute phase within 48 hours of onset.In previous studies, the electrocardiogram taking time used to calculate QRS scores was much longer which is more than 7 days from onset.13,14This is due to the extent of infarct tissue, and Selvester QRS scores as its parameter can undergo significant changes up to 30 days from the onset of STEMI.The reason is that in the acute post-infarction phase there is still viable myocardial tissue, which undergoes a stunning and hibernating phase, where this tissue has not caused changes in the QRS complex that can be detected by the Selvester QRS scoring system.Myocardial tissue, the majority of which is infarcted tissue, will cause visible changes in the QRS complex, but the electrocardiogram recording and the scoring system cannot distinguish and detect areas consisting of a mixture of necrotic and viable tissue.More complex and sophisticated methods for detecting infarct areas, such as cardiac magnetic resonance imaging, can distinguish and detect these partial differences. 15here are not many previous studies have compared the infarct area based on different revascularization methods by using the scoring system of the electrocardiogram recording.Previous studies have generally compared infarct size between conservatively managed versus revascularized patients, with either fibrinolysis or PCI. 16n our study, after adjusting with confounding variables, it showed that the difference in the simplified Selvester QRS score between the pharmacoinvasive group and the primary PCI group was not independent.Other variables affected the QRS score, namely the onset of ischemia, the location of the ST segment elevation in the anterior leads, and the number of leads with ST-segment elevation > 3 leads.This means that the confounding factors mentioned above, such as the onset of STEMI and anterior infarction, will affect the patient's Selvester QRS score, apart from the revascularization procedure of STEMI in patients, both pharmacoinvasive and primary PCI.Following the results of previous studies, where a longer onset means a longer ischemic time which will lead to a wider infarct area, which is reflected in a higher QRS score.
The number of ST segment elevations that appear in the ECG leads also indicates the area of ischemia, where the more leads with ST-segment elevation, the larger the area of ischemia and infarction. 6,17his research has several limitations.The main limitation of this study was the use of electrocardiogram recordings from medical records or registry data retrospectively so it was not possible to take further patient electrocardiograms at a longer time after the onset.Another limitation was related to the measuring instruments used in the study, namely the electrocardiogram and its QRS scoring system in estimating the area of infarct tissue and distinguishing viable and non-viable myocardial tissue, which was not a gold standard. 18

Conclusion
There was a significant difference in infarct area measured from the simplified Selvester QRS score between STEMI patients who underwent pharmacoinvasive method compared to primary PCI, which was lower in the pharmacoinvasive group.The association between simplified Selvester QRS score with a different method of revascularization was not independent, since there were confounding variables for this association, namely onset of STEMI, anterior location of infarction, and ST-segment elevation in electrocardiogram recording > 3 leads.

AcutePrimaryFigure 1 .
Figure 1.The flow of research subjects enrollment and sample selection.

Table 1 .
Basic characteristics of subjects from the pharmacoinvasive and primary PCI group.

Table 2 .
The difference in the value of the simplified Selvester QRS score between the pharmacoinvasive method and primary PCI groups.

Table 3 .
Bivariate analysis of numerical confounding variables using Pearson and Spearman tests.

Table 4 .
Bivariate analysis of categorical variables withMann-Whitney test or Independent sample T Test.

Table 5 .
Results of multivariate analysis of confounding variables using the backward.model.