Risk Factors for Nonischemic St-segment Elevation in Patients With Electrocardiographic Left Ventricular Hypertrophy

Background ST-segment elevation (STE) is not a specic change for ST-segment elevation myocardial infarction (STEMI). This may lead to a mistaken diagnosis of STEMI and false-positive cardiac catheterization laboratory activation. We aimed to investigate risk factors for STE secondary to electrocardiographic LVH in order to provide more information for differential diagnosis. Methods A total of 1,590 inpatients with electrocardiographic LVH without confounding factors (such as myocardial infarction) were enrolled in this study. Data on potential risk factors and patient characteristics were collected. Logistic regression analysis and receiver operating characteristic curve (ROC) were used to identify the risk of STE in patients with LVH. Results After reviewing the ECGs, 1590 cases of electrocardiographic LVH were divided into an ST-segment elevation group (STE group, 81 cases) and non-ST segment elevation group (1509 cases). Eighty-seven cases were randomly selected from the non-ST segment elevation group to form a new non-ST segment elevation group (non-STE group, 87 cases) for further analysis. The mean age of the 168 participants (119 men, 70.83%) was 62.33 ± 16.27. Multivariate analysis showed that stroke, infection, and the value of S V1 +R V5 were signicantly associated with STE secondary to LVH. The area under the receiver operating characteristic curve showed that the optimal value of S V1 +R V5 cut-off for predicting STE was 4.805 (sensitivity: 40.74%; specicity: 80.46%; AUC: 0.634; 95% CI: 0.550–0.719; P < 0.05). Conclusions A value of S V1 +R V5 larger than 4.8 mV, stroke, and infection are independent risk factors for STE in patients with electrocardiographic LVH. fever (body temperature greater than 38 ℃ ), infection (required antibiotic treatment), acute chest pain, stroke (cerebral hemorrhage, cerebral infarction), biliary system disease (cholecystitis, biliary obstruction), history of hypertension, history of diabetes, history of coronary heart disease, smoking history, heart rate, blood pressure, white blood cells, hemoglobin, platelets, serum potassium ions, serum sodium ions, serum calcium ions, alanine acid transaminase (ALT), serum creatinine, cardiac enzymes (kinase isoenzyme (CK-MB), cardiac troponin I (cTnI)) and S V1 + R V5 values. This study was approved Ethics Committee are as mean ± SD or as Categorical variables are presented as absolute values and proportions. Continuous variables were compared using one-way analysis of variance. Categorical variables were compared using the χ2-test or Fisher’s exact test, as appropriate. Logistic regression analysis was used to estimate odds ratios (OR) for STE and their corresponding 95% condence intervals (CI). Univariate regression analysis was performed to evaluate the association between clinical variables and STE secondary to LVH.


Background
ST-segment elevation myocardial infarction (STEMI) is usually caused by unstable plaque rupture of the coronary arteries, resulting in myocardial necrosis in the corresponding perfusion area, and is one of the most common fatal chest pains in the emergency department [1][2][3][4] . Rapid diagnosis and shortening of reperfusion therapy time are essential for reducing mortality and improving outcomes in STEMI patients 1,5 . The diagnosis of STEMI is based primarily on symptoms, electrocardiograms (ECG), and myocardial markers 1,5 . ECG is a powerful clinical tool to assist physicians in arriving at the correct diagnosis and determining proper therapy, in particular the reperfusion therapy 6 . The typical ECG changes of myocardial infarction are characterized by ST-segment elevation (STE). However, ECG is an imperfect tool in this setting 6 . STE is not a speci c for a change in SETMI. STE can also appear in benign, nonischemic presentations, such as electrocardiographic left ventricular hypertrophy (LVH) 7,8 . Patients with electrocardiographic LVH can develop complex ST-T changes (including STE), which may lead to a mistaken diagnosis of STEMI [9][10][11][12][13][14] . Previous studies have shown that electrocardiographic LVH is the most signi cant independent risk factor in leading to a mistaken diagnosis of STEMI and false-positive cardiac catheterization laboratory activation. 11,12 .
Current guidelines fail to provide a fast and effective method to identify STEMI in patients who have combined electrocardiographic LVH with STE 1 . Most studies look for ECG characteristics and clinical features of STEMI patients combined with LVH to help identify cases of myocardial infarction 6 Electrocardiographic features were collected by experienced cardiovascular physicians. ST-segment elevation is de ned as an STE that appeared in any ECG lead and an elevation amplitude greater than 0.05 mV 1 . According to the STE status in the electrocardiogram, participants were divided into an STE group and non-STE group. Eighty-seven patients were randomly selected from the non-STE group by computer to form a new non-STE group for analysis. Data collection Clinical data of patients in the STE group and new non-STE group were collected. We considered the following covariates as potential confounders: gender, age, fever (body temperature greater than 38℃), infection (required antibiotic treatment), acute chest pain, stroke (cerebral hemorrhage, cerebral infarction), biliary system disease (cholecystitis, biliary obstruction), history of hypertension, history of diabetes, history of coronary heart disease, smoking history, heart rate, blood pressure, white blood cells, hemoglobin, platelets, serum potassium ions, serum sodium ions, serum calcium ions, alanine acid transaminase (ALT), serum creatinine, cardiac enzymes (kinase isoenzyme (CK-MB), cardiac troponin I (cTnI)) and S V1 + R V5 values. This study was approved by the Ethics Committee of the First A liated Hospital of Shantou University Medical College.
Statistical analysis Continuous variables are expressed as mean ± SD or medians (interquartile range) as appropriate. Categorical variables are presented as absolute values and proportions. Continuous variables were compared using one-way analysis of variance. Categorical variables were compared using the χ2-test or Fisher's exact test, as appropriate. Logistic regression analysis was used to estimate odds ratios (OR) for STE and their corresponding 95% con dence intervals (CI). Univariate regression analysis was performed to evaluate the association between clinical variables and STE secondary to LVH.
Variables that were statistically signi cant in the univariate analysis were entered into a multivariable logistic model. The area under the receiver operating characteristic curve (AUC) was determined to assess the discriminative ability of clinical variables in predicting STE. All statistical analyses were performed using SPSS Statistics version 23.0 (IBM Corporation, Armonk, NY, USA) and HemI 1.0 (Huazhong University of Science and Technology, Hubei, China). An α level of 0.05 was deemed signi cant for all analyses.

Results
Baseline characteristics A total of 67,589 electrocardiograms from the First A liated Hospital of Shantou University Medical College between July 2015 and June 2017 were collected. After screening based on our inclusion and exclusion criteria, 21,054 inpatient ECGs were enrolled for further analysis. After reviewing the ECGs, 1590 cases of electrocardiographic LVH, but without confounding factors, were collected and divided into an ST-segment elevation group (STE group, 81 cases) and non-ST segment elevation group (1509 cases). Eighty-seven cases were randomly selected from the non-ST segment elevation group to form a new non-ST segment elevation group (non-STE group, 87 cases) for analysis. The detailed process for participants selection is detailed in Fig. 1.
The clinical characteristics are shown in Table 1. The mean age of the 168 participants (119 men, 70.83%) was 62.33 ± 16.27 years. There were signi cant differences in ratio of fever, stroke, infection, heart rate, and S V1 +R V5 between STE and non-STE groups (P < 0.05, Table 1). Comparing with the non-STE groups, the ratio of fever, stroke, heart rate and infection were higher in the STE group. Other factors such as gender ratio, age, biliary tract disease, acute chest pain, history of hypertension, history of diabetes, history of coronary heart disease, and smoking history did not have signi cant difference between the STE group and non-STE group. Association between clinical variables and STE Univariate regression analysis showed that the ratio of fever, stroke, infectious disease, heart rate, and value of S V1 +R V5 were signi cantly associated with STE (Table 2). remained independently associated with the outcome in our multivariable model (Fig. 2).

Discussion
LVH alters the measurement of cardiac repolarization. Characteristic ECG changes associated with LVH including STE, prominent septal Q waves, T-wave inversion, and ST-segment depression, which may lead to misdiagnosis of acute myocardial infarction 6 . The presence of LVH, when associated with STE, has been demonstrated as a risk factor for false-positive ST-segment elevation myocardial infarction diagnoses, commonly leading to unnecessary reperfusion therapy [9][10][11][12][13][14] . Chest pain centers have been built in many countries. Chest pain centers have been shown to shorten the reperfusion therapy time, which is associated with better outcomes for myocardial infarction patients 3 . However, in order to shorten the reperfusion time, physicians at chest pain centers often do not have enough time for careful differential diagnosis 12 . Previous studies have shown that in the early stages of chest pain center construction, the incidence of false activation of cardiac catheterization laboratory increased, and the occurrence of false activation events was closely related to electrocardiographic LVH 12,14 . More seriously, previous publications have shown that LVH with STE is an independent risk factor for misdiagnosing STEMI in patients with aortic dissection 19,20 . Aortic dissection is a catastrophic disease with rapid onset, rapid progression, high mortality, and poor natural prognosis 19 . Patients with aortic dissection who are misdiagnosed as STEMI will lead to serious consequences 21,22 . Therefore, physicians should be keenly aware of the possibility of LVH confounding the ability to recognize true STEMI 6  in leads V 2 -V 3 for women and 2 mm (0.2 mV) in the same leads for men) 23 . Interestingly, these guidelines de ne diagnostic thresholds in the absence of LVH, but the guidelines do not clarify how to diagnose STEMI in the case of LVH, which might result in many unnecessary, potentially dangerous coronary angiographies.
Based on the de ciencies of current clinical guidelines, we need to nd more methods to help clinicians make a differential diagnosis. There have been many studies looking for ECG characteristics and clinical features of STEMI patients with LVH to help identify cases of myocardial infarction. For example, a study by Armstrong et al. suggested using a ratio of ST-segment to R-S-wave magnitude ≥ 25% as a diagnostic criteria for STEMI to improve speci city of diagnosis in patients with anterior territory STE 17 . However, using these additional ECG diagnostic criteria and clinical features to identify STEMI may reduce the sensitivity of the ECG for STEMI diagnosis. To date, no clinical studies have been performed to identify independent predictors of STE in nonischemic patients with LVH. Therefore, we tried to investigate the risk factors for STE in patients with LVH but without STEMI in order to provide more information for differential diagnosis.
Risk factors for STE After researching the literature, we collected and analyzed common potential risk factors that may affect the ST segment. According to multivariate logistic regression analysis, a value of S V1 +R V5 larger than 4.8 mV is an independent risk factor for STE in patients with electrocardiographic LVH. S V1 +R V5 is the main indicator of the Sokolow-Lyon standard in diagnosing LVH, which re ects the projection of the largest vector of the left ventricular depolarization on the horizontal plane 6,18 . Our study further nds that the value of S V1 +R V5 could not only be used to diagnose electrocardiographic LVH, but also that the magnitude of S V1 +R V5 is positively correlated with the risk of nonischemic STE. Therefore, LVH patients combined with STE and an elevated S V1 +R V5 (larger than 4.8 mV) should be highly suspect for the possibility of secondary nonischemic STE. Conversely, if STE occurs when S V1 +R V5 is not signi cantly elevated (less than 4.8 mV), then STEMI might be considered to avoid misdiagnosis and delaying primary angioplasty. To determine how the S V1 +R V5 performs to differentiate between ischemic STE and nonischemic STE secondary to LVH, a threshold of 4.8 mV was chosen by optimizing receiver operating characteristic (ROC) curves. For a patient with an S V1 +R V5 larger than 4.8 mV, physicians should be keenly aware of the possibility of nonischemic STE secondary to LVH. Otherwise, STEMI should be considered. The speci city is high (0.8) but the sensitivity is low (0.4) on the ROC curve. Given that missed diagnosis of STEMI could lead to serious consequences, high speci city is needed to minimize the possibility of missed diagnosis. Considering that the magnitude of S V1 +R V5 is an additional auxiliary differential diagnosis method, lower sensitivity is acceptable. Therefore, a value of S V1 +R V5 greater than 4.8 mV could be used for clinical differential diagnosis and is worth further research.
We also found that stroke is related with the incidence of STE in patients with LVH. Patients with stroke are more likely to have STE, and the incidence of STE is 2.11-fold higher than that without stroke. The mechanism is not clear. We believe this might be related to brain-heart interaction, a group of heart disease secondary to central nervous system disorders, such as stroke [24][25][26] . Previous publications have reported that ECG abnormalities in stroke occur in 70% -80% of cases, and are mainly represented as ST-T abnormalities (including STE) [27][28][29][30] . In the current study, we found that superimposed stroke based on electrocardiographic LVH results in a signi cant increase in the incidence of STE.
Previous publications indicate that infection has important effects on the cardiovascular system. That pneumonia is a risk factor for acute cardiac complications has been documented thoroughly in several large cohorts 31,32 . Therefore, we included infection as a potential risk factor in the logistic regression analysis. According to the results of the multiple logistic regression analysis, patients combined with infection and electrocardiographic LVH are more prone to STE, and the incidence of STE is increased by 108% compared with non-infected patients. the Sokolow-Lyon standard, Cornell standard, and Gubner-Ungerleider standard. This study only used the Sokolow-Lyon standard to diagnose ECG LVH. However, the Sokolow-Lyon criterion has a higher diagnostic speci city, which is recommended by the guidelines for hypertension 30,36 . Fourth, the potential in uencing factors we included may not be comprehensive, and there may be other potential in uencing factors for ST segment elevation that have not been included and analyzed. However, the factors we analyzed cover common clinical indicators and factors already mentioned in the literature.
More importantly, the factors we include in the analysis are easy to obtain, being routine indicators of clinical laboratory tests. Therefore, the potential in uencing factors we discussed could be applied to different grades of hospitals, enabling our conclusions to have broader application.

Conclusion
Physicians should be keenly aware of the possibility of nonischemic STE secondary to electrocardiographic LVH. An S V1 +R V5 value larger than 4.8 mV, stroke, and infection are independent risk factors for STE in patients with electrocardiographic LVH.

Declarations
Ethics approval and consent to participate The study was approved by the ethics committee of the First A liated Hospital of Shantou University Medical College. The need for consent was waived because of the retrospective data. Dr. Xiansheng Huang granted administrative permission to access the raw data.

Figure 2
Multivariate logistic regressions analysis for STE secondary to LVH LVH= left ventricular hypertrophy; STE= ST-segment elevation.