Construction of a predictive model for acute myocardial infarction combined with acute heart failure based on electrocardiogram parameters and exploration of the correlations of these parameters with the Killip grade

Construction of a predictive model for acute myocardial infarction combined with acute heart failure based on electrocardiogram parameters and exploration of the correlations of these parameters with the Killip grade | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Construction of a predictive model for acute myocardial infarction combined with acute heart failure based on electrocardiogram parameters and exploration of the correlations of these parameters with the Killip grade Xuefen Guo, Guoping Yan, Hua He, Aihui Xu, Xueting Zhan, Guoliang Gao This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7286686/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 18 Feb, 2026 Read the published version in Scientific Reports → Version 1 posted 13 You are reading this latest preprint version Abstract Acute myocardial infarction (AMI) is a leading cause of acute heart failure (AHF), which markedly increases patient mortality and readmission rates. This study aimed to develop a practical, ECG-based model for early AHF risk prediction in AMI patients. We retrospectively analyzed clinical and electrocardiogram data from 301 AMI patients (October 2022–March 2025). Six predictors—prolonged QTc interval, abnormal Q wave, heart rate>100 bpm, reduced left ventricular ejection fraction, male sex, and age (60–75 years)—were selected via LASSO and incorporated into a logistic regression nomogram. The model demonstrated strong discrimination (AUC: 0.840; internal validation AUC: 0.823) and clinical utility (threshold probability: 0.13–0.65). Killip grade was positively correlated with QTc interval, heart rate, age, and QRS duration, and negatively with P-wave duration and dispersion. This nomogram offers a reliable and resource-efficient tool for early identification of AHF risk in AMI patients, especially in low-resource or primary care settings. Health sciences/Cardiology Health sciences/Diseases Health sciences/Medical research Health sciences/Risk factors Acute heart failure Electrocardiogram Acute myocardial infarction Nomogram model Killip classification Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Introduction Acute myocardial infarction, otherwise known as a heart attack, continues to be the leading cause of mortality worldwide. The development of acute heart failure in the wake of acute myocardial infarction has substantial implications for a patient's survival and likelihood of rehospitalization, despite the fact that mortality specifically due to myocardial infarction following medical interventions and pharmaceutical treatments is decreasing. Statistics indicate that approximately one-fifth to one-third of all individuals admitted to a hospital for myocardial infarction also experience acute heart failure, which considerably prolongs the duration of hospitalization and increases the risk of readmission and death, especially in primary care facilities.[ 1 ] As a routine examination method, electrocardiogram (ECG) plays an important role in reflecting the severity of myocardial injury and predicting late adverse outcomes. Recent studies have shown that certain ECG parameters are positively associated with the occurrence of acute heart failure. For example, a prolonged QTc interval not only reflects electrical conduction disorders but also significantly increases[ 2 ] the risk of heart failure after myocardial infarction. Fragmented QRS complex (fQRS) has shown good predictive efficacy[ 3 ] in myocardial infarction patients both with and without ST-segment elevation, and this result[ 4 ] has been confirmed by other scholars. In addition, QRS-T angle enlargement is often predictive of myocardial structural remodeling and instability of electrical activity, which often cause heart failure or death.[ 5 ] Zorlu[ 6 ] et al. reported that the QRS-T angle is closely related to rehospitalization for chronic heart failure. Turkmen[ 4 ] also confirmed the important role of ECG parameters in predicting arrhythmia and left ventricular remodeling after myocardial infarction by analyzing QT dispersion and the Tp-e interval. Although Martínez-Sellés[ 7 ] et al. attempted to use artificial intelligence models to integrate multiple ECG features for risk prediction, several problems, such as computing platform dependence, dataset resources, and technical thresholds, need to be solved before the clinical application of such models. Therefore, multiple traditional 12-lead ECG parameters (e.g., the QTc interval, fQRS, the QRS-T angle and the Tp-e interval) were combined according to practical clinical requirements, and common biomarkers (e.g., general clinical data and the left ventricular ejection fraction) were selected via LASSO regression. A simple and clear structural nomogram model was then constructed to both achieve the early identification of acute heart failure and quantify the risk of acute heart failure in patients with myocardial infarction. The purpose of this study was to provide a set of practical and feasible risk assessment tools for primary hospitals and to promote the accurate hierarchical management of cardiovascular diseases. Methods Study Subjects Patients receiving standard treatment for acute myocardial infarction at our hospital between October 2022 and March 2025 were included in this single-center, retrospective, observational study (Fig. 1 ). The inclusion criteria were as follows: 1) AMI according to the diagnostic criteria from the European Heart Association's 2023 guidelines[ 8 ]; 2) age of 18 years or older; and 3) complete clinical data from examinations at our hospital. The exclusion criteria were as follows: 1) malignant tumor, severe infection, autoimmune disease, or end-stage liver or kidney dysfunction; 2) death due to noncardiac causes during hospitalization; and 3) incomplete or missing key data. Acute heart failure was diagnosed according to the 2023 Focus Update of the European Society of Cardiology Guidelines[ 9 ] for Heart Failure from 2021. The Killip grade was used to assess the severity of acute heart failure in AMI patients; acute heart failure was classified as grade I, II, III, or IV based on the criteria first proposed by Killip and Kimball in 1967. In this study, the Killip grade was used to assess the severity of acute heart failure in patients with acute heart failure upon admission according to clinical manifestations. Clinical data collection General demographic data (sex, age), data related to the history of underlying diseases (hypertension, diabetes, hyperlipidemia, arrhythmia, heart failure, etc.), test results obtained upon admission (troponin levels, liver and kidney function, etc.) and auxiliary examination reports (electrocardiogram, echocardiography, etc.) were collected. Related indicators were uniformly tested at our hospital. All patients underwent standard 12-lead electrocardiogram within 24 hours after admission. The equipment and procedures used were specified by the hospital. Two physician-qualified electrocardiologists independently judged the results, and discrepancies were reviewed by a third expert. The 10 extracted ECG parameters included the following: fragmented QRS complex (representing a myocardial scar or conduction problem), abnormal Q wave (Q-wave duration ≥ 40 ms and amplitude ≥ 0.1 mV fror two adjacent leads), low-voltage QRS complex (QRS amplitude 40 ms), QRS complex prolongation (≥ 120 ms), QTc interval prolongation (QTc interval > 450 ms in men or > 460 ms in women, corrected by Bazett's formula), QT dispersion (maximum and minimum QT interval difference > 60) ms), T peak-to-end interval (Tp-e interval, reflecting the waveform difference during ventricular repolarization), and heart rate (> 100 bpm or ≤ 100 bpm). All the parameters were converted to binary variables according to clinical definitions for statistical analysis and to assess their ability to predict the risk of acute myocardial infarction complicated by acute heart failure. To standardize variable definitions and improve the clinical interpretability of the model, the latest international guidelines and high-level research were referenced, and a common clinical classification method for some continuous variables was adopted. For ECG parameters, QTc interval prolongation was defined as a QTc interval > 460 ms (> 450 ms in men and > 460 ms in women) according to the 2022 European Heart Rhythm Society guidelines[ 10 ] for the management of arrhythmias. A QRS duration > 120 msec was defined as an intraventricular conduction delay, as indicated by the 2023 Focus Update of the ESC guidelines[ 9 ] for the diagnosis and treatment of heart failure. Resting tachycardia was defined as a heart rate > 100 beats/min; a P wave duration > 120 ms was used as an indicator of left atrial enlargement and AF risk. QT dispersion > 50 ms and a Tp-e interval > 90 ms indicated increased dispersion of ventricular repolarization and an increased risk of arrhythmia, respectively. Fragmented QRS (fQRS) and low-voltage QRS were identified by electrocardiogram (ECG), and relevant definitions were provided by Villa.[ 11 ] In accordance with the 2023 Focus Update of the ESC guidelines[ 9 ] for the diagnosis and treatment of heart failure, patients were divided into three age groups, namely, < 60 years, 60–74 years and ≥ 75 years, to evaluate differences in the risk of acute heart failure across different age groups. Observation Endpoints The primary outcome was the occurrence of acute heart failure and the Killip grade during AMI. According to the 2023 Focused Update of the ESC guidelines, acute heart failure was defined as dyspnea[ 9 ] with pulmonary rales at rest, pulmonary congestion or edema on imaging, or a significant increase in the NT-proBNP level requiring intervention (e.g., the use of diuretics, vasodilators or inotropic drugs). All endpoint events were independently judged by two senior cardiologists, and if the results were inconsistent, a third physician resolved disputes. Model construction and evaluation The aim of our research was to develop a predictive model for assessing acute heart failure risk in patients recently diagnosed with acute myocardial infarction. By analyzing clinical data from a large cohort of myocardial infarction patients, we utilized LASSO regression to identify key variables associated with subsequent heart failure. These factors were then incorporated into a logistic regression model for risk calculation. Model performance was assessed using receiver operating characteristic curves and calculating the area under the curve. We also performed bootstrapping with 2000 iterations to generate a calibration curve and evaluate model fit. Additionally, decision curve analysis was used to examine whether the model provided worthwhile clinical benefit across a range of risk thresholds. A diverse collection of sentences was used to aid model interpretation while maintaining the overall word count and conceptual similarity to the original text. Statistical Methods A variety of statistical software and packages were utilized for the diverse data analysis. LASSO regression was carried out using the glmnet package to screen for key predictors, whereas the rms package was used to construct a multivariate logistic regression model and generate a calibration curve to assess model fit. The pROC package was used to generate the ROC curve and calculate the AUC for determining predictive ability. Decision curve analysis was performed to evaluate the net clinical benefit under different thresholds using the rmda package. Spearman rank correlation analysis was used to evaluate the correlation between the Killip grade (I-IV) as the rank variable and each parameter, and correlation coefficients and P values were calculated. For meaningfully linked factors, violin plots were further constructed to visualized their distribution patterns across the different Killip grades. Continuous variables are expressed as the means ± standard deviations or medians [quartiles] based on the data distribution. Independent samples t tests or Mann‒Whitney U tests were used to compare the two groups. Categorical variables are presented as frequencies and proportions and were compared via chi-square tests or Fisher's exact tests. P < 0.05 was considered to indicate statistical significance. Results Comparison of baseline characteristics Table 1 displays the clinical and electrocardiographic data of 301 registered individuals with AMI grouped according to the presence or absence of AHF. Of the 301 registered individuals, 185 (61.5%) were in the AHF group, and 116 (38.5%) were within the non-AHF (NAHF) group. A chi-square test or Mann‒Whitney U test was used to compare variables between the 2 groups according to the type of variable. The results showed notable variations in a variety of clinical and ECG parameters. There were significantly more women in the AHF group (26.9%) than in the NAHF group (11.3%). Atrial fibrillation (14.6%) and fQRS (17.2%) were much more common in the AHF group than in the NAHF group. Moreover, abnormal Q waves, a history of smoking, low-voltage QRS and a prolonged QTc interval (> 460 ms) were more common in the AHF group than in the NAHF group (all P < 0.05) and were closely related to AHF. Critically, the median LVEF in the AHF group, which was 47.0% [38.5, 51.7], was considerably lower than that in the NAHF group, which was 58.5% [56.2, 63.0] (P = 0.05). Table 1 Baseline characteristics of patients with acute myocardial infarction complicated with acute heart failure Variable Total (n = 301) AHF (n = 185) NAHF (n = 116) P value Sex 0.002 * Male 238 (79.1%) 136 (73.1%) 102 (88.7%) Female 63 (20.9%) 50 (26.9%) 13 (11.3%) Atrial Fibrillation 0.004 * Yes 31 (10.3%) 27 (14.5%) 4 (3.5%) No 270 (89.7%) 159 (85.5%) 111 (96.5%) Presence of fragmented QRS (fQRS) 0.009 * Yes 39 (13.0%) 32 (17.2%) 7 (6.1%) No 262 (87.0%) 154 (82.8%) 108 (93.9%) Abnormal Q Wave 0.000 * Yes 161 (53.5%) 118 (63.4%) 43 (37.4%) No 140 (46.5%) 68 (36.6%) 72 (62.6%) Low-voltage QRS 0.002 * Yes 54 (17.9%) 44 (23.7%) 10 (8.7%) No 247 (82.1%) 142 (76.3%) 105 (91.3%) Smoking 0.002 * Yes 145 (48.2%) 76 (40.9%) 69 (60.0%) No 156 (51.8%) 110 (59.1%) 46 (40.0%) Alcohol use 0.002 * Yes 63 (20.9%) 28 (15.1%) 35 (30.4%) No 238 (79.1%) 158 (84.9%) 80 (69.6%) Hypertension 0.817 Yes 187 (62.1%) 117 (62.9%) 70 (60.9%) No 114 (37.9%) 69 (37.1%) 45 (39.1%) Hyperglycemia 0.133 Yes 87 (28.9%) 60 (32.3%) 27 (23.5%) No 214 (71.1%) 126 (67.7%) 88 (76.5%) Hyperlipidemia 0.401 Yes 42 (14.0%) 23 (12.4%) 19 (16.5%) No 259 (86.0%) 163 (87.6%) 96 (83.5%) P-wave duration > 120 ms 0.908 Yes 52 (17.3%) 33 (17.7%) 19 (16.5%) No 249 (82.7%) 153 (82.3%) 96 (83.5%) Prolonged P-wave dispersion 0.067 Yes 19 (6.3%) 16 (8.6%) 3 (2.6%) No 282 (93.7%) 170 (91.4%) 112 (97.4%) QRS-wave duration > 120 ms 0.016 * Yes 27 (9.0%) 23 (12.4%) 4 (3.5%) No 274 (91.0%) 163 (87.6%) 111 (96.5%) QTc interval > 460 ms 0.000 * Yes 105 (34.9%) 89 (47.8%) 16 (13.9%) No 196 (65.1%) 97 (52.2%) 99 (86.1%) QT dispersion (QTd) > 50 ms 0.337 Yes 111 (36.9%) 73 (39.2%) 38 (33.0%) No 190 (63.1%) 113 (60.8%) 77 (67.0%) Tp-e interval > 90 ms 0.599 Yes 232 (77.1%) 141 (75.8%) 91 (79.1%) No 69 (22.9%) 45 (24.2%) 24 (20.9%) Heart rate > 100 bpm 0.000 * Yes 38 (12.6%) 34 (18.3%) 4 (3.5%) No 263 (87.4%) 152 (81.7%) 111 (96.5%) Left ventricular ejection fraction (LVEF), median (Q1, Q3) 55 (49, 59) 50 (47, 57) 58 (55, 62) 0.000 * Age group 0.000 * 75 83 (27.6%) 57 (30.6%) 26 (22.6%) N-terminal pro-B-type natriuretic peptide (NT-proBNP), median (Q1, Q3) 1255 (445, 3231) 374.5 (211.57, 660.5) 2651 (1295, 5166) 0.000 * AHF: acute heart failure; NAHF: nonacute heart failure; ∗P < 0.05. Table 2 Results of univariate and multivariate regression analyses of the risk of acute heart failure in patients with acute myocardial infarction Variable Univariate OR (95% CI) Univariate P Multivariate OR (95% CI) Multivariate P Sex 0.350 (0.170–0.650) 0.002 * 0.342 (0.141–0.789) 0.014 * Atrial fibrillation 4.710 (1.780–16.270) 0.005 * – – fQRS 3.210 (1.440–8.150) 0.007 * – – Abnormal Q wave 2.910 (1.800–4.730) 0.000 * 2.390 (1.316–4.402) 0.004 * Low-voltage QRS 3.250 (1.620–7.120) 0.002 * – – Alcohol use 0.410 (0.230–0.710) 0.002 * – – Hyperglycemia 1.550 (0.920–2.660) 0.104 – – LVEF 0.860 (0.820–0.890) 0.000 * 0.864 (0.819–0.906) 0.000 * Heart rate 6.210 (2.390–21.240) 0.001 * 4.004 (1.235–16.350) 0.032 * QTc interval 5.680 (3.180–10.670) 0.000 * 2.252 (1.092–4.756) 0.030 * Tp-e interval 0.830 (0.470–1.440) 0.505 – – Age 60–75 1.960 (1.180–3.310) 0.010 * 2.607 (1.393–5.001) 0.003 * Age < 60 0.380 (0.240–0.620) 0.000 * – – LVEF: left ventricular ejection fraction; ∗P < 0.05. Feature variable selection by LASSO regression analysis Figure 2 shows the results of LASSO regression analysis of all the clinical and ECG variables; 10-fold cross-validation was used to determine the optimal regularization parameter λ for screening the variables with predictive value for acute heart failure. In the figure, panel A shows the LASSO regression paths, with each curve representing the change trend of the regression coefficient of a variable at different λ values. As λ gradually increases, the penalty imposed by the model on the variables is strengthened, and the coefficients of most variables gradually approach 0, indicating that their contribution to the model is small. Panel B shows the average binomial deviation corresponding to different λ values obtained via cross-validation, with red dots representing the mean and gray error bars representing the standard error (min=-3.9), as well as the simplest model parameter within a standard error range (λ.1se=-3.3). Finally, the model corresponding to λ.1se was selected to achieve the best generalizability and model simplification. At this λ value, there were 8 variables with nonzero coefficients, namely, sex, abnormal Q wave, left ventricular ejection fraction, heart rate, QTc interval, 60–75 years of age, fQRS and NT-proBNP, which were ultimately included in the subsequent multivariate regression model. These variables were considered key factors in predicting the occurrence of acute heart failure. Univariate and multivariate Logistic regression analyses Univariate logistic regression revealed that a variety of clinical and ECG parameters were strongly related to acute heart failure (AHF). Individuals with atrial fibrillation (OR = 4.710, 95% CI: 1.780–16.270, P = 0.005) exhibited a significantly increased risk of AHF, which was related to the presence of a fragmented QRS complex (fQRS, OR = 3.210, 95% CI: 1.440–8.150, P = 0.007), an abnormal Q wave (OR = 2.910, 95% CI: 1.800–4.730, P < 0.001), and low-voltage QRS (OR = 3.250, 95% CI: 1.620–7.120, P = 0.002). Moreover, a heart rate exceeding 100 beats per minute (OR = 6.210, 95% CI: 2.390–21.240, P = 0.001) also substantially increased the risk of AHF. Conversely, the risk of AHF was significantly lower in females (OR = 0.350, 95% CI: 0.170–0.650, P = 0.002) as well as those who consumed alcohol (OR = 0.410, 95% CI: 0.230–0.710, P = 0.002). The left ventricular ejection fraction (LVEF) was a constant variable, and each unit increase in the LVEF was related to a significantly lower risk of AHF (OR = 0.860, 95% CI: 0.820–0.890; P < 0.001). The aforementioned statistically significant variables were subsequently included in the multivariate logistic regression model for additional verification. The outcomes revealed that female sex (OR = 0.342, 95% CI: 0.141–0.789, P = 0.014) was still associated with a lower risk of AHF. An abnormal Q wave (OR = 2.390, 95% CI: 1.316–4.402, P = 0.004), a prolonged QTc interval (OR = 2.252, 95% CI: 1.092–4.756, P = 0.030), a heart rate exceeding 100 beats per minute (OR = 4.004, 95% CI: 1.235–16.350, P = 0.032), and age 60–75 years (OR = 2.607, 95% CI: 1.393–5.001, P = 0.003) were independent risk factors for AHF. The LVEF remained significant in the multivariate model (OR = 0.864, 95% CI: 0.819–0.906, P < 0.001), indicating that it had a stable protective effect against AHF. Construction and interpretation of the nomogram model Figure 3 shows the nomogram that was constructed to predict the risk of acute heart failure (AHF) in patients with acute myocardial infarction based on the results of the multivariate logistic regression model. Five categorical variables (male sex, an abnormal Q wave, heart rate > 100 beats/min, a prolonged QTc interval, and age between 60 and 75 years) and one continuous variable (left ventricular ejection fraction, LVEF) were included in the model. In the nomogram, each variable corresponds to a scoring axis, and different risk scores are assigned according to its clinical situation. By adding the score for each variable to obtain the total score, the probability of AHF can be predicted according to the total number of points. As shown in the figure, the LVEF has the greatest weight in the model, and the range of the LVEF score is wide, indicating that it contributes the most to the prediction of AHF risk. This is followed by the QTc interval and heart rate, sex, Q wave abnormality, and age range. The probability range of the nomogram is 0.1–0.9, which indicates good risk stratification. Thus, the nomogram could help clinicians quickly evaluate the risk of AHF in an individual and aid early identification and precise treatment of AHF. Model discrimination ability and calibration performance evaluation Figure 4 shows the performance of the established model in terms of discrimination ability and degree of calibration. As shown in Fig. 4 A, the area under the ROC curve (AUC) of the combined model was 0.8404, which was greater than that of any single indicator, including fQRS (AUC = 0.5525), QTc interval (AUC = 0.7403), QT dispersion (QTd, AUC = 0.5715) and left ventricular ejection fraction (LVEF, AUC = 0.7712), indicating that this multivariate model had better discrimination power in identifying acute heart failure. The red curve shows the LOWESS smooth fitting results, the blue points and error bars represent the actual incidence and 95% confidence intervals after grouping, and the gray dashed line represents the ideal calibration reference line. The results showed that the predicted probability of the model was highly consistent with the actual incidence, indicating good calibration of the model in effectively predicting the probability of disease in an individual and its strong potential for clinical application. Internal validation results of the model Figure 5 displays the validation of the constructed prognostic paradigm, which performed well in terms of both discrimination and calibration. As shown in Fig. 5 A, the AUC was 0.823, with a 95% confidence interval ranging from 0.776 to 0.871, indicating that the model could effectively identify acute heart failure. The blue shaded area indicates the 95% confidence interval, and the dashed horizontal blue line represents the benchmark error of sensitivity at shifting limits, with the model maintaining high sensitivity across most of the range. Panel B illustrates the calibration curve for the model generated via the bootstrap technique (bootstrapping iterations, B = 2000). The green line represents the fitting result of the predicted risk corrected by the observed rate, the red dotted line corresponds to the observed results, and the yellow dotted line represents the ideal calibration curve. Overall, the calibration curve was very close to the ideal calibration curve, indicating that the prognostic outcomes of the model were in good agreement with the real outcomes and that the model demonstrated high calibration accuracy. Clinical utility of the line chart according to decision curve analysis To further verify the practical value of the established model in clinical decision-making, we performed decision curve analysis, as shown in Fig. 6 . The horizontal axis shows the predicted risk threshold, and the vertical axis shows the net benefit. The results revealed that the model (orange curve) had a greater net benefit than "all intervention" (blue curve) and "no intervention" (black baseline) when the threshold probability was in the range of approximately 0.13 to 0.65, indicating its good clinical utility within this probability range. Especially within the common decision-making risk range of 0.1–0.5, the model was consistently superior to the traditional treatment strategy, indicating that it can serve as an auxiliary basis for clinical judgment and help achieve more reasonable patient stratification and management. Figure 7 Violin plots showing the correlations of the Killip grade of acute heart failure with electrocardiogram parameters Correlation analysis of the Killip grade and electrocardiogram parameters in patients with acute myocardial infarction complicated with acute heart failure Table 3 shows the correlations between the Killip grade and six key clinical and computerized ECG parameters. Spearman correlation analysis revealed that QTc interval (ρ = 0.392, P = 0.001), heart rate (ρ = 0.391, P = 0.001) and age (ρ = 0.265, P = 0.003) were significantly positively correlated with the Killip grade, whereas P-wave duration (ρ=-0.257, P = 0.004) and P-wave dispersion (ρ=-0.226, P = 0.012) and QRS duration were weakly but still statistically significantly correlated (ρ = 0.184, P = 0.043). Figure 7 further shows the distribution of each parameter in patients with disease of different Killip grades via violin plots. The results indicate that (A and B) the QTc interval and heart rate increased with increasing Killip grade, (C) age increased slightly at high Killip grades, and (D and E) the P-wave duration and P-wave dispersion decreased but (F) the QRS duration slightly differed but tended to increase with increasing Killip grade. The above results suggest that some ECG parameters have potential reference value in determining the severity of heart failure. Table 3 Spearman correlation analysis of the Killip grade and electrocardiogram parameters in patients with acute heart failure Variable Spearman coefficient P value QTc Interval 0.392 0.001 Heart Rate 0.391 0.001 Age 0.265 0.003 P-wave Duration -0.257 0.004 P-wave Dispersion -0.226 0.012 QRS Duration 0.184 0.043 Discussion In this study, we developed a risk prediction model for acute heart failure based on routine electrocardiogram (ECG) data in a primary population of AMI patients. QT prolongation, abnormal Q waves, tachycardia, a reduced LVEF, an advanced age and male sex were found to be independent risk factors for acute heart failure after AMI. The discrimination power (AUC) of the model was 0.84. Further analysis revealed that the Killip grade was significantly correlated with the QTc interval, heart rate, age and many other ECG parameters, whereas the P-wave duration was negatively correlated with dispersion, suggesting that these indicators not only have predictive value but also reflect the severity of heart failure, which strengthens the rationality and clinical value of the model variables. The main findings are discussed in the context of the literature, and the model is compared with other models based on different parameters to enhance the clinical application value and theoretical significance of the present model. Interpretation and comparison of the main findings QTc interval prolongation is considered the most prominent cardiac risk indicator on electrocardiography. The QTc interval reflects the full duration of ventricular repolarization. If the QTc interval is prolonged, the conduction velocity of the myocardial action potential is uneven, which is an important cause of heart failure and atrial fibrillation. Welten[ 2 ] et al. found that in patients with acute myocardial infarction, a QTc interval longer than 460 ms was significantly associated with left ventricular dysfunction and that the risk of heart failure was increased by approximately 11% for every 10 ms of QTc interval prolongation. In contrast, in the present study, a clear cutoff value was not set, but the QTc interval was identified as a key variable via the LASSO method to better predict clinical risk. Abnormal Q waves usually indicate a large area of extensive myocardial necrosis, which is closely related to both ventricular dyskinesis and cardiac remodeling. Houdmont[ 12 ] found that in patients with acute myocardial infarction in the anterior wall, the depth of the Q wave was positively correlated with the left ventricular end-diastolic volume, which independently predicted the likelihood of heart failure during the waiting period. The results of this study are consistent with these findings and suggest that Q-wave characteristics can also be used to roughly assess risk in primary care settings. The increase in heart rate during acute myocardial infarction may result from sympathetic tension or from compensatory mechanisms of cardiac output, and tachycardia may induce heart failure. Chan[ 13 ] reported that decreased heart rate variability and an elevated heart rate were strongly associated with in-hospital heart failure. In this study, the resting heart rate was used as a variable to show that the heart rate itself is a valuable indicator, even when heart rate variability monitoring devices are not used. Reduced left ventricular systolic function is directly related to heart failure, especially in patients experiencing acute myocardial infarction for the first time. Park[ 14 ] confirmed that an LVEF < 45% is the strongest predictor of major adverse cardiac events within a year. This study further highlights the irreplaceable role of cardiac ultrasound in predicting heart failure. Interesting findings regarding other parameters This study revealed several interesting findings, particularly regarding the interaction between cardiac electrophysiological features and personal characteristics such as sex and age. First, a fragmented QRS (fQRS), a marker of myocardial scarring and electrical conduction abnormalities, performs better in predicting acute heart failure after AMI than other variables and is more common in men than in women, suggesting that it may reflect sex-related differences[ 15 , 16 ] in structural remodeling. Moreover, prolongation of the Tp-e interval and QT dispersion indicate increased heterogeneity in ventricular repolarization, and these repolarization parameters have greater predictive sensitivity in female patients, suggesting a difference in electrophysiological responses by sex.[ 17 ] In addition, the predictive performance of electrophysiological parameters increased with age, especially in elderly patients, indicating that age is not only an independent risk factor but also may change the risk assessment mode[ 18 ] by affecting the expression of electrophysiological parameters. Notably, the combination of multiple ECG parameters was superior to any single variable in the prediction of heart failure, which strengthens the potential of multidimensional electrical signal analysis. However, some traditional clinical parameters (such as atrial fibrillation or simple heart rate) have limited predictive value, further highlighting the advantages[ 19 ] of electrophysiological characteristics in individualized risk assessment. P-wave dispersion in patients was not directly explored in this study, but related studies suggest that it may have predictive power. Yang[ 20 ] reported that P-wave dispersion > 40 ms may predict heart failure after acute myocardial infarction, which may be related to heart failure caused by atrial fibrillation. This finding suggests that changes in P-wave dispersion could be incorporated in future models to improve sensitivity. This study did not include parameters related to ST-segment changes, but Kazemi[ 21 ] study revealed that an ST shift in lead aVR can predict left ventricular dysfunction in patients with multivesicular disease. Siren[ 22 ] et al. found that incomplete ST-segment resolution after PCI was associated with a higher rate of new or worsening heart failure, suggesting ST resolution is a valuable predictor for post-infarction outcomes and that continuous ECG monitoring may enhance early risk stratification. The Tp-e interval was not evaluated in this study, but Özbek[ 23 ] reported that its prolongation predicts increased dysregulation of compensatory mechanisms and is an important indicator of heart failure and malignant arrhythmias after acute myocardial infarction. Galloway[ 24 ] et al. established an algorithm for identifying T-wave alternation based on electrocardiogram (ECG) images and deep learning models and suggested that T-wave variation could predict heart failure in the absence of significant structural changes. The present study also revealed that the Killip grade was significantly correlated with ECG parameters such as the QTc interval, heart rate, and P-wave characteristics, indicating its potential value in the assessment of AHF. With increasing Killip grade, the QTc interval was also prolonged, which is consistent with El Amrawy et al.[ 25 ] observation that prolongation of the QTc interval significantly predicts a higher Killip grade and a higher risk of in-hospital mortality in STEMI patients. Heart rate was also positively correlated with the Killip grade, and Jensen[ 26 ] et al. confirmed that a higher heart rate is an independent predictor of a higher Killip grade and poor outcomes. In addition, with increasing Killip grade, the P-wave duration decreased and conductivity decreased, which may reflect atrioventricular conduction remodeling. Bulluck[ 27 ] et al. proposed that the combination of electrophysiological and imaging parameters can enhance the prediction power of the Killip grade, which indirectly supports this finding. Although the correlation of the Killip grade with the QRS duration was weak in this study, Gabaldon‒Perezi[ 28 ] et al. noted that flow disturbances in patients with Killip grade I-II disease may also cause serious complications, suggesting that electrocardiographic abnormalities also have predictive value in patients with low Killip grades. Limitations and future prospects Although this study explored the feasibility of using simple ECG parameters to identify the risk of heart failure in patients with myocardial infarction, there are numerous limitations. The sample size was limited, and the patients were from a single health care institution, which may have resulted in selection bias. The model does not consider indicators of dynamic ECG changes such as T-wave directionality, the T-wave interapical interval, or waveform analysis, which limits its generalizability. In addition, multicenter validation was not carried out, and the external reliability of the model remains to be tested. In the future, the number of samples needs to be increased, and more abnormalities should be automatically identified via artificial intelligence-based image recognition technology. Moreover, continuous indicators such as HRV should be included to improve the prediction accuracy and construct an early warning system. Conclusion In this study, a simple and practical risk prediction model for AMI complicated with AHF based on routine ECG and clinical characteristics was established. In the internal validation, the model performed well in terms of discrimination and calibration, indicating suitability for the early identification of high-risk patients in primary care settings to aid treatment. Notably, several ECG parameters, such as the Killip grade, QTc interval, heart rate and P-wave dispersion, were significantly correlated with the severity of AHF, indicating that these parameters not only have predictive value but can also assist in assessing the severity of AHF. These findings provide a new perspective for the clinical use of ECG findings beyond risk assessment and further expand the explanatory power and clinical practicability of the model. The stability and generalizability of the model could be further verified using external datasets from more hospitals in the future. Declarations Ethical approval and consent to participate This study was approved by the Medical Ethics Committee of Xuancheng People's Hospital (approval number :2025-byky001-01) and fully complied with the ethical principles of the Declaration of Helsinki and subsequent versions. All patients who participated in the study signed an informed consent form at the time of visiting the hospital, and agreed to the hospital to use their medical data for anonymous scientific research analysis. This was a retrospective observational study, and no additional interventions were performed. The study data were desensitized to protect patient privacy. Conflict of Interest Statement The authors declare no competing interests. Funding This study was funded by the universal-level Key Project of Natural Science of Bengbu Medical University in Anhui Province (project number: 2024byzd192). The project name was "Construction of prognosis prediction model of acute myocardial Infarction based on electrocardiogram multimodal data". The research was conducted by Gao Guoliang of Xuancheng People's Hospital in Anhui Province. The funding supported all stages of work, including clinical data acquisition, ECG feature extraction, multivariate statistical model construction and model visualization development. The research team independently completed the whole process from project design to data analysis to paper writing, and the funders did not participate in any part of the research implementation. This study conforms to the scientific research ethics and statistical norms. The conclusions drawn are authentic and reliable, and also have certain clinical value. Author Contribution X.G. was responsible for project establishment, project management, data interpretation, and paper writing and revision. G.G. was responsible for the collection of clinical data and the drawing of statistical analysis charts. G.Y. and A.X. made ECG judgment, searched literature and wrote the first draft of the manuscript. H.H. and X.Z. established the prediction model and result visualization, and participated in the interpretation of the results and revision of the paper. All the authors have read and agreed to the final draft and are jointly responsible for the authenticity and completeness of the study results. Acknowledgement Thanks to the kindness and care of all the teachers, I have received help and support from many aspects in this research work. Special thanks to the team of Cardiology Department of Xuancheng People's Hospital, Anhui Province, who provided great assistance in sample screening and data collection; We would also like to thank Xuancheng People's Hospital of Anhui Province and Bengbu Medical University for funding of the research project; At the same time, we would also like to thank the patients and their families who participated in this study, who provided valuable data and were an important basis for this study. Data Availability The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request. References Allescher, J. et al. QRS fragmentation does not predict mortality in survivors of acute myocardial infarction. Clin. Cardiol. 47 , e24218 (2024). Welten, S. et al. Prolongation of the QTc interval is associated with an increased risk of cardiovascular diseases: The Hoorn study. J. Electrocardiol. 80 , 133–138 (2023). Xu, Y. et al. Predicting efficacy of combined assessment with fragmented QRS and severely depressed heart rate variability on outcome of patients with acute myocardial infarction. Heart Vessels . 37 , 239–249 (2022). Türkmen, S., Bozkurt, M., Hoşoğlu, Y. & Göl, M. Significance of fragmented QRS and predictors of outcome in ST-elevation myocardial infarction. J. Res. Med. Sci. 29 , 23 (2024). Han, X. et al. Prognostic significance of QRS distortion and frontal QRS-T angle in patients with ST-elevation myocardial infarction. Int. J. Cardiol. 345 , 1–6 (2021). Zorlu, Ç., Açıkel, B., Ömür, S. E. & Frontal Plane QRS - T Angle Is a Predictor of Ventricular Arrhythmia in Heart Failure With Preserved Ejection Fraction. Ann. Noninvasive Electrocardiol. 30 , e70062 (2025). Martínez-Sellés, M. & Marina-Breysse, M. Current and Future Use of Artificial Intelligence in Electrocardiography. Journal Cardiovasc. Dev. disease 10 , (2023). Byrne, R. A. et al. 2023 ESC Guidelines for the management of acute coronary syndromes. Eur. Heart J. 44 , 3720–3826 (2023). McDonagh, T. A. et al. Focused Update of the 2021 ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure. Eur Heart J . 44, 3627–3639 (2023). (2023). Zeppenfeld, K. et al. ESC Guidelines for the management of patients with ventricular arrhythmias and the prevention of sudden cardiac death. Eur Heart J . 43, 3997–4126 (2022). (2022). Villa, A. et al. A machine learning algorithm for electrocardiographic fQRS quantification validated on multi-center data. Sci. Rep. 12 , 6783 (2022). Houdmont, M. et al. Pathological Q waves at presentation of anterior ST segment elevation myocardial infarction predict heart failure: a Southeast Asian perspective. Coron. Artery Dis. 36 , 378–383 (2025). Chan, J. S. K. et al. Fragmented QRS Is Independently Predictive of Long-Term Adverse Clinical Outcomes in Asian Patients Hospitalized for Heart Failure: A Retrospective Cohort Study. Front. Cardiovasc. Med. 8 , 738417 (2021). Park, C. S. et al. Left Ventricular Ejection Fraction 1 Year After Acute Myocardial Infarction Identifies the Benefits of the Long-Term Use of β-Blockers: Analysis of Data From the KAMIR-NIH Registry. Circ. Cardiovasc. Interv . 14 , e010159 (2021). Luo, G. et al. The Predictive Value of Fragmented QRS for Cardiovascular Events in Acute Myocardial Infarction: A Systematic Review and Meta-Analysis. Front. Physiol. 11 , 1027 (2020). Terho, H. K. et al. Prevalence and prognostic significance of fragmented QRS complex in middle-aged subjects with and without clinical or electrocardiographic evidence of cardiac disease. Am. J. Cardiol. 114 , 141–147 (2014). Iwahashi, N. et al. Global Strain Measured by Three-Dimensional Speckle Tracking Echocardiography Is a Useful Predictor for 10-Year Prognosis After a First ST-Elevation Acute Myocardial Infarction. Circ. J. 85 , 1735–1743 (2021). Hempel, P. et al. Explainable AI associates ECG aging effects with increased cardiovascular risk in a longitudinal population study. NPJ Digit. Med. 8 , 25 (2025). Lee, M. S. et al. Artificial intelligence applied to electrocardiogram to rule out acute myocardial infarction: the ROMIAE multicentre study. Eur. Heart J. 46 , 1917–1929 (2025). Yang, N. et al. The Role of P-Wave Variables in Enhancing Prediction of New-Onset Atrial Fibrillation in Patients With Acute Myocardial Infarction. Ann. Noninvasive Electrocardiol. 30 , e70041 (2025). Kazemi, E. et al. The prognostic effect of ST-elevation in lead aVR on coronary artery disease, and outcome in acute coronary syndrome patients: a systematic review and meta-analysis. Eur. J. Med. Res. 27 , 302 (2022). Sirén, M. et al. The prognostic significance of single-lead ST-segment resolution in ST-segment elevation myocardial infarction patients treated with primary PCI - A substudy of the randomized TOTAL trial. Am. Heart J. 269 , 149–157 (2024). Özbek, S. C. & Sökmen, E. Usefulness of Tp-Te interval and Tp-Te/QT ratio in the prediction of ventricular arrhythmias and mortality in acute STEMI patients undergoing fibrinolytic therapy. J. Electrocardiol. 56 , 100–105 (2019). Galloway, C. D. et al. Development and Validation of a Deep-Learning Model to Screen for Hyperkalemia From the Electrocardiogram. JAMA Cardiol. 4 , 428–436 (2019). El Amrawy, A. M. et al. QTc interval prolongation impact on in-hospital mortality in acute coronary syndromes patients using artificial intelligence and machine learning. Egypt. Heart J. 76 , 149 (2024). Jensen, M. T. et al. Heart rate at admission is a predictor of in-hospital mortality in patients with acute coronary syndromes: Results from 58 European hospitals: The European Hospital Benchmarking by Outcomes in acute coronary syndrome Processes study. Eur. Heart J. Acute Cardiovasc. Care . 7 , 149–157 (2018). Bulluck, H. et al. Independent Predictors of Cardiac Mortality and Hospitalization for Heart Failure in a Multi-Ethnic Asian ST-segment Elevation Myocardial Infarction Population Treated by Primary Percutaneous Coronary Intervention. Sci. Rep. 9 , 10072 (2019). Gabaldón-Pérez, A. et al. Prognostic value of cardiac magnetic resonance early after ST-segment elevation myocardial infarction in older patients. Age Ageing 51 , (2022). Additional Declarations No competing interests reported. Supplementary Files GraphicalAbstract.tif Cite Share Download PDF Status: Published Journal Publication published 18 Feb, 2026 Read the published version in Scientific Reports → Version 1 posted Editorial decision: Revision requested 14 Oct, 2025 Reviews received at journal 18 Sep, 2025 Reviews received at journal 17 Sep, 2025 Reviewers agreed at journal 08 Sep, 2025 Reviewers agreed at journal 07 Sep, 2025 Reviews received at journal 15 Aug, 2025 Reviewers agreed at journal 13 Aug, 2025 Reviewers agreed at journal 10 Aug, 2025 Reviewers invited by journal 08 Aug, 2025 Editor invited by journal 07 Aug, 2025 Editor assigned by journal 05 Aug, 2025 Submission checks completed at journal 05 Aug, 2025 First submitted to journal 04 Aug, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7286686","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":498227560,"identity":"43f599c5-b6d1-43e2-bdcf-86fd71d84bab","order_by":0,"name":"Xuefen Guo","email":"","orcid":"","institution":"Wannan Medical College","correspondingAuthor":false,"prefix":"","firstName":"Xuefen","middleName":"","lastName":"Guo","suffix":""},{"id":498227561,"identity":"39cabc17-4660-4bf8-8f51-b1a60946444d","order_by":1,"name":"Guoping Yan","email":"","orcid":"","institution":"Wannan Medical College","correspondingAuthor":false,"prefix":"","firstName":"Guoping","middleName":"","lastName":"Yan","suffix":""},{"id":498227562,"identity":"7cfc48c4-4e92-49b5-8cb5-ad7ab31b4d3b","order_by":2,"name":"Hua He","email":"","orcid":"","institution":"Wannan Medical College","correspondingAuthor":false,"prefix":"","firstName":"Hua","middleName":"","lastName":"He","suffix":""},{"id":498227563,"identity":"c70164f6-cc0e-43b1-b6bf-7f565c099b2f","order_by":3,"name":"Aihui Xu","email":"","orcid":"","institution":"Langxi Hospital Affiliated to the school of clinical medicine of Hangzhou Normal University","correspondingAuthor":false,"prefix":"","firstName":"Aihui","middleName":"","lastName":"Xu","suffix":""},{"id":498227564,"identity":"cc070612-ee3b-4d37-84af-71ca13e5c988","order_by":4,"name":"Xueting Zhan","email":"","orcid":"","institution":"Wannan Medical College","correspondingAuthor":false,"prefix":"","firstName":"Xueting","middleName":"","lastName":"Zhan","suffix":""},{"id":498227565,"identity":"557a74ed-3e45-44fe-b83c-6c6e5d119bda","order_by":5,"name":"Guoliang Gao","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAv0lEQVRIiWNgGAWjYDACZhBRYAMkeEjSYpAG02JArD6DwyRoMTjOe/jlF4Pzif3sZw8wF1T8IcZ8vjRrGYPbiTN78hKYZ5whwhbJZh4zYwmglg0HcgyYeduI13Iucf/5N0At/4jQws/MY/zwg8GBxA0SIFsaiNLCl8bMYJBsPOPGG4PDPMeMCWth4z97+OOPCjvZ/v4cw8c8NXKEtQCjg00aGCOODUDmAWLUg7Qwf/zBwGBPpOpRMApGwSgYiQAAiyE1Ke4rrswAAAAASUVORK5CYII=","orcid":"","institution":"Wannan Medical College","correspondingAuthor":true,"prefix":"","firstName":"Guoliang","middleName":"","lastName":"Gao","suffix":""}],"badges":[],"createdAt":"2025-08-04 04:08:16","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7286686/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7286686/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1038/s41598-026-40600-9","type":"published","date":"2026-02-18T15:57:06+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":89066472,"identity":"04b4ff9b-5985-4689-8b23-8f3007577664","added_by":"auto","created_at":"2025-08-14 10:42:41","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":4507599,"visible":true,"origin":"","legend":"\u003cp\u003eFlowchart of the inclusion criteria for acute myocardial infarction patients with or without acute heart failure.\u003c/p\u003e","description":"","filename":"fig1.png","url":"https://assets-eu.researchsquare.com/files/rs-7286686/v1/cea6e65db8d1c4623ec75e99.png"},{"id":89066523,"identity":"9dcc98ac-03a1-4d18-a8aa-b3710f14218d","added_by":"auto","created_at":"2025-08-14 10:42:43","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":3623954,"visible":true,"origin":"","legend":"\u003cp\u003eLASSO regression results for 20 risk variables.\u003c/p\u003e\n\u003cp\u003eA: Coefficient path diagram of risk variables. B: Cross-validation LASSO regression curve.\u003c/p\u003e\n\u003cp\u003ePanel A shows the regression coefficient paths for each variable according to LASSO regression for different log(λ) values. As the value of λ decreased, the variable gradually entered the model. The blue dashed line indicates the minimum value of λ (log(λ) = -3.91), which corresponds to the least squares error; the red dotted line is the λ value (log(λ) = -3.27), which represents the simplest model. Panel B shows the 10-fold cross-validation curve, with red dots representing the average error at each λ value and gray error bars representing ±1 standard error. Finally, the λ values within 1 standard error were selected to construct the model, and a total of 8 variables with coefficients other than 0 were screened for modeling.\u003c/p\u003e","description":"","filename":"fig2.png","url":"https://assets-eu.researchsquare.com/files/rs-7286686/v1/29f9cae22fff9a99d315bfe3.png"},{"id":89066081,"identity":"c96eb413-abdb-454a-9f3d-3543981073ac","added_by":"auto","created_at":"2025-08-14 10:42:33","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":252634,"visible":true,"origin":"","legend":"\u003cp\u003eNomogram model for predicting the occurrence of acute myocardial infarction accompanied by acute heart failure.\u003c/p\u003e","description":"","filename":"fig3.png","url":"https://assets-eu.researchsquare.com/files/rs-7286686/v1/cd0cc7fa20762bc1eec38d91.png"},{"id":89066728,"identity":"d0622369-a163-49db-82fd-22d38b93200e","added_by":"auto","created_at":"2025-08-14 10:42:49","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":4532670,"visible":true,"origin":"","legend":"\u003cp\u003eROC curve and calibration curve of the line graph model.\u003c/p\u003e\n\u003cp\u003eA: ROC curves for the line chart and other independent predictors. B: Calibration curves before internal validation of the line chart.\u003c/p\u003e","description":"","filename":"fig4.png","url":"https://assets-eu.researchsquare.com/files/rs-7286686/v1/4d9b71a965ca36fe94570ea5.png"},{"id":89066424,"identity":"e71b0fdf-cb46-4982-a0f7-ff3096432b42","added_by":"auto","created_at":"2025-08-14 10:42:40","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":3807805,"visible":true,"origin":"","legend":"\u003cp\u003eROC curve and calibration curve of the internal validation-based line chart\u003c/p\u003e\n\u003cp\u003eA: ROC curve after internal validation of the bar chart. B: Calibration curve after internal validation of the bar chart.\u003c/p\u003e","description":"","filename":"fig5.png","url":"https://assets-eu.researchsquare.com/files/rs-7286686/v1/aa2fe519a2561ff97dcad48b.png"},{"id":89066548,"identity":"82c4064b-289b-4988-bf66-deb9185d9697","added_by":"auto","created_at":"2025-08-14 10:42:43","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":217652,"visible":true,"origin":"","legend":"\u003cp\u003eClinical decision analysis curve for the line chart.\u003c/p\u003e","description":"","filename":"fig6.png","url":"https://assets-eu.researchsquare.com/files/rs-7286686/v1/f0318cf7b607f3e82e3b2f23.png"},{"id":89066991,"identity":"bc2081d7-4dc3-4cbe-8d5e-b4090ebf4b53","added_by":"auto","created_at":"2025-08-14 10:43:00","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":644815,"visible":true,"origin":"","legend":"\u003cp\u003eDistributions of six key clinical and ECG parameters in patients with different Killip grades (I-IV). (A) QTc interval. (B) Heart rate. (C) Age. (D) P-wave duration. (E) P-wave dispersion. (F) QRS duration. The distribution characteristics of each variable in patients with disease of different Killip grades are presented in the form of violin plots combined with boxplots. The P value represents the monotonic correlation between the variable and Killip grade, and the significance level is expressed as follows: *P\u0026lt; 0.05, **P\u0026lt; 0.01, and ***P\u0026lt; 0.001.\u003c/p\u003e","description":"","filename":"fig7.png","url":"https://assets-eu.researchsquare.com/files/rs-7286686/v1/465868ee0c48583c4cf73611.png"},{"id":103251022,"identity":"38d0681a-d724-4b69-85e5-e95eda397237","added_by":"auto","created_at":"2026-02-23 16:01:13","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":16547997,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7286686/v1/a2718554-ac1d-4a7e-a286-cb2ef62b87da.pdf"},{"id":89066758,"identity":"90b7ab5d-9ee3-4206-9c55-dc6283a81e2f","added_by":"auto","created_at":"2025-08-14 10:42:51","extension":"tif","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":717388,"visible":true,"origin":"","legend":"","description":"","filename":"GraphicalAbstract.tif","url":"https://assets-eu.researchsquare.com/files/rs-7286686/v1/8627a0bd71703ad1f969ede4.tif"}],"financialInterests":"No competing interests reported.","formattedTitle":"Construction of a predictive model for acute myocardial infarction combined with acute heart failure based on electrocardiogram parameters and exploration of the correlations of these parameters with the Killip grade","fulltext":[{"header":"Introduction","content":"\u003cp\u003eAcute myocardial infarction, otherwise known as a heart attack, continues to be the leading cause of mortality worldwide. The development of acute heart failure in the wake of acute myocardial infarction has substantial implications for a patient's survival and likelihood of rehospitalization, despite the fact that mortality specifically due to myocardial infarction following medical interventions and pharmaceutical treatments is decreasing. Statistics indicate that approximately one-fifth to one-third of all individuals admitted to a hospital for myocardial infarction also experience acute heart failure, which considerably prolongs the duration of hospitalization and increases the risk of readmission and death, especially in primary care facilities.[\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]\u003c/p\u003e\u003cp\u003eAs a routine examination method, electrocardiogram (ECG) plays an important role in reflecting the severity of myocardial injury and predicting late adverse outcomes. Recent studies have shown that certain ECG parameters are positively associated with the occurrence of acute heart failure. For example, a prolonged QTc interval not only reflects electrical conduction disorders but also significantly increases[\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e] the risk of heart failure after myocardial infarction. Fragmented QRS complex (fQRS) has shown good predictive efficacy[\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e] in myocardial infarction patients both with and without ST-segment elevation, and this result[\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e] has been confirmed by other scholars. In addition, QRS-T angle enlargement is often predictive of myocardial structural remodeling and instability of electrical activity, which often cause heart failure or death.[\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e] Zorlu[\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e] et al. reported that the QRS-T angle is closely related to rehospitalization for chronic heart failure. Turkmen[\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e] also confirmed the important role of ECG parameters in predicting arrhythmia and left ventricular remodeling after myocardial infarction by analyzing QT dispersion and the Tp-e interval. Although Martínez-Sellés[\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e] et al. attempted to use artificial intelligence models to integrate multiple ECG features for risk prediction, several problems, such as computing platform dependence, dataset resources, and technical thresholds, need to be solved before the clinical application of such models.\u003c/p\u003e\u003cp\u003eTherefore, multiple traditional 12-lead ECG parameters (e.g., the QTc interval, fQRS, the QRS-T angle and the Tp-e interval) were combined according to practical clinical requirements, and common biomarkers (e.g., general clinical data and the left ventricular ejection fraction) were selected via LASSO regression. A simple and clear structural nomogram model was then constructed to both achieve the early identification of acute heart failure and quantify the risk of acute heart failure in patients with myocardial infarction. The purpose of this study was to provide a set of practical and feasible risk assessment tools for primary hospitals and to promote the accurate hierarchical management of cardiovascular diseases.\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003e\u003cb\u003eStudy Subjects\u003c/b\u003e\u003c/p\u003e\u003cp\u003ePatients receiving standard treatment for acute myocardial infarction at our hospital between October 2022 and March 2025 were included in this single-center, retrospective, observational study (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). The inclusion criteria were as follows: 1) AMI according to the diagnostic criteria from the European Heart Association's 2023 guidelines[\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]; 2) age of 18 years or older; and 3) complete clinical data from examinations at our hospital. The exclusion criteria were as follows: 1) malignant tumor, severe infection, autoimmune disease, or end-stage liver or kidney dysfunction; 2) death due to noncardiac causes during hospitalization; and 3) incomplete or missing key data. Acute heart failure was diagnosed according to the 2023 Focus Update of the European Society of Cardiology Guidelines[\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e] for Heart Failure from 2021. The Killip grade was used to assess the severity of acute heart failure in AMI patients; acute heart failure was classified as grade I, II, III, or IV based on the criteria first proposed by Killip and Kimball in 1967. In this study, the Killip grade was used to assess the severity of acute heart failure in patients with acute heart failure upon admission according to clinical manifestations.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003cb\u003eClinical data collection\u003c/b\u003e\u003c/p\u003e\u003cp\u003eGeneral demographic data (sex, age), data related to the history of underlying diseases (hypertension, diabetes, hyperlipidemia, arrhythmia, heart failure, etc.), test results obtained upon admission (troponin levels, liver and kidney function, etc.) and auxiliary examination reports (electrocardiogram, echocardiography, etc.) were collected. Related indicators were uniformly tested at our hospital. All patients underwent standard 12-lead electrocardiogram within 24 hours after admission. The equipment and procedures used were specified by the hospital. Two physician-qualified electrocardiologists independently judged the results, and discrepancies were reviewed by a third expert. The 10 extracted ECG parameters included the following: fragmented QRS complex (representing a myocardial scar or conduction problem), abnormal Q wave (Q-wave duration ≥ 40 ms and amplitude ≥ 0.1 mV fror two adjacent leads), low-voltage QRS complex (QRS amplitude \u0026lt; 5 mm for all limb leads), a prolonged P-wave duration (maximum P-wave duration ≥ 120) ms), P-wave dispersion (maximum and minimum P-wave duration difference \u0026gt; 40 ms), QRS complex prolongation (≥ 120 ms), QTc interval prolongation (QTc interval \u0026gt; 450 ms in men or \u0026gt; 460 ms in women, corrected by Bazett's formula), QT dispersion (maximum and minimum QT interval difference \u0026gt; 60) ms), T peak-to-end interval (Tp-e interval, reflecting the waveform difference during ventricular repolarization), and heart rate (\u0026gt; 100 bpm or ≤ 100 bpm). All the parameters were converted to binary variables according to clinical definitions for statistical analysis and to assess their ability to predict the risk of acute myocardial infarction complicated by acute heart failure.\u003c/p\u003e\u003cp\u003e To standardize variable definitions and improve the clinical interpretability of the model, the latest international guidelines and high-level research were referenced, and a common clinical classification method for some continuous variables was adopted. For ECG parameters, QTc interval prolongation was defined as a QTc interval \u0026gt; 460 ms (\u0026gt; 450 ms in men and \u0026gt; 460 ms in women) according to the 2022 European Heart Rhythm Society guidelines[\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e] for the management of arrhythmias. A QRS duration \u0026gt; 120 msec was defined as an intraventricular conduction delay, as indicated by the 2023 Focus Update of the ESC guidelines[\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e] for the diagnosis and treatment of heart failure. Resting tachycardia was defined as a heart rate \u0026gt; 100 beats/min; a P wave duration \u0026gt; 120 ms was used as an indicator of left atrial enlargement and AF risk. QT dispersion \u0026gt; 50 ms and a Tp-e interval \u0026gt; 90 ms indicated increased dispersion of ventricular repolarization and an increased risk of arrhythmia, respectively. Fragmented QRS (fQRS) and low-voltage QRS were identified by electrocardiogram (ECG), and relevant definitions were provided by Villa.[\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e] In accordance with the 2023 Focus Update of the ESC guidelines[\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e] for the diagnosis and treatment of heart failure, patients were divided into three age groups, namely, \u0026lt; 60 years, 60–74 years and ≥ 75 years, to evaluate differences in the risk of acute heart failure across different age groups.\u003c/p\u003e\u003cp\u003e\u003cb\u003eObservation Endpoints\u003c/b\u003e\u003c/p\u003e\u003cp\u003eThe primary outcome was the occurrence of acute heart failure and the Killip grade during AMI. According to the 2023 Focused Update of the ESC guidelines, acute heart failure was defined as dyspnea[\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e] with pulmonary rales at rest, pulmonary congestion or edema on imaging, or a significant increase in the NT-proBNP level requiring intervention (e.g., the use of diuretics, vasodilators or inotropic drugs). All endpoint events were independently judged by two senior cardiologists, and if the results were inconsistent, a third physician resolved disputes.\u003c/p\u003e\u003cp\u003e\u003cb\u003eModel construction and evaluation\u003c/b\u003e\u003c/p\u003e\u003cp\u003eThe aim of our research was to develop a predictive model for assessing acute heart failure risk in patients recently diagnosed with acute myocardial infarction. By analyzing clinical data from a large cohort of myocardial infarction patients, we utilized LASSO regression to identify key variables associated with subsequent heart failure. These factors were then incorporated into a logistic regression model for risk calculation. Model performance was assessed using receiver operating characteristic curves and calculating the area under the curve. We also performed bootstrapping with 2000 iterations to generate a calibration curve and evaluate model fit. Additionally, decision curve analysis was used to examine whether the model provided worthwhile clinical benefit across a range of risk thresholds. A diverse collection of sentences was used to aid model interpretation while maintaining the overall word count and conceptual similarity to the original text.\u003c/p\u003e\u003cp\u003e\u003cb\u003eStatistical Methods\u003c/b\u003e\u003c/p\u003e\u003cp\u003eA variety of statistical software and packages were utilized for the diverse data analysis. LASSO regression was carried out using the glmnet package to screen for key predictors, whereas the rms package was used to construct a multivariate logistic regression model and generate a calibration curve to assess model fit. The pROC package was used to generate the ROC curve and calculate the AUC for determining predictive ability. Decision curve analysis was performed to evaluate the net clinical benefit under different thresholds using the rmda package. Spearman rank correlation analysis was used to evaluate the correlation between the Killip grade (I-IV) as the rank variable and each parameter, and correlation coefficients and P values were calculated. For meaningfully linked factors, violin plots were further constructed to visualized their distribution patterns across the different Killip grades. Continuous variables are expressed as the means ± standard deviations or medians [quartiles] based on the data distribution. Independent samples t tests or Mann‒Whitney U tests were used to compare the two groups. Categorical variables are presented as frequencies and proportions and were compared via chi-square tests or Fisher's exact tests. P \u0026lt; 0.05 was considered to indicate statistical significance.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003e\u003cb\u003eComparison of baseline characteristics\u003c/b\u003e\u003c/p\u003e\u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e displays the clinical and electrocardiographic data of 301 registered individuals with AMI grouped according to the presence or absence of AHF. Of the 301 registered individuals, 185 (61.5%) were in the AHF group, and 116 (38.5%) were within the non-AHF (NAHF) group. A chi-square test or Mann‒Whitney U test was used to compare variables between the 2 groups according to the type of variable. The results showed notable variations in a variety of clinical and ECG parameters. There were significantly more women in the AHF group (26.9%) than in the NAHF group (11.3%). Atrial fibrillation (14.6%) and fQRS (17.2%) were much more common in the AHF group than in the NAHF group. Moreover, abnormal Q waves, a history of smoking, low-voltage QRS and a prolonged QTc interval (\u0026gt;\u0026thinsp;460 ms) were more common in the AHF group than in the NAHF group (all P\u0026thinsp;\u0026lt;\u0026thinsp;0.05) and were closely related to AHF. Critically, the median LVEF in the AHF group, which was 47.0% [38.5, 51.7], was considerably lower than that in the NAHF group, which was 58.5% [56.2, 63.0] (P\u0026thinsp;=\u0026thinsp;0.05).\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eBaseline characteristics of patients with acute myocardial infarction complicated with acute heart failure\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"5\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eVariable\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eTotal (n\u0026thinsp;=\u0026thinsp;301)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eAHF (n\u0026thinsp;=\u0026thinsp;185)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eNAHF (n\u0026thinsp;=\u0026thinsp;116)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003eP value\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSex\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.002 *\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMale\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e238 (79.1%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e136 (73.1%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e102 (88.7%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eFemale\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e63 (20.9%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e50 (26.9%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e13 (11.3%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAtrial Fibrillation\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.004 *\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eYes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e31 (10.3%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e27 (14.5%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e4 (3.5%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNo\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e270 (89.7%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e159 (85.5%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e111 (96.5%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePresence of fragmented QRS (fQRS)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.009 *\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eYes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e39 (13.0%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e32 (17.2%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e7 (6.1%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNo\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e262 (87.0%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e154 (82.8%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e108 (93.9%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAbnormal Q Wave\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.000 *\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eYes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e161 (53.5%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e118 (63.4%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e43 (37.4%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNo\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e140 (46.5%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e68 (36.6%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e72 (62.6%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLow-voltage QRS\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.002 *\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eYes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e54 (17.9%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e44 (23.7%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e10 (8.7%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNo\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e247 (82.1%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e142 (76.3%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e105 (91.3%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSmoking\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.002 *\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eYes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e145 (48.2%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e76 (40.9%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e69 (60.0%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNo\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e156 (51.8%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e110 (59.1%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e46 (40.0%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAlcohol use\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.002 *\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eYes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e63 (20.9%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e28 (15.1%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e35 (30.4%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNo\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e238 (79.1%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e158 (84.9%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e80 (69.6%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eHypertension\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.817\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eYes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e187 (62.1%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e117 (62.9%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e70 (60.9%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNo\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e114 (37.9%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e69 (37.1%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e45 (39.1%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eHyperglycemia\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.133\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eYes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e87 (28.9%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e60 (32.3%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e27 (23.5%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNo\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e214 (71.1%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e126 (67.7%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e88 (76.5%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eHyperlipidemia\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.401\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eYes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e42 (14.0%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e23 (12.4%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e19 (16.5%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNo\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e259 (86.0%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e163 (87.6%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e96 (83.5%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eP-wave duration\u0026thinsp;\u0026gt;\u0026thinsp;120 ms\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.908\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eYes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e52 (17.3%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e33 (17.7%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e19 (16.5%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNo\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e249 (82.7%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e153 (82.3%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e96 (83.5%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eProlonged P-wave dispersion\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.067\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eYes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e19 (6.3%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e16 (8.6%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e3 (2.6%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNo\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e282 (93.7%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e170 (91.4%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e112 (97.4%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eQRS-wave duration\u0026thinsp;\u0026gt;\u0026thinsp;120 ms\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.016 *\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eYes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e27 (9.0%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e23 (12.4%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e4 (3.5%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNo\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e274 (91.0%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e163 (87.6%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e111 (96.5%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eQTc interval\u0026thinsp;\u0026gt;\u0026thinsp;460 ms\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.000 *\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eYes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e105 (34.9%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e89 (47.8%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e16 (13.9%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNo\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e196 (65.1%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e97 (52.2%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e99 (86.1%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eQT dispersion (QTd)\u0026thinsp;\u0026gt;\u0026thinsp;50 ms\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.337\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eYes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e111 (36.9%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e73 (39.2%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e38 (33.0%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNo\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e190 (63.1%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e113 (60.8%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e77 (67.0%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTp-e interval\u0026thinsp;\u0026gt;\u0026thinsp;90 ms\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.599\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eYes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e232 (77.1%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e141 (75.8%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e91 (79.1%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNo\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e69 (22.9%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e45 (24.2%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e24 (20.9%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eHeart rate\u0026thinsp;\u0026gt;\u0026thinsp;100 bpm\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.000 *\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eYes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e38 (12.6%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e34 (18.3%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e4 (3.5%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNo\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e263 (87.4%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e152 (81.7%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e111 (96.5%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLeft ventricular ejection fraction (LVEF), median (Q1, Q3)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e55 (49, 59)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e50 (47, 57)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e58 (55, 62)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.000 *\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAge group\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.000 *\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u0026lt;60\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e115 (38.2%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e55 (29.6%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e60 (52.2%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e60\u0026ndash;75\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e103 (34.2%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e74 (39.8%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e29 (25.2%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u0026gt;75\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e83 (27.6%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e57 (30.6%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e26 (22.6%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eN-terminal pro-B-type natriuretic peptide (NT-proBNP), median (Q1, Q3)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e1255 (445, 3231)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e374.5 (211.57, 660.5)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e2651 (1295, 5166)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.000 *\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003ctfoot\u003e\u003ctr\u003e\u003ctd colspan=\"5\"\u003eAHF: acute heart failure; NAHF: nonacute heart failure; \u0026lowast;P\u0026thinsp;\u0026lt;\u0026thinsp;0.05.\u003c/td\u003e\u003c/tr\u003e\u003c/tfoot\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eResults of univariate and multivariate regression analyses of the risk of acute heart failure in patients with acute myocardial infarction\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"5\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eVariable\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eUnivariate OR (95% CI)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eUnivariate P\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eMultivariate OR (95% CI)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003eMultivariate P\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSex\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.350 (0.170\u0026ndash;0.650)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.002 *\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.342 (0.141\u0026ndash;0.789)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.014 *\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAtrial fibrillation\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e4.710 (1.780\u0026ndash;16.270)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.005 *\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u0026ndash;\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u0026ndash;\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003efQRS\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e3.210 (1.440\u0026ndash;8.150)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.007 *\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u0026ndash;\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u0026ndash;\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAbnormal Q wave\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e2.910 (1.800\u0026ndash;4.730)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.000 *\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e2.390 (1.316\u0026ndash;4.402)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.004 *\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLow-voltage QRS\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e3.250 (1.620\u0026ndash;7.120)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.002 *\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u0026ndash;\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u0026ndash;\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAlcohol use\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.410 (0.230\u0026ndash;0.710)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.002 *\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u0026ndash;\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u0026ndash;\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eHyperglycemia\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e1.550 (0.920\u0026ndash;2.660)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.104\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u0026ndash;\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u0026ndash;\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLVEF\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.860 (0.820\u0026ndash;0.890)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.000 *\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.864 (0.819\u0026ndash;0.906)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.000 *\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eHeart rate\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e6.210 (2.390\u0026ndash;21.240)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.001 *\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e4.004 (1.235\u0026ndash;16.350)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.032 *\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eQTc interval\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e5.680 (3.180\u0026ndash;10.670)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.000 *\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e2.252 (1.092\u0026ndash;4.756)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.030 *\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTp-e interval\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.830 (0.470\u0026ndash;1.440)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.505\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u0026ndash;\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u0026ndash;\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAge 60\u0026ndash;75\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e1.960 (1.180\u0026ndash;3.310)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.010 *\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e2.607 (1.393\u0026ndash;5.001)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.003 *\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAge\u0026thinsp;\u0026lt;\u0026thinsp;60\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.380 (0.240\u0026ndash;0.620)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.000 *\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u0026ndash;\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u0026ndash;\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003ctfoot\u003e\u003ctr\u003e\u003ctd colspan=\"5\"\u003eLVEF: left ventricular ejection fraction; \u0026lowast;P\u0026thinsp;\u0026lt;\u0026thinsp;0.05.\u003c/td\u003e\u003c/tr\u003e\u003c/tfoot\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003e\u003cb\u003eFeature variable selection by LASSO regression analysis\u003c/b\u003e\u003c/p\u003e\u003cp\u003eFigure \u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e shows the results of LASSO regression analysis of all the clinical and ECG variables; 10-fold cross-validation was used to determine the optimal regularization parameter λ for screening the variables with predictive value for acute heart failure. In the figure, panel A shows the LASSO regression paths, with each curve representing the change trend of the regression coefficient of a variable at different λ values. As λ gradually increases, the penalty imposed by the model on the variables is strengthened, and the coefficients of most variables gradually approach 0, indicating that their contribution to the model is small. Panel B shows the average binomial deviation corresponding to different λ values obtained via cross-validation, with red dots representing the mean and gray error bars representing the standard error (min=-3.9), as well as the simplest model parameter within a standard error range (λ.1se=-3.3). Finally, the model corresponding to λ.1se was selected to achieve the best generalizability and model simplification. At this λ value, there were 8 variables with nonzero coefficients, namely, sex, abnormal Q wave, left ventricular ejection fraction, heart rate, QTc interval, 60\u0026ndash;75 years of age, fQRS and NT-proBNP, which were ultimately included in the subsequent multivariate regression model. These variables were considered key factors in predicting the occurrence of acute heart failure.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003cb\u003eUnivariate and multivariate Logistic regression analyses\u003c/b\u003e\u003c/p\u003e\u003cp\u003eUnivariate logistic regression revealed that a variety of clinical and ECG parameters were strongly related to acute heart failure (AHF). Individuals with atrial fibrillation (OR\u0026thinsp;=\u0026thinsp;4.710, 95% CI: 1.780\u0026ndash;16.270, P\u0026thinsp;=\u0026thinsp;0.005) exhibited a significantly increased risk of AHF, which was related to the presence of a fragmented QRS complex (fQRS, OR\u0026thinsp;=\u0026thinsp;3.210, 95% CI: 1.440\u0026ndash;8.150, P\u0026thinsp;=\u0026thinsp;0.007), an abnormal Q wave (OR\u0026thinsp;=\u0026thinsp;2.910, 95% CI: 1.800\u0026ndash;4.730, P\u0026thinsp;\u0026lt;\u0026thinsp;0.001), and low-voltage QRS (OR\u0026thinsp;=\u0026thinsp;3.250, 95% CI: 1.620\u0026ndash;7.120, P\u0026thinsp;=\u0026thinsp;0.002). Moreover, a heart rate exceeding 100 beats per minute (OR\u0026thinsp;=\u0026thinsp;6.210, 95% CI: 2.390\u0026ndash;21.240, P\u0026thinsp;=\u0026thinsp;0.001) also substantially increased the risk of AHF. Conversely, the risk of AHF was significantly lower in females (OR\u0026thinsp;=\u0026thinsp;0.350, 95% CI: 0.170\u0026ndash;0.650, P\u0026thinsp;=\u0026thinsp;0.002) as well as those who consumed alcohol (OR\u0026thinsp;=\u0026thinsp;0.410, 95% CI: 0.230\u0026ndash;0.710, P\u0026thinsp;=\u0026thinsp;0.002). The left ventricular ejection fraction (LVEF) was a constant variable, and each unit increase in the LVEF was related to a significantly lower risk of AHF (OR\u0026thinsp;=\u0026thinsp;0.860, 95% CI: 0.820\u0026ndash;0.890; P\u0026thinsp;\u0026lt;\u0026thinsp;0.001).\u003c/p\u003e\u003cp\u003eThe aforementioned statistically significant variables were subsequently included in the multivariate logistic regression model for additional verification. The outcomes revealed that female sex (OR\u0026thinsp;=\u0026thinsp;0.342, 95% CI: 0.141\u0026ndash;0.789, P\u0026thinsp;=\u0026thinsp;0.014) was still associated with a lower risk of AHF. An abnormal Q wave (OR\u0026thinsp;=\u0026thinsp;2.390, 95% CI: 1.316\u0026ndash;4.402, P\u0026thinsp;=\u0026thinsp;0.004), a prolonged QTc interval (OR\u0026thinsp;=\u0026thinsp;2.252, 95% CI: 1.092\u0026ndash;4.756, P\u0026thinsp;=\u0026thinsp;0.030), a heart rate exceeding 100 beats per minute (OR\u0026thinsp;=\u0026thinsp;4.004, 95% CI: 1.235\u0026ndash;16.350, P\u0026thinsp;=\u0026thinsp;0.032), and age 60\u0026ndash;75 years (OR\u0026thinsp;=\u0026thinsp;2.607, 95% CI: 1.393\u0026ndash;5.001, P\u0026thinsp;=\u0026thinsp;0.003) were independent risk factors for AHF. The LVEF remained significant in the multivariate model (OR\u0026thinsp;=\u0026thinsp;0.864, 95% CI: 0.819\u0026ndash;0.906, P\u0026thinsp;\u0026lt;\u0026thinsp;0.001), indicating that it had a stable protective effect against AHF.\u003c/p\u003e\u003cp\u003e\u003cb\u003eConstruction and interpretation of the nomogram model\u003c/b\u003e\u003c/p\u003e\u003cp\u003eFigure \u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e shows the nomogram that was constructed to predict the risk of acute heart failure (AHF) in patients with acute myocardial infarction based on the results of the multivariate logistic regression model. Five categorical variables (male sex, an abnormal Q wave, heart rate\u0026thinsp;\u0026gt;\u0026thinsp;100 beats/min, a prolonged QTc interval, and age between 60 and 75 years) and one continuous variable (left ventricular ejection fraction, LVEF) were included in the model. In the nomogram, each variable corresponds to a scoring axis, and different risk scores are assigned according to its clinical situation. By adding the score for each variable to obtain the total score, the probability of AHF can be predicted according to the total number of points. As shown in the figure, the LVEF has the greatest weight in the model, and the range of the LVEF score is wide, indicating that it contributes the most to the prediction of AHF risk. This is followed by the QTc interval and heart rate, sex, Q wave abnormality, and age range. The probability range of the nomogram is 0.1\u0026ndash;0.9, which indicates good risk stratification. Thus, the nomogram could help clinicians quickly evaluate the risk of AHF in an individual and aid early identification and precise treatment of AHF.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003cb\u003eModel discrimination ability and calibration performance evaluation\u003c/b\u003e\u003c/p\u003e\u003cp\u003eFigure \u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e shows the performance of the established model in terms of discrimination ability and degree of calibration. As shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eA, the area under the ROC curve (AUC) of the combined model was 0.8404, which was greater than that of any single indicator, including fQRS (AUC\u0026thinsp;=\u0026thinsp;0.5525), QTc interval (AUC\u0026thinsp;=\u0026thinsp;0.7403), QT dispersion (QTd, AUC\u0026thinsp;=\u0026thinsp;0.5715) and left ventricular ejection fraction (LVEF, AUC\u0026thinsp;=\u0026thinsp;0.7712), indicating that this multivariate model had better discrimination power in identifying acute heart failure. The red curve shows the LOWESS smooth fitting results, the blue points and error bars represent the actual incidence and 95% confidence intervals after grouping, and the gray dashed line represents the ideal calibration reference line. The results showed that the predicted probability of the model was highly consistent with the actual incidence, indicating good calibration of the model in effectively predicting the probability of disease in an individual and its strong potential for clinical application.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003cb\u003eInternal validation results of the model\u003c/b\u003e\u003c/p\u003e\u003cp\u003eFigure \u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e displays the validation of the constructed prognostic paradigm, which performed well in terms of both discrimination and calibration. As shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eA, the AUC was 0.823, with a 95% confidence interval ranging from 0.776 to 0.871, indicating that the model could effectively identify acute heart failure. The blue shaded area indicates the 95% confidence interval, and the dashed horizontal blue line represents the benchmark error of sensitivity at shifting limits, with the model maintaining high sensitivity across most of the range. Panel B illustrates the calibration curve for the model generated via the bootstrap technique (bootstrapping iterations, B\u0026thinsp;=\u0026thinsp;2000). The green line represents the fitting result of the predicted risk corrected by the observed rate, the red dotted line corresponds to the observed results, and the yellow dotted line represents the ideal calibration curve. Overall, the calibration curve was very close to the ideal calibration curve, indicating that the prognostic outcomes of the model were in good agreement with the real outcomes and that the model demonstrated high calibration accuracy.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003cb\u003eClinical utility of the line chart according to decision curve analysis\u003c/b\u003e\u003c/p\u003e\u003cp\u003eTo further verify the practical value of the established model in clinical decision-making, we performed decision curve analysis, as shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e. The horizontal axis shows the predicted risk threshold, and the vertical axis shows the net benefit. The results revealed that the model (orange curve) had a greater net benefit than \"all intervention\" (blue curve) and \"no intervention\" (black baseline) when the threshold probability was in the range of approximately 0.13 to 0.65, indicating its good clinical utility within this probability range. Especially within the common decision-making risk range of 0.1\u0026ndash;0.5, the model was consistently superior to the traditional treatment strategy, indicating that it can serve as an auxiliary basis for clinical judgment and help achieve more reasonable patient stratification and management.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eFigure \u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003e Violin plots showing the correlations of the Killip grade of acute heart failure with electrocardiogram parameters\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003cb\u003eCorrelation analysis of the Killip grade and electrocardiogram parameters in patients with acute myocardial infarction complicated with acute heart failure\u003c/b\u003e\u003c/p\u003e\u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e shows the correlations between the Killip grade and six key clinical and computerized ECG parameters. Spearman correlation analysis revealed that QTc interval (ρ\u0026thinsp;=\u0026thinsp;0.392, P\u0026thinsp;=\u0026thinsp;0.001), heart rate (ρ\u0026thinsp;=\u0026thinsp;0.391, P\u0026thinsp;=\u0026thinsp;0.001) and age (ρ\u0026thinsp;=\u0026thinsp;0.265, P\u0026thinsp;=\u0026thinsp;0.003) were significantly positively correlated with the Killip grade, whereas P-wave duration (ρ=-0.257, P\u0026thinsp;=\u0026thinsp;0.004) and P-wave dispersion (ρ=-0.226, P\u0026thinsp;=\u0026thinsp;0.012) and QRS duration were weakly but still statistically significantly correlated (ρ\u0026thinsp;=\u0026thinsp;0.184, P\u0026thinsp;=\u0026thinsp;0.043). Figure\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003e further shows the distribution of each parameter in patients with disease of different Killip grades via violin plots. The results indicate that (A and B) the QTc interval and heart rate increased with increasing Killip grade, (C) age increased slightly at high Killip grades, and (D and E) the P-wave duration and P-wave dispersion decreased but (F) the QRS duration slightly differed but tended to increase with increasing Killip grade. The above results suggest that some ECG parameters have potential reference value in determining the severity of heart failure.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eSpearman correlation analysis of the Killip grade and electrocardiogram parameters in patients with acute heart failure\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"3\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eVariable\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eSpearman coefficient\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eP value\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eQTc Interval\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.392\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.001\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eHeart Rate\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.391\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.001\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAge\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.265\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.003\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eP-wave Duration\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e-0.257\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.004\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eP-wave Dispersion\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e-0.226\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.012\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eQRS Duration\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.184\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.043\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eIn this study, we developed a risk prediction model for acute heart failure based on routine electrocardiogram (ECG) data in a primary population of AMI patients. QT prolongation, abnormal Q waves, tachycardia, a reduced LVEF, an advanced age and male sex were found to be independent risk factors for acute heart failure after AMI. The discrimination power (AUC) of the model was 0.84. Further analysis revealed that the Killip grade was significantly correlated with the QTc interval, heart rate, age and many other ECG parameters, whereas the P-wave duration was negatively correlated with dispersion, suggesting that these indicators not only have predictive value but also reflect the severity of heart failure, which strengthens the rationality and clinical value of the model variables. The main findings are discussed in the context of the literature, and the model is compared with other models based on different parameters to enhance the clinical application value and theoretical significance of the present model.\u003c/p\u003e\u003cp\u003e\u003cb\u003eInterpretation and comparison of the main findings\u003c/b\u003e\u003c/p\u003e\u003cp\u003eQTc interval prolongation is considered the most prominent cardiac risk indicator on electrocardiography. The QTc interval reflects the full duration of ventricular repolarization. If the QTc interval is prolonged, the conduction velocity of the myocardial action potential is uneven, which is an important cause of heart failure and atrial fibrillation. Welten[\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e] et al. found that in patients with acute myocardial infarction, a QTc interval longer than 460 ms was significantly associated with left ventricular dysfunction and that the risk of heart failure was increased by approximately 11% for every 10 ms of QTc interval prolongation. In contrast, in the present study, a clear cutoff value was not set, but the QTc interval was identified as a key variable via the LASSO method to better predict clinical risk. Abnormal Q waves usually indicate a large area of extensive myocardial necrosis, which is closely related to both ventricular dyskinesis and cardiac remodeling. Houdmont[\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e] found that in patients with acute myocardial infarction in the anterior wall, the depth of the Q wave was positively correlated with the left ventricular end-diastolic volume, which independently predicted the likelihood of heart failure during the waiting period. The results of this study are consistent with these findings and suggest that Q-wave characteristics can also be used to roughly assess risk in primary care settings. The increase in heart rate during acute myocardial infarction may result from sympathetic tension or from compensatory mechanisms of cardiac output, and tachycardia may induce heart failure. Chan[\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e] reported that decreased heart rate variability and an elevated heart rate were strongly associated with in-hospital heart failure. In this study, the resting heart rate was used as a variable to show that the heart rate itself is a valuable indicator, even when heart rate variability monitoring devices are not used. Reduced left ventricular systolic function is directly related to heart failure, especially in patients experiencing acute myocardial infarction for the first time. Park[\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e] confirmed that an LVEF\u0026thinsp;\u0026lt;\u0026thinsp;45% is the strongest predictor of major adverse cardiac events within a year. This study further highlights the irreplaceable role of cardiac ultrasound in predicting heart failure.\u003c/p\u003e\u003cp\u003e\u003cb\u003eInteresting findings regarding other parameters\u003c/b\u003e\u003c/p\u003e\u003cp\u003eThis study revealed several interesting findings, particularly regarding the interaction between cardiac electrophysiological features and personal characteristics such as sex and age. First, a fragmented QRS (fQRS), a marker of myocardial scarring and electrical conduction abnormalities, performs better in predicting acute heart failure after AMI than other variables and is more common in men than in women, suggesting that it may reflect sex-related differences[\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e, \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e] in structural remodeling. Moreover, prolongation of the Tp-e interval and QT dispersion indicate increased heterogeneity in ventricular repolarization, and these repolarization parameters have greater predictive sensitivity in female patients, suggesting a difference in electrophysiological responses by sex.[\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e] In addition, the predictive performance of electrophysiological parameters increased with age, especially in elderly patients, indicating that age is not only an independent risk factor but also may change the risk assessment mode[\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e] by affecting the expression of electrophysiological parameters. Notably, the combination of multiple ECG parameters was superior to any single variable in the prediction of heart failure, which strengthens the potential of multidimensional electrical signal analysis. However, some traditional clinical parameters (such as atrial fibrillation or simple heart rate) have limited predictive value, further highlighting the advantages[\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e] of electrophysiological characteristics in individualized risk assessment.\u003c/p\u003e\u003cp\u003eP-wave dispersion in patients was not directly explored in this study, but related studies suggest that it may have predictive power. Yang[\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e] reported that P-wave dispersion\u0026thinsp;\u0026gt;\u0026thinsp;40 ms may predict heart failure after acute myocardial infarction, which may be related to heart failure caused by atrial fibrillation. This finding suggests that changes in P-wave dispersion could be incorporated in future models to improve sensitivity. This study did not include parameters related to ST-segment changes, but Kazemi[\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e] study revealed that an ST shift in lead aVR can predict left ventricular dysfunction in patients with multivesicular disease. Siren[\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e] et al. found that incomplete ST-segment resolution after PCI was associated with a higher rate of new or worsening heart failure, suggesting ST resolution is a valuable predictor for post-infarction outcomes and that continuous ECG monitoring may enhance early risk stratification. The Tp-e interval was not evaluated in this study, but \u0026Ouml;zbek[\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e] reported that its prolongation predicts increased dysregulation of compensatory mechanisms and is an important indicator of heart failure and malignant arrhythmias after acute myocardial infarction. Galloway[\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e] et al. established an algorithm for identifying T-wave alternation based on electrocardiogram (ECG) images and deep learning models and suggested that T-wave variation could predict heart failure in the absence of significant structural changes.\u003c/p\u003e\u003cp\u003eThe present study also revealed that the Killip grade was significantly correlated with ECG parameters such as the QTc interval, heart rate, and P-wave characteristics, indicating its potential value in the assessment of AHF. With increasing Killip grade, the QTc interval was also prolonged, which is consistent with El Amrawy et al.[\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e] observation that prolongation of the QTc interval significantly predicts a higher Killip grade and a higher risk of in-hospital mortality in STEMI patients. Heart rate was also positively correlated with the Killip grade, and Jensen[\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e] et al. confirmed that a higher heart rate is an independent predictor of a higher Killip grade and poor outcomes. In addition, with increasing Killip grade, the P-wave duration decreased and conductivity decreased, which may reflect atrioventricular conduction remodeling. Bulluck[\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e] et al. proposed that the combination of electrophysiological and imaging parameters can enhance the prediction power of the Killip grade, which indirectly supports this finding. Although the correlation of the Killip grade with the QRS duration was weak in this study, Gabaldon‒Perezi[\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e] et al. noted that flow disturbances in patients with Killip grade I-II disease may also cause serious complications, suggesting that electrocardiographic abnormalities also have predictive value in patients with low Killip grades.\u003c/p\u003e\u003cp\u003e\u003cb\u003eLimitations and future prospects\u003c/b\u003e\u003c/p\u003e\u003cp\u003eAlthough this study explored the feasibility of using simple ECG parameters to identify the risk of heart failure in patients with myocardial infarction, there are numerous limitations. The sample size was limited, and the patients were from a single health care institution, which may have resulted in selection bias. The model does not consider indicators of dynamic ECG changes such as T-wave directionality, the T-wave interapical interval, or waveform analysis, which limits its generalizability. In addition, multicenter validation was not carried out, and the external reliability of the model remains to be tested. In the future, the number of samples needs to be increased, and more abnormalities should be automatically identified via artificial intelligence-based image recognition technology. Moreover, continuous indicators such as HRV should be included to improve the prediction accuracy and construct an early warning system.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eIn this study, a simple and practical risk prediction model for AMI complicated with AHF based on routine ECG and clinical characteristics was established. In the internal validation, the model performed well in terms of discrimination and calibration, indicating suitability for the early identification of high-risk patients in primary care settings to aid treatment. Notably, several ECG parameters, such as the Killip grade, QTc interval, heart rate and P-wave dispersion, were significantly correlated with the severity of AHF, indicating that these parameters not only have predictive value but can also assist in assessing the severity of AHF. These findings provide a new perspective for the clinical use of ECG findings beyond risk assessment and further expand the explanatory power and clinical practicability of the model. The stability and generalizability of the model could be further verified using external datasets from more hospitals in the future.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003ch2\u003eEthical approval and consent to participate\u003c/h2\u003e\u003cp\u003e This study was approved by the Medical Ethics Committee of Xuancheng People's Hospital (approval number :2025-byky001-01) and fully complied with the ethical principles of the Declaration of Helsinki and subsequent versions. All patients who participated in the study signed an informed consent form at the time of visiting the hospital, and agreed to the hospital to use their medical data for anonymous scientific research analysis. This was a retrospective observational study, and no additional interventions were performed. The study data were desensitized to protect patient privacy.\u003c/p\u003e\u003c/p\u003e\u003cp\u003e\u003ch2\u003eConflict of Interest Statement\u003c/h2\u003e\u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e\u003c/p\u003e\u003ch2\u003eFunding\u003c/h2\u003e\u003cp\u003eThis study was funded by the universal-level Key Project of Natural Science of Bengbu Medical University in Anhui Province (project number: 2024byzd192). The project name was \"Construction of prognosis prediction model of acute myocardial Infarction based on electrocardiogram multimodal data\". The research was conducted by Gao Guoliang of Xuancheng People's Hospital in Anhui Province. The funding supported all stages of work, including clinical data acquisition, ECG feature extraction, multivariate statistical model construction and model visualization development. The research team independently completed the whole process from project design to data analysis to paper writing, and the funders did not participate in any part of the research implementation. This study conforms to the scientific research ethics and statistical norms. The conclusions drawn are authentic and reliable, and also have certain clinical value.\u003c/p\u003e\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eX.G. was responsible for project establishment, project management, data interpretation, and paper writing and revision. G.G. was responsible for the collection of clinical data and the drawing of statistical analysis charts. G.Y. and A.X. made ECG judgment, searched literature and wrote the first draft of the manuscript. H.H. and X.Z. established the prediction model and result visualization, and participated in the interpretation of the results and revision of the paper. All the authors have read and agreed to the final draft and are jointly responsible for the authenticity and completeness of the study results.\u003c/p\u003e\u003ch2\u003eAcknowledgement\u003c/h2\u003e\u003cp\u003eThanks to the kindness and care of all the teachers, I have received help and support from many aspects in this research work. Special thanks to the team of Cardiology Department of Xuancheng People's Hospital, Anhui Province, who provided great assistance in sample screening and data collection; We would also like to thank Xuancheng People's Hospital of Anhui Province and Bengbu Medical University for funding of the research project; At the same time, we would also like to thank the patients and their families who participated in this study, who provided valuable data and were an important basis for this study.\u003c/p\u003e\u003ch2\u003eData Availability\u003c/h2\u003e\u003cp\u003eThe datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eAllescher, J. et al. QRS fragmentation does not predict mortality in survivors of acute myocardial infarction. \u003cem\u003eClin. Cardiol.\u003c/em\u003e \u003cb\u003e47\u003c/b\u003e, e24218 (2024).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWelten, S. et al. Prolongation of the QTc interval is associated with an increased risk of cardiovascular diseases: The Hoorn study. \u003cem\u003eJ. Electrocardiol.\u003c/em\u003e \u003cb\u003e80\u003c/b\u003e, 133\u0026ndash;138 (2023).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eXu, Y. et al. Predicting efficacy of combined assessment with fragmented QRS and severely depressed heart rate variability on outcome of patients with acute myocardial infarction. \u003cem\u003eHeart Vessels\u003c/em\u003e. \u003cb\u003e37\u003c/b\u003e, 239\u0026ndash;249 (2022).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eT\u0026uuml;rkmen, S., Bozkurt, M., Hoşoğlu, Y. \u0026amp; G\u0026ouml;l, M. Significance of fragmented QRS and predictors of outcome in ST-elevation myocardial infarction. \u003cem\u003eJ. Res. Med. Sci.\u003c/em\u003e \u003cb\u003e29\u003c/b\u003e, 23 (2024).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHan, X. et al. Prognostic significance of QRS distortion and frontal QRS-T angle in patients with ST-elevation myocardial infarction. \u003cem\u003eInt. J. Cardiol.\u003c/em\u003e \u003cb\u003e345\u003c/b\u003e, 1\u0026ndash;6 (2021).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZorlu, \u0026Ccedil;., A\u0026ccedil;ıkel, B., \u0026Ouml;m\u0026uuml;r, S. E. \u0026amp; Frontal Plane QRS - T Angle Is a Predictor of Ventricular Arrhythmia in Heart Failure With Preserved Ejection Fraction. \u003cem\u003eAnn. Noninvasive Electrocardiol.\u003c/em\u003e \u003cb\u003e30\u003c/b\u003e, e70062 (2025).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMart\u0026iacute;nez-Sell\u0026eacute;s, M. \u0026amp; Marina-Breysse, M. Current and Future Use of Artificial Intelligence in Electrocardiography. \u003cem\u003eJournal Cardiovasc. Dev. disease\u003c/em\u003e \u003cb\u003e10\u003c/b\u003e, (2023).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eByrne, R. A. et al. 2023 ESC Guidelines for the management of acute coronary syndromes. \u003cem\u003eEur. Heart J.\u003c/em\u003e \u003cb\u003e44\u003c/b\u003e, 3720\u0026ndash;3826 (2023).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMcDonagh, T. A. et al. Focused Update of the 2021 ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure. \u003cem\u003eEur Heart J\u003c/em\u003e. 44, 3627\u0026ndash;3639 (2023). (2023).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZeppenfeld, K. et al. ESC Guidelines for the management of patients with ventricular arrhythmias and the prevention of sudden cardiac death. \u003cem\u003eEur Heart J\u003c/em\u003e. 43, 3997\u0026ndash;4126 (2022). (2022).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eVilla, A. et al. A machine learning algorithm for electrocardiographic fQRS quantification validated on multi-center data. \u003cem\u003eSci. Rep.\u003c/em\u003e \u003cb\u003e12\u003c/b\u003e, 6783 (2022).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHoudmont, M. et al. Pathological Q waves at presentation of anterior ST segment elevation myocardial infarction predict heart failure: a Southeast Asian perspective. \u003cem\u003eCoron. Artery Dis.\u003c/em\u003e \u003cb\u003e36\u003c/b\u003e, 378\u0026ndash;383 (2025).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eChan, J. S. K. et al. Fragmented QRS Is Independently Predictive of Long-Term Adverse Clinical Outcomes in Asian Patients Hospitalized for Heart Failure: A Retrospective Cohort Study. \u003cem\u003eFront. Cardiovasc. Med.\u003c/em\u003e \u003cb\u003e8\u003c/b\u003e, 738417 (2021).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePark, C. S. et al. Left Ventricular Ejection Fraction 1 Year After Acute Myocardial Infarction Identifies the Benefits of the Long-Term Use of β-Blockers: Analysis of Data From the KAMIR-NIH Registry. \u003cem\u003eCirc. Cardiovasc. Interv\u003c/em\u003e. \u003cb\u003e14\u003c/b\u003e, e010159 (2021).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLuo, G. et al. The Predictive Value of Fragmented QRS for Cardiovascular Events in Acute Myocardial Infarction: A Systematic Review and Meta-Analysis. \u003cem\u003eFront. Physiol.\u003c/em\u003e \u003cb\u003e11\u003c/b\u003e, 1027 (2020).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTerho, H. K. et al. Prevalence and prognostic significance of fragmented QRS complex in middle-aged subjects with and without clinical or electrocardiographic evidence of cardiac disease. \u003cem\u003eAm. J. Cardiol.\u003c/em\u003e \u003cb\u003e114\u003c/b\u003e, 141\u0026ndash;147 (2014).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eIwahashi, N. et al. Global Strain Measured by Three-Dimensional Speckle Tracking Echocardiography Is a Useful Predictor for 10-Year Prognosis After a First ST-Elevation Acute Myocardial Infarction. \u003cem\u003eCirc. J.\u003c/em\u003e \u003cb\u003e85\u003c/b\u003e, 1735\u0026ndash;1743 (2021).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHempel, P. et al. Explainable AI associates ECG aging effects with increased cardiovascular risk in a longitudinal population study. \u003cem\u003eNPJ Digit. Med.\u003c/em\u003e \u003cb\u003e8\u003c/b\u003e, 25 (2025).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLee, M. S. et al. Artificial intelligence applied to electrocardiogram to rule out acute myocardial infarction: the ROMIAE multicentre study. \u003cem\u003eEur. Heart J.\u003c/em\u003e \u003cb\u003e46\u003c/b\u003e, 1917\u0026ndash;1929 (2025).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eYang, N. et al. The Role of P-Wave Variables in Enhancing Prediction of New-Onset Atrial Fibrillation in Patients With Acute Myocardial Infarction. \u003cem\u003eAnn. Noninvasive Electrocardiol.\u003c/em\u003e \u003cb\u003e30\u003c/b\u003e, e70041 (2025).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKazemi, E. et al. The prognostic effect of ST-elevation in lead aVR on coronary artery disease, and outcome in acute coronary syndrome patients: a systematic review and meta-analysis. \u003cem\u003eEur. J. Med. Res.\u003c/em\u003e \u003cb\u003e27\u003c/b\u003e, 302 (2022).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSir\u0026eacute;n, M. et al. The prognostic significance of single-lead ST-segment resolution in ST-segment elevation myocardial infarction patients treated with primary PCI - A substudy of the randomized TOTAL trial. \u003cem\u003eAm. Heart J.\u003c/em\u003e \u003cb\u003e269\u003c/b\u003e, 149\u0026ndash;157 (2024).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003e\u0026Ouml;zbek, S. C. \u0026amp; S\u0026ouml;kmen, E. Usefulness of Tp-Te interval and Tp-Te/QT ratio in the prediction of ventricular arrhythmias and mortality in acute STEMI patients undergoing fibrinolytic therapy. \u003cem\u003eJ. Electrocardiol.\u003c/em\u003e \u003cb\u003e56\u003c/b\u003e, 100\u0026ndash;105 (2019).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGalloway, C. D. et al. Development and Validation of a Deep-Learning Model to Screen for Hyperkalemia From the Electrocardiogram. \u003cem\u003eJAMA Cardiol.\u003c/em\u003e \u003cb\u003e4\u003c/b\u003e, 428\u0026ndash;436 (2019).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eEl Amrawy, A. M. et al. QTc interval prolongation impact on in-hospital mortality in acute coronary syndromes patients using artificial intelligence and machine learning. \u003cem\u003eEgypt. Heart J.\u003c/em\u003e \u003cb\u003e76\u003c/b\u003e, 149 (2024).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eJensen, M. T. et al. Heart rate at admission is a predictor of in-hospital mortality in patients with acute coronary syndromes: Results from 58 European hospitals: The European Hospital Benchmarking by Outcomes in acute coronary syndrome Processes study. \u003cem\u003eEur. Heart J. Acute Cardiovasc. Care\u003c/em\u003e. \u003cb\u003e7\u003c/b\u003e, 149\u0026ndash;157 (2018).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBulluck, H. et al. Independent Predictors of Cardiac Mortality and Hospitalization for Heart Failure in a Multi-Ethnic Asian ST-segment Elevation Myocardial Infarction Population Treated by Primary Percutaneous Coronary Intervention. \u003cem\u003eSci. Rep.\u003c/em\u003e \u003cb\u003e9\u003c/b\u003e, 10072 (2019).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGabald\u0026oacute;n-P\u0026eacute;rez, A. et al. Prognostic value of cardiac magnetic resonance early after ST-segment elevation myocardial infarction in older patients. \u003cem\u003eAge Ageing\u003c/em\u003e \u003cb\u003e51\u003c/b\u003e, (2022).\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Acute heart failure, Electrocardiogram, Acute myocardial infarction, Nomogram model, Killip classification","lastPublishedDoi":"10.21203/rs.3.rs-7286686/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7286686/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eAcute myocardial infarction (AMI) is a leading cause of acute heart failure (AHF), which markedly increases patient mortality and readmission rates. This study aimed to develop a practical, ECG-based model for early AHF risk prediction in AMI patients. We retrospectively analyzed clinical and electrocardiogram data from 301 AMI patients (October 2022–March 2025). Six predictors—prolonged QTc interval, abnormal Q wave, heart rate\u0026gt;100 bpm, reduced left ventricular ejection fraction, male sex, and age (60–75 years)—were selected via LASSO and incorporated into a logistic regression nomogram. The model demonstrated strong discrimination (AUC: 0.840; internal validation AUC: 0.823) and clinical utility (threshold probability: 0.13–0.65). Killip grade was positively correlated with QTc interval, heart rate, age, and QRS duration, and negatively with P-wave duration and dispersion. This nomogram offers a reliable and resource-efficient tool for early identification of AHF risk in AMI patients, especially in low-resource or primary care settings.\u003c/p\u003e","manuscriptTitle":"Construction of a predictive model for acute myocardial infarction combined with acute heart failure based on electrocardiogram parameters and exploration of the correlations of these parameters with the Killip grade","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-08-14 10:00:03","doi":"10.21203/rs.3.rs-7286686/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-10-14T10:10:21+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-09-18T09:11:14+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-09-17T13:34:11+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"74205282139174873411710800105630259158","date":"2025-09-08T04:00:10+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"31946121994467278157335133931130702892","date":"2025-09-08T03:54:28+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-08-16T01:26:39+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"99561563123388530087043669491442310462","date":"2025-08-14T00:45:15+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"179443862345782171849031646398103057496","date":"2025-08-10T06:54:22+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-08-08T06:06:36+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2025-08-07T05:44:11+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-08-05T06:16:06+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-08-05T04:20:28+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2025-08-04T04:05:02+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"f59ddcc2-c11e-4b28-b6e2-9f704ffc0a91","owner":[],"postedDate":"August 14th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":52922345,"name":"Health sciences/Cardiology"},{"id":52922346,"name":"Health sciences/Diseases"},{"id":52922347,"name":"Health sciences/Medical research"},{"id":52922348,"name":"Health sciences/Risk factors"}],"tags":[],"updatedAt":"2026-02-23T16:00:13+00:00","versionOfRecord":{"articleIdentity":"rs-7286686","link":"https://doi.org/10.1038/s41598-026-40600-9","journal":{"identity":"scientific-reports","isVorOnly":false,"title":"Scientific Reports"},"publishedOn":"2026-02-18 15:57:06","publishedOnDateReadable":"February 18th, 2026"},"versionCreatedAt":"2025-08-14 10:00:03","video":"","vorDoi":"10.1038/s41598-026-40600-9","vorDoiUrl":"https://doi.org/10.1038/s41598-026-40600-9","workflowStages":[]},"version":"v1","identity":"rs-7286686","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7286686","identity":"rs-7286686","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}