{"paper_id":"4b222faa-13cb-4968-9fad-e73f856b6e36","body_text":"Mortality Prediction in the elderly patients with Coronary Artery Disease and Atrial Fibrillation: A Retrospective Machine Learning Approach | 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 Research Article Mortality Prediction in the elderly patients with Coronary Artery Disease and Atrial Fibrillation: A Retrospective Machine Learning Approach Yuyan Wang, Yangxun Wu, Yuting Zou, Shuai Mao, Yanzhu Yao, Yifan Wang, and 4 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8337799/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Background: Elderly patients with coexisting coronary artery disease (CAD) and atrial fibrillation (AF) are at significantly increased risk of mortality. Accurate risk stratification is crucial for improving clinical management, yet a dedicated predictive tool for this specific population is lacking. The widely used CHA₂DS₂-VASc score demonstrates limited performance in predicting all-cause mortality in this complex comorbid group. Methods: A cohort of elderly inpatients (≥65 years) diagnosed with CAD and AF were retrospectively enrolled from the Department of Cardiology at the Chinese PLA General Hospital between January 2010 and December 2017. Baseline clinical data during hospitalization were collected, and all patients were followed up for all-cause mortality. Patients were randomly divided into a training set (70%) and a validation set (30%). Variable selection was performed using LASSO-Cox regression. Predictive models were established through a Cox proportional hazards model (via LASSO-Cox) and a random survival forest model to predict all-cause mortality risk. Model performance was evaluated using the area under the receiver operating characteristic curve (AUC), calibration curves, and decision curve analysis (DCA), and compared against the CHA₂DS₂-VASc score. Results: A total of 1678 elderly patients with CAD and AF were randomly divided into a training set (n=1174) and a validation set (n=504). Through LASSO‑Cox regression, 17 variables were identified as predictors associated with all‑cause mortality. Two distinct models were developed: LASSO‑Cox Model A included factors such as age, left ventricular ejection fraction, blood glucose, plasma fibrinogen, D‑dimer, prothrombin time, NT‑proBNP, hemoglobin, and hematocrit; LASSO‑Cox Model B incorporated all variables from Model A plus acute myocardial infarction, history of myocardial infarction, renal insufficiency, heart failure, diabetes, stroke, chronic obstructive pulmonary disease, and malignant tumor. The analysis of Cox proportional hazards regression models showed that in the training sets, the Area Under Curves (AUC) of model A, model B and CHA₂DS₂-VASc scores for predicting 1-year all-cause death were 0.83, 0.85 and 0.66, respectively; the AUC for 5-year all-cause death were 0.74, 0.76 and 0.62, respectively. In the validation sets, the AUC of model A, model B and the CHA₂DS₂-VASc score for predicting 1-year all-cause mortality were 0.79, 0.78 and 0.57, respectively; the AUC for 5-year all-cause death were 0.74, 0.75 and 0.57, respectively. Model B demonstrated better calibration and provided greater net clinical benefit in DCA. Conclusion: The LASSO‑Cox machine learning model demonstrates superior predictive performance for all cause mortality relative to the traditional CHA₂DS₂-VASc score in elderly patients with CAD and AF. Elderly Coronary artery disease Atrial fibrillation Mortality Machine learning Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Introduction Coronary artery disease (CAD) and atrial fibrillation (AF) are highly prevalent cardiovascular diseases in China, characterized by an increasing incidence with advancing age [ 1 , 2 ]. CAD is a leading cause of mortality in both developing and developed countries. As an age-related arrhythmia, AF is associated with a significantly elevated risk of death, with approximately half of all deaths in AF patients being attributable to cardiac causes [ 3 ]. CAD and AF share common risk factors and pathophysiological features and therefore often coexist. The population aged 65 and above in China has risen to 264 million, accounting for 18.70% of the total population. The prevalence of comorbidities in the elderly, including these conditions, is increasing along with the accelerating aging of the population. Patients with concomitant CAD and AF have a significantly higher rate of adverse cardiovascular events compared to those with either condition alone [ 4 , 5 ]. Furthermore, among elderly patients (≥ 70 years), both the co-prevalence of these diseases and the risk of adverse clinical outcomes are markedly higher than in younger patients [ 6 ]. The combination of CAD and AF in the elderly is associated with higher mortality [ 7 ], posing a severe threat to life and health. Consequently, predicting the all-cause mortality risk in elderly patients with CAD and AF is of paramount importance for clinical management. The CHA₂DS₂-VASc score model is the most widely used clinical prediction tool for assessing thromboembolic risk and guiding antithrombotic therapy in AF patients [ 8 ]. This scoring system can also predict the risk of adverse outcomes such as all-cause mortality and major adverse cardiac and cerebrovascular events [ 9 – 14 ]. However, it has certain limitations in predicting adverse clinical outcomes specifically in elderly patients with coexisting CAD and AF [ 7 ]. The 2022 European Society of Cardiology guidelines explicitly recommend that a systematic assessment of thrombotic risk in elderly patients with CAD or AF should integrate various types of risk predictors [ 15 ]. Due to the widespread adoption of machine learning algorithms and their significant advantages in predictive performance, several studies have applied these algorithms to develop clinical models for predicting all-cause death in patients with either CAD or AF, demonstrating superior predictive value compared to traditional clinical prediction models. For instance, Chen et al. utilized LASSO regression and the random survival forest algorithm to select variables from 1,223 AF patients aged 55 and above. They employed six machine learning algorithms, including Cox regression, to build models predicting 1-year all-cause mortality. Their results indicated that the model using LASSO regression for variable selection performed the best, exhibiting higher predictive power and clinical net benefit than the traditional CHA₂DS₂-VASc model [ 16 ]. Consequently, using LASSO regression to select clinical risk factors combined with machine learning algorithms to build all-cause mortality risk models has been proven effective and accurate in cardiovascular disease patients aged 50 and above, characterized by CAD or AF, and is significantly superior to traditional clinical prediction models. Among these approaches, combining variable selection via LASSO regression with a Cox proportional hazards model has been shown to achieve optimal results. LASSO regression is favored for variable selection due to its high accuracy and stability, and is commonly used in modeling for cancer, heart failure, and AF populations [ 17 – 19 ]. Furthermore, the LASSO-Cox regression method builds upon LASSO regression by estimating the relationship between the selected variables and the time to endpoint events. It serves as a method to prevent model overfitting and has been widely incorporated into various types of machine learning algorithms [ 20 , 21 ]. Currently, there is a lack of established risk prediction models specifically for all-cause mortality in the elderly population with concomitant CAD and AF. Therefore, this study aims to address this gap by establishing a clinical follow-up cohort database and leveraging advanced big data analytics and machine learning model algorithms. It seeks to thoroughly investigate the performance of a risk assessment tool for all-cause mortality in this specific patient group. The development of such a model holds significant clinical utility and practical value for improving the prognosis of elderly patients with CAD and AF. Materials and methods Study population A consecutive cohort of elderly patients diagnosed with both CAD and AF during hospitalization in the Department of Cardiology, Chinese People's Liberation Army (PLA) General Hospital, between January 2010 and December 2017, was enrolled. The patients were randomly divided into a training set and a validation set at a 7:3 ratio. During the dataset division, stratified sampling was employed to ensure the proportion of samples with and without the endpoint event was identical between the training and validation sets. Inclusion criteria : (1)Age ≥ 65 years; (2) Clinical diagnosis of CAD combined with AF. Exclusion criteria : (1) Inability to undergo follow-up for at least one year; (2) Missing crucial information in electronic medical records (including data on blood biochemistry and past medical history); (3) Participation in other clinical trials; (4) Terminal patients during the hospitalization period. This study was reviewed and approved by the Ethics Committee of the Chinese PLA General Hospital. Informed consent was obtained from all enrolled patients. Definition of disease CAD was defined as stable coronary artery disease (including stable angina pectoris, prior myocardial infarction, and ischemic cardiomyopathy) [ 22 ] or acute coronary syndrome (including unstable angina and acute myocardial infarction) [ 23 ]. AF was defined as a cardiac rhythm characterized by irregular, discrete P waves and irregular RR intervals (in the absence of atrioventricular block) recorded by a standard 12-lead electrocardiogram or a single-lead rhythm strip lasting > 30 seconds [ 8 ], encompassing paroxysmal, persistent, or permanent AF. Collection of clinical information Clinical data of the enrolled subjects were obtained from inpatient medical records or outpatient systems. The collected information included:(1) Demographics: Sex, age, etc. (2) Lifestyle factors: History of smoking and alcohol consumption, etc. (3) Past medical history and comorbidities: History of renal dysfunction, stroke, chronic obstructive pulmonary disease (COPD), peripheral arterial disease, malignant tumors, etc. (4) Previous medication history: History of cardiovascular medication use, etc. (5) CAD type: Stable angina, acute coronary syndrome (ACS) (including unstable angina, acute ST-segment elevation myocardial infarction, acute non-ST-segment elevation myocardial infarction), etc. (6) AF type: Paroxysmal AF, persistent AF, permanent AF. (7) Concomitant medications: Antiplatelet agents, anticoagulants, beta-blockers, calcium channel blockers, statins, amiodarone, etc. (8) Laboratory tests and examinations: Complete blood count, liver and kidney function tests, coagulation profile, left ventricular ejection fraction (LVEF), etc. (9) Interventional procedures and surgeries: Percutaneous coronary intervention (PCI), radiofrequency ablation, coronary artery bypass grafting (CABG), etc. (10) Discharge information: Discharge diagnoses and discharge medications. (11) In-hospital adverse events. Patient follow-up and endpoint events All enrolled patients were followed for a minimum of one year. The primary endpoint of this study was all-cause mortality. All-cause mortality was defined and categorized according to the International Classification of Diseases, Tenth Revision (ICD-10) [ 24 ], including both cardiac and non-cardiac death. Cardiac death was defined as death due to fatal myocardial infarction, sudden cardiac death, or stroke. Non-cardiac death referred to death attributable to causes other than those listed above, including mortality from malignant tumors, infection, respiratory diseases, trauma/accident, or other non-vascular causes. If the cause of death could not be determined based on available information, it was classified as undetermined. Follow-up was conducted by reviewing and recording information from patient re-admission records, performing telephone interviews with patients or their family members, and verifying death certificates by contacting the respective hospital where the death occurred. All death events were independently adjudicated by two members of the event adjudication committee. Statistical methods In this study, descriptive statistics were first employed to summarize the baseline characteristics of the enrolled elderly patients with CAD and AF. For continuous variables, normality was assessed using appropriate statistical tests. Variables with a normal distribution are reported as mean ± standard deviation, while non-normally distributed variables are presented as median with interquartile range. Categorical variables are described using frequencies and percentages. Comparisons for normally distributed continuous variables were made using the independent samples t-test, and the Mann-Whitney U test was used for non-normally distributed continuous variables. To identify predictors significantly associated with all-cause mortality from a comprehensive set of clinical variables, the Least Absolute Shrinkage and Selection Operator (LASSO) method was employed for variable selection. Based on the variables previously selected by LASSO-Cox regression, two predictive models were constructed: (1) Model A: Variables comprising general clinical characteristics + laboratory test indicators. (2) Model B: Variables comprising general clinical characteristics + laboratory test indicators + comorbidity indicators. Nomograms were developed based on the LASSO-Cox regression models to visualize the prediction of 1-year and 5-year survival probabilities. The predictive performance of the LASSO-Cox models and the CHA₂DS₂-VASc score model for all-cause mortality was evaluated and compared using the Area Under the Receiver Operating Characteristic Curve (AUC), calibration curves, and Decision Curve Analysis (DCA). A two-sided p value < 0.05 was considered statistically significant. Results Baseline characteristics of the patients A total of 2,437 elderly patients (≥ 65 years) with CAD and AF, hospitalized in the Department of Cardiology at the Chinese PLA General Hospital between January 2010 and December 2017, were initially enrolled in this study. The mean follow-up duration was 5.4 ± 2.5 years, with a follow-up rate of 93.27%. After excluding 4 cases with unspecified time of death, 1,678 patients were included for developing the clinical prediction model for all-cause mortality risk assessment. These included cases were randomly divided into a training set (n = 1,174) and a validation set (n = 504) in a 7:3 ratio. Except for unstable angina (37.65% vs. 45.63%, P = 0.003), no statistically significant differences were observed in other general clinical data, laboratory findings, or comorbidities between the training and validation sets ( p > 0.05) (Table 1 ). Table 1 Comparison of General Clinical Characteristics in Elderly Patients with CAD and AF Clinical Characteristics Training Set (n = 1174) Validation Set (n = 504) P value Demographics Female sex, n (%) 499 (42.50) 221 (43.85) 0.648 Age (years), mean ± SD 77.28 ± 6.67 77.56 ± 6.82 0.436 BMI (kg/m2), mean ± SD 24.82 ± 3.79 24.76 ± 3.93 0.814 Laboratory Examinations Left ventricular ejection fraction (%), median (Q1, Q3) 57 (52,62) 58 (52.25,63) 0.725 Uric acid (µmol/L), mean ± SD 360.69 ± 121.14 353.91 ± 115.98 0.287 Creatinine (µmol/L), mean ± SD 97.39 ± 58.84 92.83 ± 41.93 0.115 Blood glucose (mmol/L), mean ± SD 6.50 ± 2.56 6.57 ± 2.85 0.653 AST (U/L), mean ± SD 29.82 ± 60.27 28.00 ± 44.23 0.540 ALT (U/L), mean ± SD 26.69 ± 69.33 22.88 ± 29.88 0.235 Plasma fibrinogen (g/L), mean ± SD 3.54 ± 1.05 3.54 ± 0.98 0.923 Plasma D-dimer (µg/mL), mean ± SD 1.24 ± 2.23 1.15 ± 1.85 0.454 Prothrombin time (s), mean ± SD 15.28 ± 6.97 15.09 ± 3.37 0.562 NT-proBNP (pg/mL), M(Q1,Q3) 1202 (47-81.40,2932.25) 1265.50 (569.68,2790.75) 0.621 Platelet count (109/L), mean ± SD 185.48 (62.09) 189.03 (65.20) 0.291 Hemoglobin (g/L), mean ± SD 128.05 ± 19.21 129.38 ± 20.36 0.201 Hematocrit (L/L), mean ± SD 0.38 ± 0.06 0.39 ± 0.06 0.333 Triglycerides (mmol/L), mean ± SD 1.25 ± 0.83 1.19 ± 0.61 0.143 Total cholesterol (mmol/L), mean ± SD 3.83 ± 0.95 3.81 ± 0.92 0.825 LDL (mmol/L), mean ± SD 2.23 ± 0.78 2.26 ± 0.76 0.428 HDL (mmol/L), mean ± SD 1.16 ± 0.35 1.14 ± 0.34 0.256 Type of CAD, n (%) Unstable angina 442 (37.65) 230 (45.63) 0.003 Stable angina 109 (9.28) 35 (6.94) 0.141 Acute myocardial infarction 86 (7.33) 42 (8.33) 0.540 Ischemic cardiomyopathy 18 (1.53) 7 (1.39) 0.997 Type of Atrial Fibrillation, n (%) Persistent atrial fibrillation 220 (18.74) 93 (18.45) 0.944 Paroxysmal atrial fibrillation 554 (47.19) 244 (48.41) 0.684 Comorbidities, n (%) Coronary artery disease 214 (18.23) 90 (17.86) 0.911 Hypertension 886 (75.47) 371 (73.62) 0.457 Dyslipidemia 267 (22.74) 112 (22.22) 0.865 History of myocardial infarction 154 (13.12) 62 (12.30) 0.705 Renal dysfunction 178 (15.16) 67 (13.29) 0.359 Heart failure 452 (38.50) 195 (38.69) 0.985 Diabetes mellitus 354 (30.15) 161 (31.94) 0.502 Stroke 287 (24.45) 118 (23.41) 0.696 Chronic obstructive pulmonary disease 48 (4.09) 17 (3.37) 0.577 Malignant tumor 146 (12.44) 63 (12.50) 1.000 Rheumatic heart disease 44 (3.75) 14 (2.78) 0.395 Valvular heart disease 41 (3.49) 14 (2.78) 0.546 Note : Data are presented as median (interquartile range), mean ± standard deviation, or number (percentage), as appropriate. Abbreviations : AST, aspartate aminotransferase; ALT, alanine aminotransferase; NT-proBNP, N-terminal pro-B-type natriuretic peptide; LDL, low-density lipoprotein; HDL, high-density lipoprotein. Selection of relevant factors and model development Relevant clinical factors associated with all-cause mortality in elderly CAD-AF patients were screened using LASSO-Cox regression. Employing ten-fold cross-validation (Fig. 1 ), the model ultimately retained 17 variables: age, left ventricular ejection fraction, blood glucose, plasma fibrinogen, plasma D-dimer, prothrombin time, NT-proBNP, hemoglobin, hematocrit, acute myocardial infarction, history of myocardial infarction, renal dysfunction, heart failure, diabetes, stroke, chronic obstructive pulmonary disease, and malignant tumors. Using these variables, two models were constructed: LASSO-Cox Model A Included age, left ventricular ejection fraction, blood glucose, plasma fibrinogen, plasma D-dimer, prothrombin time, NT-proBNP, hemoglobin, and hematocrit. LASSO-Cox Model B Included all variables from Model A plus acute myocardial infarction, history of myocardial infarction, renal dysfunction, heart failure, diabetes, stroke, chronic obstructive pulmonary disease, and malignant tumors. The predictive performance of LASSO-Cox Models A and B was described and compared against the CHA₂DS₂-VASc score using the Area Under the Curve (AUC) (Fig. 2 ). For predicting 1-year all-cause mortality risk, LASSO-Cox Model B demonstrated the highest predictive performance in the training set (AUC: 0.85, 95% CI: 0.81–0.89). Furthermore, the predictive ability of the CHA₂DS₂-VASc score model for 1-year all-cause mortality was significantly lower than that of both LASSO-Cox Model A and Model B, in both the training set (AUC: 0.66, 95% CI: 0.62–0.71) and the validation set (AUC: 0.57, 95% CI: 0.48–0.67). Similarly, for predicting 5-year all-cause mortality risk, the predictive ability of the CHA₂DS₂-VASc score model remained significantly inferior to both LASSO-Cox models in both the training set (AUC: 0.62, 95% CI: 0.58–0.66) and the validation set (AUC: 0.57, 95% CI: 0.50–0.63). Since LASSO-Cox Model B exhibited the best predictive performance across both the training and validation sets, it was selected as the final optimal model. Construction of a nomogram for predicting all-cause mortality risk in elderly CAD-AF patients A Cox proportional hazards regression model was developed based on the 17 independent predictors (LASSO-Cox Model B) in the training set to create a nomogram for predicting the 1-year and 5-year all-cause mortality probabilities in elderly patients with CAD and AF (Fig. 3 ). Figure 3 shows that age contributed the most points to the nomogram score, followed by prothrombin time, blood glucose, plasma D-dimer, left ventricular ejection fraction, and other variables. For an individual patient, the points corresponding to each predictor can be obtained from the nomogram. The sum of these points yields a total score, which can then be converted into the predicted probability of 1-year and 5-year survival for elderly patients with CAD and AF. Evaluation of model performance for predicting all-cause mortality in elderly CAD-AF patients The calibration curves of the predictive models are shown in Fig. 4 . For predicting 1-year all-cause mortality in the training set, all three models demonstrated good agreement with observed outcomes when the predicted survival probability was above 0.9. In the validation set for 1-year all-cause mortality, good agreement was observed across all three models when the predicted survival probability exceeded 0.95. All models also showed favorable calibration for predicting 5-year all-cause mortality in both datasets. DCA revealed that compared to the CHA₂DS₂-VASc score model, the LASSO-Cox machine learning models provided a greater net benefit for predicting both 1-year and 5-year all-cause mortality risk (Fig. 5 ). Discussion This study found that a Cox proportional hazards regression model, developed via the LASSO-Cox algorithm and incorporating 17 clinical risk factors, accurately predicted the risk of all-cause mortality in elderly patients with coexisting CAD and AF. Compared to the CHA₂DS₂-VASc score model, the prediction model established in this study via the LASSO-Cox machine learning algorithm demonstrated superior discriminatory ability and provided a greater net clinical benefit in both the training and validation sets. Currently, there is a lack of machine learning-based prediction models for all-cause mortality risk specifically targeted at elderly patients aged 65 and above with concurrent CAD and AF. This research serves as a valuable reference for machine learning-based risk assessment of all-cause mortality in this specific patient population. The prediction model developed in this study incorporates three categories of clinical indicators: general patient characteristics, laboratory test results, and patient comorbidities. Our results identified age, left ventricular ejection fraction, blood glucose, plasma fibrinogen, plasma D-dimer, hemoglobin, hematocrit, history of myocardial infarction, history of heart failure, and renal dysfunction as risk factors for all-cause mortality in patients with coexisting CAD and AF. These findings share common ground with the model proposed by Dong Min et al. However, our model additionally includes prothrombin time, NT-proBNP, blood glucose, diabetes, stroke, chronic obstructive pulmonary disease, and malignant tumors as risk factors influencing all-cause mortality in elderly patients with CAD and AF. Among these, NT-proBNP emerged as the most significant laboratory predictor for all-cause mortality and adverse cardiovascular events. NT-proBNP concentration is associated with left ventricular filling pressure and wall stress [ 25 – 27 ]. Its levels increase during left ventricular dysfunction and acute myocardial infarction, and it counteracts heart failure through diuretic, natriuretic, and antihypertensive effects [ 28 , 29 ]. However, the lack of a specific threshold hinders its application in clinical risk prediction [ 30 ]. Therefore, our prediction model provides a more effective tool for utilizing NT-proBNP in assessing all-cause mortality risk in elderly patients with CAD and AF. A Swedish study indicated that malignant tumors and renal failure can increase the risk of all-cause mortality in AF patients [ 31 ]. Elevated blood glucose has also been reported to be associated with adverse cardiovascular events [ 32 , 33 ]. The present study not only corroborates these established associations but further quantifies their predictive value within a dedicated model for elderly patients with CAD and AF. Specifically, variables representing these conditions: malignant tumor, renal dysfunction, and blood glucose were consistently selected into our LASSO-Cox model (Model B) as significant predictors. It underscores that our model successfully captures and integrates these critical laboratory and comorbidity-based risk dimensions. Consequently, greater attention should be paid to laboratory findings and comorbidities in elderly patients with CAD and AF. Early diagnosis and timely intervention targeting these factors are crucial for improving mortality outcomes and prolonging survival. In a study of 2,174 ACS patients aged 50 and above, Li et al. utilized LASSO regression to select six variables and subsequently developed a Cox regression model to predict 3-year all-cause mortality. Their results showed that this model achieved a predictive performance (AUC) of 0.768 and a calibration of 0.711, both of which were superior to those of the traditional GRACE model (AUC: 0.701; calibration: 0.203) [ 34 ]. Similarly, Sun et al., working with 468 heart failure patients over 50, applied LASSO regression to select 26 variables and then employed four machine learning algorithms—Extreme Gradient Boosting (XGBoost), Random Survival Forest, Multi-layer Perceptron, and Support Vector Machine—to build models predicting the risk of Major Adverse Cardiovascular Events (MACEs) within six months post-discharge. They found that the model built using the XGBoost algorithm performed best, with an AUC of 0.84 and a calibration of 0.77 [ 35 ]. These studies collectively affirm the utility of LASSO-based variable selection combined with regression or machine learning modeling in cardiovascular prognosis. Our study extends this methodological approach to a distinct and clinically complex cohort—elderly patients with CAD and AF—a population at notably high risk yet lacking a dedicated prognostic tool. By applying LASSO-Cox regression, we not only confirmed the feasibility of this approach in a multimorbid setting but also developed a model (Model B) that demonstrated robust predictive performance (AUC up to 0.85 for 1-year mortality), thereby providing a tailored risk assessment instrument for this patient group. The Cox proportional hazards model developed in this study demonstrated significantly superior discriminative ability for all-cause mortality compared to the conventional CHA₂DS₂-VASc score in elderly patients with coexisting CAD and AF. Specifically, our model exhibited robust predictive performance, with an AUC of 0.85 for 1-year mortality in the training set, and maintained stable, clinically relevant accuracy in the independent validation set (1-year AUC: 0.80; sustained AUC > 0.75 during long-term follow-up). This performance markedly outperformed the CHA₂DS₂-VASc score, which showed limited discriminative power in the same cohort (AUCs between 0.57 and 0.66). More importantly, decision curve analysis confirmed that our model, which integrates over ten clinical risk factors selected via the LASSO-Cox algorithm, provides a greater net clinical benefit across a wide range of risk thresholds. These findings collectively indicate that a multivariate model incorporating diverse clinical, laboratory, and comorbidity data offers a more accurate and clinically useful risk stratification tool for this high-risk, complex population than the traditional score-based approach. Limitations This study has several limitations. First, the types of variables available in the clinical follow-up database for model selection were limited. A lack of structured information, such as details regarding surgical interventions and imaging data, restricted the inclusion of other predictors that might significantly influence clinical outcomes. Secondly, the study exclusively employed the LASSO-Cox machine learning algorithm for model construction. The randomness in variable selection can differ across various machine learning algorithms. Therefore, future work should involve comparing models built using other algorithms, including Random Forest, to further validate and refine our approach. Furthermore, as this was a single-center, retrospective study, subsequent external validation using cases from other institutions is necessary to verify the model's predictive performance and generalizability. Conclusion The Cox proportional hazards regression model established based on the machine learning algorithm demonstrated superior predictive ability for all-cause mortality risk in elderly patients with CAD and AF, showing both higher predictive performance and greater net benefit compared to the CHA₂DS₂-VASc score model. Declarations Ethics approval and consent to participate The study was approved by the Ethics Committee of Chinese PLA General Hospital (S2024-442-02). Informed consent was obtained from all subjects. Data Avaliability Date is available upon author‘s request. Conflicts of interest No potential conflict of interest was reported by the author(s). Funding and acknowledgements This research was financially supported by grants from the National Natural Science Foundation of China (No. 81870262, 82170352). Author Contribution Yuyan Wang: Conceptualization, Methodology, Formal analysis, Writing – Original Draft.Yangxun Wu: Data Curation, Software, Validation.Yuting Zou: Investigation, Resources.Shuai Mao: Visualization, Writing – Review & Editing.Yanzhu Yao: Investigation, Data Curation.Yifan Wang: Methodology, Validation.Yunzhang Zhao: Formal analysis, Software.Chao Lv: Resources, Project Administration.Li Fan: Supervision, Writing – Review & Editing, Funding Acquisition.Tong Yin: Supervision, Writing – Review & Editing, Funding Acquisition, Project Administration. References Fadah K, Hechanova A, Mukherjee D. Epidemiology, Pathophysiology, and Management of Coronary Artery Disease in the Elderly. Int J Angiol. 2022;31(4):244–50. Zhang J, et al. Epidemiology of Atrial Fibrillation: Geographic/Ecological Risk Factors, Age, Sex, Genetics. Card Electrophysiol Clin. 2021;13(1):1–23. Steinhubl SR, et al. Early and sustained dual oral antiplatelet therapy following percutaneous coronary intervention: a randomized controlled trial. JAMA. 2002;288(19):2411–20. Steensig K, et al. CAD Is an Independent Risk Factor for Stroke Among Patients With Atrial Fibrillation. J Am Coll Cardiol. 2018;72(20):2540–2. Saglietto A, et al. Prognostic implications of atrial fibrillation in patients with stable coronary artery disease: a systematic review and meta-analysis of adjusted observational studies. Rev Cardiovasc Med. 2021;22(2):439–44. Petersen JK, et al. Incidence of ischaemic stroke and mortality in patients with acute coronary syndrome and first-time detected atrial fibrillation: a nationwide study. Eur Heart J. 2021;42(44):4553–61. Wu Y, et al. Assessment of the CHA(2)DS(2)-VASc Score for the Prediction of Death in Elderly Patients With Coronary Artery Disease and Atrial Fibrillation. Front Cardiovasc Med. 2021;8:805234. Hindricks G et al. 2020 ESC Guidelines for the diagnosis and management of atrial fibrillation developed in collaboration with the European Association for Cardio-Thoracic Surgery (EACTS): The Task Force for the diagnosis and management of atrial fibrillation of the European Society of Cardiology (ESC) Developed with the special contribution of the European Heart Rhythm Association (EHRA) of the ESC. Eur Heart J, 2021. 42(5): pp. 373–498. Kim D, et al. Long-Term Anticoagulation Discontinuation After Catheter Ablation for Atrial Fibrillation: The ALONE-AF Randomized Clinical Trial. JAMA. 2025;334(14):1246–54. Shuvy M, et al. Value of the CHA(2) DS(2) -VASc score for predicting outcome in patients with heart failure. ESC Heart Fail. 2020;7(5):2553–60. Gök M, et al. Relationship Between CHA2DS2-VASc Score and Right Ventricular Dysfunction in Patients With Acute Pulmonary Thromboembolism. Clin Appl Thromb Hemost. 2018;24(9suppl):s56–62. Li CY, et al. Assessment of CHA2DS2-VASc score for predicting cardiovascular and cerebrovascular outcomes in acute myocardial infarction patients. Med (Baltim). 2018;97(28):e11230. Topaz G, et al. CHA(2) DS(2) -VASc score and clinical outcomes of patients with chest pain discharged from internal medicine wards following acute coronary syndrome rule-out. Clin Cardiol. 2018;41(4):539–43. Ma X, et al. Prognostic value of CHADS2 and CHA2DS2-VASc scores for post-discharge outcomes in patients with acute coronary syndrome undergoing percutaneous coronary intervention. Med (Baltim). 2020;99(30):e21321. Andreotti F, et al. Acute, periprocedural and longterm antithrombotic therapy in older adults: 2022 Update by the ESC Working Group on Thrombosis. Eur Heart J. 2023;44(4):262–79. Chen Y, et al. Predicting All-Cause Mortality Risk in Atrial Fibrillation Patients: A Novel LASSO-Cox Model Generated From a Prospective Dataset. Front Cardiovasc Med. 2021;8:730453. Hill NR, et al. Predicting atrial fibrillation in primary care using machine learning. PLoS ONE. 2019;14(11):e0224582. Lanfear DE, et al. Targeted Metabolomic Profiling of Plasma and Survival in Heart Failure Patients. JACC Heart Fail. 2017;5(11):823–32. Mavaddat N, et al. Polygenic Risk Scores for Prediction of Breast Cancer and Breast Cancer Subtypes. Am J Hum Genet. 2019;104(1):21–34. Vrints C, et al. 2024 ESC Guidelines for the management of chronic coronary syndromes. Eur Heart J. 2024;45(36):3415–537. Lim M, Hastie T. Learning interactions via hierarchical group-lasso regularization. J Comput Graph Stat. 2015;24(3):627–54. Knuuti J, et al. 2019 ESC Guidelines for the diagnosis and management of chronic coronary syndromes. Eur Heart J. 2020;41(3):407–77. Lip GYH et al. 2018 Joint European consensus document on the management of antithrombotic therapy in atrial fibrillation patients presenting with acute coronary syndrome and/or undergoing percutaneous cardiovascular interventions: a joint consensus document of the European Heart Rhythm Association (EHRA), European Society of Cardiology Working Group on Thrombosis, European Association of Percutaneous Cardiovascular Interventions (EAPCI), and European Association of Acute Cardiac Care (ACCA) endorsed by the Heart Rhythm Society (HRS), Asia-Pacific Heart Rhythm Society (APHRS), Latin America Heart Rhythm Society (LAHRS), and Cardiac Arrhythmia Society of Southern Africa (CASSA). Europace, 2019. 21(2): pp. 192–193. ICD-10: there's a code for that. Lancet, 2015. 386(10002): p. 1420. Nishi H, et al. N-terminal pro-brain natriuretic peptide and cardiorenal outcome in patients with anaemia in chronic kidney disease. ESC Heart Fail. 2025;12(2):848–58. Januzzi JL, Mebazaa A, Di Somma S. ST2 and prognosis in acutely decompensated heart failure: the International ST2 Consensus Panel. Am J Cardiol. 2015;115(7 Suppl):b26–31. Januzzi JL Jr., et al. Measurement of the interleukin family member ST2 in patients with acute dyspnea: results from the PRIDE (Pro-Brain Natriuretic Peptide Investigation of Dyspnea in the Emergency Department) study. J Am Coll Cardiol. 2007;50(7):607–13. Booth RA, et al. Performance of BNP and NT-proBNP for diagnosis of heart failure in primary care patients: a systematic review. Heart Fail Rev. 2014;19(4):439–51. Januzzi JL, et al. NT-proBNP testing for diagnosis and short-term prognosis in acute destabilized heart failure: an international pooled analysis of 1256 patients: the International Collaborative of NT-proBNP Study. Eur Heart J. 2006;27(3):330–7. Li X, et al. Plasma NT pro-BNP, hs-CRP and big-ET levels at admission as prognostic markers of survival in hospitalized patients with dilated cardiomyopathy: a single-center cohort study. BMC Cardiovasc Disord. 2014;14:67. Andersson T, et al. All-cause mortality in 272,186 patients hospitalized with incident atrial fibrillation 1995–2008: a Swedish nationwide long-term case-control study. Eur Heart J. 2013;34(14):1061–7. Yatsuya H, et al. Risk Factor of Cardiovascular Disease Among Older Individuals. J Atheroscler Thromb. 2017;24(3):258–61. Liu X, et al. Relationship between the triglyceride-glucose index and risk of cardiovascular diseases and mortality in the general population: a systematic review and meta-analysis. Cardiovasc Diabetol. 2022;21(1):124. Li YM, et al. A LASSO-derived risk model for long-term mortality in Chinese patients with acute coronary syndrome. J Transl Med. 2020;18(1):157. Sun Q, et al. A prediction model for major adverse cardiovascular events in patients with heart failure based on high-throughput echocardiographic data. Front Cardiovasc Med. 2022;9:1022658. Additional Declarations No competing interests reported. Cite Share Download PDF Status: Posted Version 1 posted 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-8337799\",\"acceptedTermsAndConditions\":true,\"allowDirectSubmit\":true,\"archivedVersions\":[],\"articleType\":\"Research Article\",\"associatedPublications\":[],\"authors\":[{\"id\":572451839,\"identity\":\"2ad7de4d-687a-4dba-bebf-08647707fb6f\",\"order_by\":0,\"name\":\"Yuyan Wang\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Chinese PLA General Hospital\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Yuyan\",\"middleName\":\"\",\"lastName\":\"Wang\",\"suffix\":\"\"},{\"id\":572451844,\"identity\":\"b5dbfe40-f7d0-457f-acba-8bb3a891b6a5\",\"order_by\":1,\"name\":\"Yangxun Wu\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Chinese PLA General Hospital\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Yangxun\",\"middleName\":\"\",\"lastName\":\"Wu\",\"suffix\":\"\"},{\"id\":572451845,\"identity\":\"429c14a3-b144-4e43-b769-8e72ab1795f3\",\"order_by\":2,\"name\":\"Yuting Zou\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Chinese PLA General Hospital\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Yuting\",\"middleName\":\"\",\"lastName\":\"Zou\",\"suffix\":\"\"},{\"id\":572451846,\"identity\":\"273699a8-069f-49a3-bf40-aa7c3b00fe65\",\"order_by\":3,\"name\":\"Shuai Mao\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Chinese PLA General Hospital\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Shuai\",\"middleName\":\"\",\"lastName\":\"Mao\",\"suffix\":\"\"},{\"id\":572451848,\"identity\":\"4d6e2b81-9f2b-4d57-952b-0285cedf17ab\",\"order_by\":4,\"name\":\"Yanzhu Yao\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Chinese PLA General Hospital\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Yanzhu\",\"middleName\":\"\",\"lastName\":\"Yao\",\"suffix\":\"\"},{\"id\":572451850,\"identity\":\"12b0966c-0fa6-4911-9b61-cbe502095271\",\"order_by\":5,\"name\":\"Yifan Wang\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Chinese PLA General Hospital\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Yifan\",\"middleName\":\"\",\"lastName\":\"Wang\",\"suffix\":\"\"},{\"id\":572451851,\"identity\":\"444b4c8e-73a6-4cba-bd9f-1a25aa16e594\",\"order_by\":6,\"name\":\"Yunzhang Zhao\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Chinese PLA General Hospital\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Yunzhang\",\"middleName\":\"\",\"lastName\":\"Zhao\",\"suffix\":\"\"},{\"id\":572451852,\"identity\":\"ee5b1178-6589-4326-ada9-7d9bf5a6338d\",\"order_by\":7,\"name\":\"Chao Lv\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Chinese PLA General Hospital\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Chao\",\"middleName\":\"\",\"lastName\":\"Lv\",\"suffix\":\"\"},{\"id\":572451853,\"identity\":\"e26b0c66-d4e2-4d55-ae77-30cf5a04e3bb\",\"order_by\":8,\"name\":\"Li Fan\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Chinese PLA General Hospital\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Li\",\"middleName\":\"\",\"lastName\":\"Fan\",\"suffix\":\"\"},{\"id\":572451854,\"identity\":\"cade0e00-ebd8-4eda-80d1-8cb49a8b12f0\",\"order_by\":9,\"name\":\"Tong Yin\",\"email\":\"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA0UlEQVRIiWNgGAWjYDACZghlx8bMfODAhx/EaWFsAFLJ/OxsiQdn9hBnD1gL48x+HuPDHGxEqDc4znv8wccdtcwGh3k+HGbgYZDnFztAQMthvsTGmWeO8xkc5t1wuMCCwXDm7ARCWngMm3nbjjGDtczgYUgwuE2kFsYNh3keHOZhI15LDePMZh4G4rRIArXMnNl2IJmfmc0AGMgShP3Cd/6MwYePbXV2bPyHH3/48MNGnl+agBaFA2DqMIwvgV85CMg3gKk6wipHwSgYBaNg5AIAEAxHIKEWXX8AAAAASUVORK5CYII=\",\"orcid\":\"\",\"institution\":\"Chinese PLA General Hospital\",\"correspondingAuthor\":true,\"prefix\":\"\",\"firstName\":\"Tong\",\"middleName\":\"\",\"lastName\":\"Yin\",\"suffix\":\"\"}],\"badges\":[],\"createdAt\":\"2025-12-11 14:38:35\",\"currentVersionCode\":1,\"declarations\":\"\",\"doi\":\"10.21203/rs.3.rs-8337799/v1\",\"doiUrl\":\"https://doi.org/10.21203/rs.3.rs-8337799/v1\",\"draftVersion\":[],\"editorialEvents\":[],\"editorialNote\":\"\",\"failedWorkflow\":false,\"files\":[{\"id\":100367631,\"identity\":\"a388e4ab-7118-4dce-8137-9baecf9802e9\",\"added_by\":\"auto\",\"created_at\":\"2026-01-16 07:57:11\",\"extension\":\"docx\",\"order_by\":0,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":1028119,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"Figures.docx\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8337799/v1/15ed062f9764bdf26e3bc2b3.docx\"},{\"id\":100124935,\"identity\":\"e16e4a1e-bc99-4b8d-bb58-0db99a4a1e6b\",\"added_by\":\"auto\",\"created_at\":\"2026-01-13 09:16:32\",\"extension\":\"docx\",\"order_by\":1,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":20011,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"Tables.docx\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8337799/v1/53a03e5e06b05c95e423b9e9.docx\"},{\"id\":100366019,\"identity\":\"2ce34f38-7366-4cfb-8834-8a888edf7476\",\"added_by\":\"auto\",\"created_at\":\"2026-01-16 07:55:52\",\"extension\":\"docx\",\"order_by\":2,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":99567,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"manuscript20251212.docx\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8337799/v1/e2c9faebc4227aacab067c99.docx\"},{\"id\":100124956,\"identity\":\"22171084-9aef-4b74-ae37-88b5bfc81364\",\"added_by\":\"auto\",\"created_at\":\"2026-01-13 09:16:33\",\"extension\":\"json\",\"order_by\":3,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":11841,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"b112a57a862c47638be1a19da20aae86.json\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8337799/v1/7e540067f7f21623fcaa82b7.json\"},{\"id\":100365956,\"identity\":\"471a082d-b262-4811-a365-85501ec0a463\",\"added_by\":\"auto\",\"created_at\":\"2026-01-16 07:55:47\",\"extension\":\"xml\",\"order_by\":4,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":98142,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"b112a57a862c47638be1a19da20aae861enriched.xml\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8337799/v1/2a4e852c28ce145fb28d8f36.xml\"},{\"id\":100124939,\"identity\":\"6d4f9959-6535-4ff0-a6f6-b2d3ebbc3a33\",\"added_by\":\"auto\",\"created_at\":\"2026-01-13 09:16:32\",\"extension\":\"png\",\"order_by\":5,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":79619,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"floatimage1.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8337799/v1/842c801bc6ac2a7d00160e62.png\"},{\"id\":100124942,\"identity\":\"9d900e8d-bb73-4372-b76d-b1cb2ea20ef4\",\"added_by\":\"auto\",\"created_at\":\"2026-01-13 09:16:32\",\"extension\":\"png\",\"order_by\":6,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":125866,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"floatimage2.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8337799/v1/5eb23395bed6da3c57dce6e0.png\"},{\"id\":100124953,\"identity\":\"699de1f1-d321-4888-bce6-f61cd8c63b7b\",\"added_by\":\"auto\",\"created_at\":\"2026-01-13 09:16:32\",\"extension\":\"jpeg\",\"order_by\":7,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":469302,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"floatimage3.jpeg\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8337799/v1/46e8446fcf92fadbacc6425f.jpeg\"},{\"id\":100366446,\"identity\":\"adbf5476-dd03-4d50-8db2-427d3fa6f298\",\"added_by\":\"auto\",\"created_at\":\"2026-01-16 07:56:18\",\"extension\":\"png\",\"order_by\":8,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":474911,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"floatimage4.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8337799/v1/b2de49f847608104f9f62aaa.png\"},{\"id\":100124950,\"identity\":\"7e9d9e2a-35b0-4fdb-88c6-571cfa09cf41\",\"added_by\":\"auto\",\"created_at\":\"2026-01-13 09:16:32\",\"extension\":\"png\",\"order_by\":9,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":152018,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"floatimage5.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8337799/v1/0e19937f2b151fa9a3113824.png\"},{\"id\":100367817,\"identity\":\"5424e6a3-1d34-4e41-a40f-3e79b27b3622\",\"added_by\":\"auto\",\"created_at\":\"2026-01-16 07:57:21\",\"extension\":\"png\",\"order_by\":10,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":18305,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"Onlinefloatimage1.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8337799/v1/8fb3fd4b5ff5cd31844dd71b.png\"},{\"id\":100124946,\"identity\":\"e80b26e1-16ed-4592-beab-6f88de2569b1\",\"added_by\":\"auto\",\"created_at\":\"2026-01-13 09:16:32\",\"extension\":\"png\",\"order_by\":11,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":29886,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"Onlinefloatimage2.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8337799/v1/3745cab2b2a9cd85d4c55eff.png\"},{\"id\":100124944,\"identity\":\"374c1f77-9bcc-4ecb-841d-87db70e945f0\",\"added_by\":\"auto\",\"created_at\":\"2026-01-13 09:16:32\",\"extension\":\"png\",\"order_by\":12,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":98943,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"Onlinefloatimage3.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8337799/v1/350538c9a634c86c7eda395b.png\"},{\"id\":100124952,\"identity\":\"ff826b4d-187a-4ab1-9e04-30b284d9f2c7\",\"added_by\":\"auto\",\"created_at\":\"2026-01-13 09:16:32\",\"extension\":\"png\",\"order_by\":13,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":59977,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"Onlinefloatimage4.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8337799/v1/80a2e4f11593f230ec4b6be7.png\"},{\"id\":100367827,\"identity\":\"34adef82-62c5-4d43-94cf-988037358cf3\",\"added_by\":\"auto\",\"created_at\":\"2026-01-16 07:57:21\",\"extension\":\"png\",\"order_by\":14,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":36687,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"Onlinefloatimage5.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8337799/v1/cedb58eaf706e1ee1ce75c2f.png\"},{\"id\":100365944,\"identity\":\"223ff9bc-ca60-4d2c-9e3c-6026cccda3af\",\"added_by\":\"auto\",\"created_at\":\"2026-01-16 07:55:46\",\"extension\":\"xml\",\"order_by\":15,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":94201,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"b112a57a862c47638be1a19da20aae861structuring.xml\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8337799/v1/0addf0ad3e039924072d76a1.xml\"},{\"id\":100124949,\"identity\":\"13383bef-14a4-4b59-9bdd-9effebd0c6a5\",\"added_by\":\"auto\",\"created_at\":\"2026-01-13 09:16:32\",\"extension\":\"html\",\"order_by\":16,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":107245,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"earlyproof.html\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8337799/v1/926629ee9565b4599d859546.html\"},{\"id\":100366558,\"identity\":\"aa2f9f49-72ae-49a4-a3b9-ba1a25ce3c68\",\"added_by\":\"auto\",\"created_at\":\"2026-01-16 07:56:21\",\"extension\":\"png\",\"order_by\":1,\"title\":\"Figure 1\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":120018,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003eSee image above for figure legend\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"1.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8337799/v1/ce0c56fb0262455f848ebb67.png\"},{\"id\":100124933,\"identity\":\"86c4155d-d700-4112-8ae4-561f71223ed4\",\"added_by\":\"auto\",\"created_at\":\"2026-01-13 09:16:32\",\"extension\":\"png\",\"order_by\":2,\"title\":\"Figure 2\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":166055,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003eSee image above for figure legend\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"2.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8337799/v1/a256e528e31775172a45a569.png\"},{\"id\":100124937,\"identity\":\"62fcde2c-6fe2-47e9-92f7-26a170f70dc6\",\"added_by\":\"auto\",\"created_at\":\"2026-01-13 09:16:32\",\"extension\":\"png\",\"order_by\":3,\"title\":\"Figure 3\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":243755,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003eSee image above for figure legend\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"3.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8337799/v1/dcc2ef7a0083c9660f3159b9.png\"},{\"id\":100124945,\"identity\":\"441d9f22-b945-4670-a5d3-39c012dce24e\",\"added_by\":\"auto\",\"created_at\":\"2026-01-13 09:16:32\",\"extension\":\"png\",\"order_by\":4,\"title\":\"Figure 4\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":441334,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003eSee image above for figure legend\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"4.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8337799/v1/4ff5069b0020e04c49dfd6c9.png\"},{\"id\":100125005,\"identity\":\"ea4ffa58-b79c-419a-bcb1-17537d371f6e\",\"added_by\":\"auto\",\"created_at\":\"2026-01-13 09:16:46\",\"extension\":\"png\",\"order_by\":5,\"title\":\"Figure 5\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":177788,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003eSee image above for figure legend\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"5.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8337799/v1/b8775142a245c81b6486810b.png\"},{\"id\":107705358,\"identity\":\"05e3250c-bae1-4a44-a209-cf109cba8574\",\"added_by\":\"auto\",\"created_at\":\"2026-04-24 09:11:51\",\"extension\":\"pdf\",\"order_by\":0,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"manuscript-pdf\",\"size\":1255216,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"manuscript.pdf\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8337799/v1/b26c19cd-b444-42b8-b034-77a70c02c878.pdf\"}],\"financialInterests\":\"No competing interests reported.\",\"formattedTitle\":\"Mortality Prediction in the elderly patients with Coronary Artery Disease and Atrial Fibrillation: A Retrospective Machine Learning Approach\",\"fulltext\":[{\"header\":\"Introduction\",\"content\":\"\\u003cp\\u003eCoronary artery disease (CAD) and atrial fibrillation (AF) are highly prevalent cardiovascular diseases in China, characterized by an increasing incidence with advancing age [\\u003cspan citationid=\\\"CR1\\\" class=\\\"CitationRef\\\"\\u003e1\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR2\\\" class=\\\"CitationRef\\\"\\u003e2\\u003c/span\\u003e]. CAD is a leading cause of mortality in both developing and developed countries. As an age-related arrhythmia, AF is associated with a significantly elevated risk of death, with approximately half of all deaths in AF patients being attributable to cardiac causes [\\u003cspan citationid=\\\"CR3\\\" class=\\\"CitationRef\\\"\\u003e3\\u003c/span\\u003e]. CAD and AF share common risk factors and pathophysiological features and therefore often coexist. The population aged 65 and above in China has risen to 264\\u0026nbsp;million, accounting for 18.70% of the total population. The prevalence of comorbidities in the elderly, including these conditions, is increasing along with the accelerating aging of the population. Patients with concomitant CAD and AF have a significantly higher rate of adverse cardiovascular events compared to those with either condition alone [\\u003cspan citationid=\\\"CR4\\\" class=\\\"CitationRef\\\"\\u003e4\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR5\\\" class=\\\"CitationRef\\\"\\u003e5\\u003c/span\\u003e]. Furthermore, among elderly patients (\\u0026ge;\\u0026thinsp;70 years), both the co-prevalence of these diseases and the risk of adverse clinical outcomes are markedly higher than in younger patients [\\u003cspan citationid=\\\"CR6\\\" class=\\\"CitationRef\\\"\\u003e6\\u003c/span\\u003e]. The combination of CAD and AF in the elderly is associated with higher mortality [\\u003cspan citationid=\\\"CR7\\\" class=\\\"CitationRef\\\"\\u003e7\\u003c/span\\u003e], posing a severe threat to life and health. Consequently, predicting the all-cause mortality risk in elderly patients with CAD and AF is of paramount importance for clinical management.\\u003c/p\\u003e \\u003cp\\u003eThe CHA₂DS₂-VASc score model is the most widely used clinical prediction tool for assessing thromboembolic risk and guiding antithrombotic therapy in AF patients [\\u003cspan citationid=\\\"CR8\\\" class=\\\"CitationRef\\\"\\u003e8\\u003c/span\\u003e]. This scoring system can also predict the risk of adverse outcomes such as all-cause mortality and major adverse cardiac and cerebrovascular events [\\u003cspan additionalcitationids=\\\"CR10 CR11 CR12 CR13\\\" citationid=\\\"CR9\\\" class=\\\"CitationRef\\\"\\u003e9\\u003c/span\\u003e\\u0026ndash;\\u003cspan citationid=\\\"CR14\\\" class=\\\"CitationRef\\\"\\u003e14\\u003c/span\\u003e]. However, it has certain limitations in predicting adverse clinical outcomes specifically in elderly patients with coexisting CAD and AF [\\u003cspan citationid=\\\"CR7\\\" class=\\\"CitationRef\\\"\\u003e7\\u003c/span\\u003e]. The 2022 European Society of Cardiology guidelines explicitly recommend that a systematic assessment of thrombotic risk in elderly patients with CAD or AF should integrate various types of risk predictors [\\u003cspan citationid=\\\"CR15\\\" class=\\\"CitationRef\\\"\\u003e15\\u003c/span\\u003e]. Due to the widespread adoption of machine learning algorithms and their significant advantages in predictive performance, several studies have applied these algorithms to develop clinical models for predicting all-cause death in patients with either CAD or AF, demonstrating superior predictive value compared to traditional clinical prediction models. For instance, Chen et al. utilized LASSO regression and the random survival forest algorithm to select variables from 1,223 AF patients aged 55 and above. They employed six machine learning algorithms, including Cox regression, to build models predicting 1-year all-cause mortality. Their results indicated that the model using LASSO regression for variable selection performed the best, exhibiting higher predictive power and clinical net benefit than the traditional CHA₂DS₂-VASc model [\\u003cspan citationid=\\\"CR16\\\" class=\\\"CitationRef\\\"\\u003e16\\u003c/span\\u003e]. Consequently, using LASSO regression to select clinical risk factors combined with machine learning algorithms to build all-cause mortality risk models has been proven effective and accurate in cardiovascular disease patients aged 50 and above, characterized by CAD or AF, and is significantly superior to traditional clinical prediction models. Among these approaches, combining variable selection via LASSO regression with a Cox proportional hazards model has been shown to achieve optimal results. LASSO regression is favored for variable selection due to its high accuracy and stability, and is commonly used in modeling for cancer, heart failure, and AF populations [\\u003cspan additionalcitationids=\\\"CR18\\\" citationid=\\\"CR17\\\" class=\\\"CitationRef\\\"\\u003e17\\u003c/span\\u003e\\u0026ndash;\\u003cspan citationid=\\\"CR19\\\" class=\\\"CitationRef\\\"\\u003e19\\u003c/span\\u003e]. Furthermore, the LASSO-Cox regression method builds upon LASSO regression by estimating the relationship between the selected variables and the time to endpoint events. It serves as a method to prevent model overfitting and has been widely incorporated into various types of machine learning algorithms [\\u003cspan citationid=\\\"CR20\\\" class=\\\"CitationRef\\\"\\u003e20\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR21\\\" class=\\\"CitationRef\\\"\\u003e21\\u003c/span\\u003e].\\u003c/p\\u003e \\u003cp\\u003eCurrently, there is a lack of established risk prediction models specifically for all-cause mortality in the elderly population with concomitant CAD and AF. Therefore, this study aims to address this gap by establishing a clinical follow-up cohort database and leveraging advanced big data analytics and machine learning model algorithms. It seeks to thoroughly investigate the performance of a risk assessment tool for all-cause mortality in this specific patient group. The development of such a model holds significant clinical utility and practical value for improving the prognosis of elderly patients with CAD and AF.\\u003c/p\\u003e\"},{\"header\":\"Materials and methods\",\"content\":\"\\u003cdiv id=\\\"Sec3\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003eStudy population\\u003c/h2\\u003e \\u003cp\\u003eA consecutive cohort of elderly patients diagnosed with both CAD and AF during hospitalization in the Department of Cardiology, Chinese People's Liberation Army (PLA) General Hospital, between January 2010 and December 2017, was enrolled. The patients were randomly divided into a training set and a validation set at a 7:3 ratio. During the dataset division, stratified sampling was employed to ensure the proportion of samples with and without the endpoint event was identical between the training and validation sets. \\u003cb\\u003eInclusion criteria\\u003c/b\\u003e: (1)Age\\u0026thinsp;\\u0026ge;\\u0026thinsp;65 years; (2) Clinical diagnosis of CAD combined with AF. \\u003cb\\u003eExclusion criteria\\u003c/b\\u003e: (1) Inability to undergo follow-up for at least one year; (2) Missing crucial information in electronic medical records (including data on blood biochemistry and past medical history); (3) Participation in other clinical trials; (4) Terminal patients during the hospitalization period. This study was reviewed and approved by the Ethics Committee of the Chinese PLA General Hospital. Informed consent was obtained from all enrolled patients.\\u003c/p\\u003e \\u003c/div\\u003e\\n\\u003ch3\\u003eDefinition of disease\\u003c/h3\\u003e\\n\\u003cp\\u003eCAD was defined as stable coronary artery disease (including stable angina pectoris, prior myocardial infarction, and ischemic cardiomyopathy) [\\u003cspan citationid=\\\"CR22\\\" class=\\\"CitationRef\\\"\\u003e22\\u003c/span\\u003e] or acute coronary syndrome (including unstable angina and acute myocardial infarction) [\\u003cspan citationid=\\\"CR23\\\" class=\\\"CitationRef\\\"\\u003e23\\u003c/span\\u003e]. AF was defined as a cardiac rhythm characterized by irregular, discrete P waves and irregular RR intervals (in the absence of atrioventricular block) recorded by a standard 12-lead electrocardiogram or a single-lead rhythm strip lasting\\u0026thinsp;\\u0026gt;\\u0026thinsp;30 seconds [\\u003cspan citationid=\\\"CR8\\\" class=\\\"CitationRef\\\"\\u003e8\\u003c/span\\u003e], encompassing paroxysmal, persistent, or permanent AF.\\u003c/p\\u003e\\n\\u003ch3\\u003eCollection of clinical information\\u003c/h3\\u003e\\n\\u003cp\\u003eClinical data of the enrolled subjects were obtained from inpatient medical records or outpatient systems. The collected information included:(1) Demographics: Sex, age, etc. (2) Lifestyle factors: History of smoking and alcohol consumption, etc. (3) Past medical history and comorbidities: History of renal dysfunction, stroke, chronic obstructive pulmonary disease (COPD), peripheral arterial disease, malignant tumors, etc. (4) Previous medication history: History of cardiovascular medication use, etc. (5) CAD type: Stable angina, acute coronary syndrome (ACS) (including unstable angina, acute ST-segment elevation myocardial infarction, acute non-ST-segment elevation myocardial infarction), etc. (6) AF type: Paroxysmal AF, persistent AF, permanent AF. (7) Concomitant medications: Antiplatelet agents, anticoagulants, beta-blockers, calcium channel blockers, statins, amiodarone, etc. (8) Laboratory tests and examinations: Complete blood count, liver and kidney function tests, coagulation profile, left ventricular ejection fraction (LVEF), etc. (9) Interventional procedures and surgeries: Percutaneous coronary intervention (PCI), radiofrequency ablation, coronary artery bypass grafting (CABG), etc. (10) Discharge information: Discharge diagnoses and discharge medications. (11) In-hospital adverse events.\\u003c/p\\u003e\\n\\u003ch3\\u003ePatient follow-up and endpoint events\\u003c/h3\\u003e\\n\\u003cp\\u003e All enrolled patients were followed for a minimum of one year. The primary endpoint of this study was all-cause mortality. All-cause mortality was defined and categorized according to the International Classification of Diseases, Tenth Revision (ICD-10) [\\u003cspan citationid=\\\"CR24\\\" class=\\\"CitationRef\\\"\\u003e24\\u003c/span\\u003e], including both cardiac and non-cardiac death. Cardiac death was defined as death due to fatal myocardial infarction, sudden cardiac death, or stroke. Non-cardiac death referred to death attributable to causes other than those listed above, including mortality from malignant tumors, infection, respiratory diseases, trauma/accident, or other non-vascular causes. If the cause of death could not be determined based on available information, it was classified as undetermined.\\u003c/p\\u003e \\u003cp\\u003eFollow-up was conducted by reviewing and recording information from patient re-admission records, performing telephone interviews with patients or their family members, and verifying death certificates by contacting the respective hospital where the death occurred. All death events were independently adjudicated by two members of the event adjudication committee.\\u003c/p\\u003e\\n\\u003ch3\\u003eStatistical methods\\u003c/h3\\u003e\\n\\u003cp\\u003eIn this study, descriptive statistics were first employed to summarize the baseline characteristics of the enrolled elderly patients with CAD and AF. For continuous variables, normality was assessed using appropriate statistical tests. Variables with a normal distribution are reported as mean\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;standard deviation, while non-normally distributed variables are presented as median with interquartile range. Categorical variables are described using frequencies and percentages. Comparisons for normally distributed continuous variables were made using the independent samples t-test, and the Mann-Whitney U test was used for non-normally distributed continuous variables. To identify predictors significantly associated with all-cause mortality from a comprehensive set of clinical variables, the Least Absolute Shrinkage and Selection Operator (LASSO) method was employed for variable selection.\\u003c/p\\u003e \\u003cp\\u003eBased on the variables previously selected by LASSO-Cox regression, two predictive models were constructed: (1) Model A: Variables comprising general clinical characteristics\\u0026thinsp;+\\u0026thinsp;laboratory test indicators. (2) Model B: Variables comprising general clinical characteristics\\u0026thinsp;+\\u0026thinsp;laboratory test indicators\\u0026thinsp;+\\u0026thinsp;comorbidity indicators.\\u003c/p\\u003e \\u003cp\\u003eNomograms were developed based on the LASSO-Cox regression models to visualize the prediction of 1-year and 5-year survival probabilities. The predictive performance of the LASSO-Cox models and the CHA₂DS₂-VASc score model for all-cause mortality was evaluated and compared using the Area Under the Receiver Operating Characteristic Curve (AUC), calibration curves, and Decision Curve Analysis (DCA). A two-sided \\u003cem\\u003ep\\u003c/em\\u003e value\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.05 was considered statistically significant.\\u003c/p\\u003e\"},{\"header\":\"Results\",\"content\":\"\\u003cdiv id=\\\"Sec9\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003eBaseline characteristics of the patients\\u003c/h2\\u003e \\u003cp\\u003eA total of 2,437 elderly patients (\\u0026ge;\\u0026thinsp;65 years) with CAD and AF, hospitalized in the Department of Cardiology at the Chinese PLA General Hospital between January 2010 and December 2017, were initially enrolled in this study. The mean follow-up duration was 5.4\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;2.5 years, with a follow-up rate of 93.27%. After excluding 4 cases with unspecified time of death, 1,678 patients were included for developing the clinical prediction model for all-cause mortality risk assessment. These included cases were randomly divided into a training set (n\\u0026thinsp;=\\u0026thinsp;1,174) and a validation set (n\\u0026thinsp;=\\u0026thinsp;504) in a 7:3 ratio. Except for unstable angina (37.65% vs. 45.63%, P\\u0026thinsp;=\\u0026thinsp;0.003), no statistically significant differences were observed in other general clinical data, laboratory findings, or comorbidities between the training and validation sets (\\u003cem\\u003ep\\u003c/em\\u003e\\u0026thinsp;\\u0026gt;\\u0026thinsp;0.05) (Table\\u0026nbsp;\\u003cspan refid=\\\"Tab1\\\" class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003e).\\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\\u003eComparison of General Clinical Characteristics in Elderly Patients with CAD and AF\\u003c/p\\u003e \\u003c/div\\u003e \\u003c/caption\\u003e \\u003ccolgroup cols=\\\"4\\\"\\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=\\\"char\\\" char=\\\".\\\" class=\\\"colspec\\\" colname=\\\"c3\\\" colnum=\\\"3\\\"\\u003e\\u003c/div\\u003e \\u003cdiv align=\\\"char\\\" char=\\\".\\\" class=\\\"colspec\\\" colname=\\\"c4\\\" colnum=\\\"4\\\"\\u003e\\u003c/div\\u003e \\u003cthead\\u003e \\u003ctr\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eClinical Characteristics\\u003c/p\\u003e \\u003c/th\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003eTraining Set\\u003c/p\\u003e \\u003cp\\u003e(n\\u0026thinsp;=\\u0026thinsp;1174)\\u003c/p\\u003e \\u003c/th\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003eValidation Set\\u003c/p\\u003e \\u003cp\\u003e(n\\u0026thinsp;=\\u0026thinsp;504)\\u003c/p\\u003e \\u003c/th\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e\\u003cem\\u003eP\\u003c/em\\u003e value\\u003c/p\\u003e \\u003c/th\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eDemographics\\u003c/p\\u003e \\u003c/th\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c2\\\"\\u003e\\u0026nbsp;\\u003c/th\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c3\\\"\\u003e\\u0026nbsp;\\u003c/th\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c4\\\"\\u003e\\u0026nbsp;\\u003c/th\\u003e \\u003c/tr\\u003e \\u003c/thead\\u003e \\u003ctbody\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eFemale sex, n (%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e499 (42.50)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e221 (43.85)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.648\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eAge (years), mean\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;SD\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e77.28\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;6.67\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e77.56\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;6.82\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.436\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eBMI (kg/m2), mean\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;SD\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e24.82\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;3.79\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e24.76\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;3.93\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.814\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003e\\u003cb\\u003eLaboratory Examinations\\u003c/b\\u003e\\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 \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eLeft ventricular ejection fraction (%),\\u003c/p\\u003e \\u003cp\\u003emedian (Q1, Q3)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e57 (52,62)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e58 (52.25,63)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.725\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eUric acid (\\u0026micro;mol/L), mean\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;SD\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e360.69\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;121.14\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e353.91\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;115.98\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.287\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eCreatinine (\\u0026micro;mol/L), mean\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;SD\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e97.39\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;58.84\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e92.83\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;41.93\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.115\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eBlood glucose (mmol/L), mean\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;SD\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e6.50\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;2.56\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e6.57\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;2.85\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.653\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eAST (U/L), mean\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;SD\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e29.82\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;60.27\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e28.00\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;44.23\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.540\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eALT (U/L), mean\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;SD\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e26.69\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;69.33\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e22.88\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;29.88\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.235\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003ePlasma fibrinogen (g/L), mean\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;SD\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e3.54\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;1.05\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e3.54\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.98\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.923\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003ePlasma D-dimer (\\u0026micro;g/mL), mean\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;SD\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e1.24\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;2.23\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e1.15\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;1.85\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.454\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eProthrombin time (s), mean\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;SD\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e15.28\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;6.97\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e15.09\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;3.37\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.562\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eNT-proBNP (pg/mL), M(Q1,Q3)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e1202\\u003c/p\\u003e \\u003cp\\u003e(47-81.40,2932.25)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e1265.50\\u003c/p\\u003e \\u003cp\\u003e(569.68,2790.75)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.621\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003ePlatelet count (109/L), mean\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;SD\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e185.48 (62.09)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e189.03 (65.20)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.291\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eHemoglobin (g/L), mean\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;SD\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e128.05\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;19.21\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e129.38\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;20.36\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.201\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eHematocrit (L/L), mean\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;SD\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e0.38\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.06\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e0.39\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.06\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.333\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eTriglycerides (mmol/L), mean\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;SD\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e1.25\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.83\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e1.19\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.61\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.143\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eTotal cholesterol (mmol/L), mean\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;SD\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e3.83\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.95\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e3.81\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.92\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.825\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eLDL (mmol/L), mean\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;SD\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e2.23\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.78\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e2.26\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.76\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.428\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eHDL (mmol/L), mean\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;SD\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e1.16\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.35\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e1.14\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.34\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.256\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003e\\u003cb\\u003eType of CAD, n (%)\\u003c/b\\u003e\\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 \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eUnstable angina\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e442 (37.65)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e230 (45.63)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e\\u003cb\\u003e0.003\\u003c/b\\u003e\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eStable angina\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e109 (9.28)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e35 (6.94)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.141\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eAcute myocardial infarction\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e86 (7.33)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e42 (8.33)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.540\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eIschemic cardiomyopathy\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e18 (1.53)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e7 (1.39)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.997\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003e\\u003cb\\u003eType of Atrial Fibrillation, n (%)\\u003c/b\\u003e\\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 \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003ePersistent atrial fibrillation\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e220 (18.74)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e93 (18.45)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.944\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eParoxysmal atrial fibrillation\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e554 (47.19)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e244 (48.41)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.684\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003e\\u003cb\\u003eComorbidities, n (%)\\u003c/b\\u003e\\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 \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eCoronary artery disease\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e214 (18.23)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e90 (17.86)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.911\\u003c/p\\u003e \\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 \\u003cp\\u003e886 (75.47)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e371 (73.62)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.457\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eDyslipidemia\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e267 (22.74)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e112 (22.22)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.865\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eHistory of myocardial infarction\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e154 (13.12)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e62 (12.30)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.705\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eRenal dysfunction\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e178 (15.16)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e67 (13.29)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.359\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eHeart failure\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e452 (38.50)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e195 (38.69)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.985\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eDiabetes mellitus\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e354 (30.15)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e161 (31.94)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.502\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eStroke\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e287 (24.45)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e118 (23.41)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.696\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eChronic obstructive pulmonary disease\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e48 (4.09)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e17 (3.37)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.577\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eMalignant tumor\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e146 (12.44)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e63 (12.50)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e1.000\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eRheumatic heart disease\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e44 (3.75)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e14 (2.78)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.395\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eValvular heart disease\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e41 (3.49)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e14 (2.78)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.546\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003c/tbody\\u003e \\u003c/colgroup\\u003e \\u003ctfoot\\u003e \\u003ctr\\u003e\\u003ctd colspan=\\\"4\\\"\\u003e\\u003cb\\u003eNote\\u003c/b\\u003e: Data are presented as median (interquartile range), mean\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;standard deviation, or number (percentage), as appropriate. \\u003cb\\u003eAbbreviations\\u003c/b\\u003e: AST, aspartate aminotransferase; ALT, alanine aminotransferase; NT-proBNP, N-terminal pro-B-type natriuretic peptide; LDL, low-density lipoprotein; HDL, high-density lipoprotein.\\u003c/td\\u003e\\u003c/tr\\u003e \\u003c/tfoot\\u003e \\u003c/table\\u003e\\u003c/div\\u003e \\u003c/p\\u003e \\u003c/div\\u003e\\n\\u003ch3\\u003eSelection of relevant factors and model development\\u003c/h3\\u003e\\n\\u003cp\\u003eRelevant clinical factors associated with all-cause mortality in elderly CAD-AF patients were screened using LASSO-Cox regression. Employing ten-fold cross-validation (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig1\\\" class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003e), the model ultimately retained 17 variables: age, left ventricular ejection fraction, blood glucose, plasma fibrinogen, plasma D-dimer, prothrombin time, NT-proBNP, hemoglobin, hematocrit, acute myocardial infarction, history of myocardial infarction, renal dysfunction, heart failure, diabetes, stroke, chronic obstructive pulmonary disease, and malignant tumors.\\u003c/p\\u003e \\u003cp\\u003e \\u003c/p\\u003e \\u003cp\\u003eUsing these variables, two models were constructed:\\u003c/p\\u003e \\u003cp\\u003e \\u003cstrong\\u003eLASSO-Cox Model A\\u003c/strong\\u003e \\u003cp\\u003eIncluded age, left ventricular ejection fraction, blood glucose, plasma fibrinogen, plasma D-dimer, prothrombin time, NT-proBNP, hemoglobin, and hematocrit.\\u003c/p\\u003e \\u003c/p\\u003e \\u003cp\\u003e \\u003cstrong\\u003eLASSO-Cox Model B\\u003c/strong\\u003e \\u003cp\\u003eIncluded all variables from Model A plus acute myocardial infarction, history of myocardial infarction, renal dysfunction, heart failure, diabetes, stroke, chronic obstructive pulmonary disease, and malignant tumors.\\u003c/p\\u003e \\u003c/p\\u003e \\u003cp\\u003eThe predictive performance of LASSO-Cox Models A and B was described and compared against the CHA₂DS₂-VASc score using the Area Under the Curve (AUC) (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig2\\\" class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003e). For predicting 1-year all-cause mortality risk, LASSO-Cox Model B demonstrated the highest predictive performance in the training set (AUC: 0.85, 95% CI: 0.81\\u0026ndash;0.89). Furthermore, the predictive ability of the CHA₂DS₂-VASc score model for 1-year all-cause mortality was significantly lower than that of both LASSO-Cox Model A and Model B, in both the training set (AUC: 0.66, 95% CI: 0.62\\u0026ndash;0.71) and the validation set (AUC: 0.57, 95% CI: 0.48\\u0026ndash;0.67). Similarly, for predicting 5-year all-cause mortality risk, the predictive ability of the CHA₂DS₂-VASc score model remained significantly inferior to both LASSO-Cox models in both the training set (AUC: 0.62, 95% CI: 0.58\\u0026ndash;0.66) and the validation set (AUC: 0.57, 95% CI: 0.50\\u0026ndash;0.63). Since LASSO-Cox Model B exhibited the best predictive performance across both the training and validation sets, it was selected as the final optimal model.\\u003c/p\\u003e \\u003cp\\u003e \\u003c/p\\u003e \\u003cdiv id=\\\"Sec11\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003eConstruction of a nomogram for predicting all-cause mortality risk in elderly CAD-AF patients\\u003c/h2\\u003e \\u003cp\\u003eA Cox proportional hazards regression model was developed based on the 17 independent predictors (LASSO-Cox Model B) in the training set to create a nomogram for predicting the 1-year and 5-year all-cause mortality probabilities in elderly patients with CAD and AF (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig3\\\" class=\\\"InternalRef\\\"\\u003e3\\u003c/span\\u003e).\\u003c/p\\u003e \\u003cp\\u003e \\u003c/p\\u003e \\u003cp\\u003eFigure \\u003cspan refid=\\\"Fig3\\\" class=\\\"InternalRef\\\"\\u003e3\\u003c/span\\u003e shows that age contributed the most points to the nomogram score, followed by prothrombin time, blood glucose, plasma D-dimer, left ventricular ejection fraction, and other variables. For an individual patient, the points corresponding to each predictor can be obtained from the nomogram. The sum of these points yields a total score, which can then be converted into the predicted probability of 1-year and 5-year survival for elderly patients with CAD and AF.\\u003c/p\\u003e \\u003c/div\\u003e \\u003cdiv id=\\\"Sec12\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003eEvaluation of model performance for predicting all-cause mortality in elderly CAD-AF patients\\u003c/h2\\u003e \\u003cp\\u003eThe calibration curves of the predictive models are shown in Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig4\\\" class=\\\"InternalRef\\\"\\u003e4\\u003c/span\\u003e. For predicting 1-year all-cause mortality in the training set, all three models demonstrated good agreement with observed outcomes when the predicted survival probability was above 0.9. In the validation set for 1-year all-cause mortality, good agreement was observed across all three models when the predicted survival probability exceeded 0.95. All models also showed favorable calibration for predicting 5-year all-cause mortality in both datasets.\\u003c/p\\u003e \\u003cp\\u003e \\u003c/p\\u003e \\u003cp\\u003eDCA revealed that compared to the CHA₂DS₂-VASc score model, the LASSO-Cox machine learning models provided a greater net benefit for predicting both 1-year and 5-year all-cause mortality risk (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig5\\\" class=\\\"InternalRef\\\"\\u003e5\\u003c/span\\u003e).\\u003c/p\\u003e \\u003cp\\u003e \\u003c/p\\u003e \\u003c/div\\u003e\"},{\"header\":\"Discussion\",\"content\":\"\\u003cp\\u003eThis study found that a Cox proportional hazards regression model, developed via the LASSO-Cox algorithm and incorporating 17 clinical risk factors, accurately predicted the risk of all-cause mortality in elderly patients with coexisting CAD and AF. Compared to the CHA₂DS₂-VASc score model, the prediction model established in this study via the LASSO-Cox machine learning algorithm demonstrated superior discriminatory ability and provided a greater net clinical benefit in both the training and validation sets.\\u003c/p\\u003e \\u003cp\\u003eCurrently, there is a lack of machine learning-based prediction models for all-cause mortality risk specifically targeted at elderly patients aged 65 and above with concurrent CAD and AF. This research serves as a valuable reference for machine learning-based risk assessment of all-cause mortality in this specific patient population.\\u003c/p\\u003e \\u003cp\\u003eThe prediction model developed in this study incorporates three categories of clinical indicators: general patient characteristics, laboratory test results, and patient comorbidities. Our results identified age, left ventricular ejection fraction, blood glucose, plasma fibrinogen, plasma D-dimer, hemoglobin, hematocrit, history of myocardial infarction, history of heart failure, and renal dysfunction as risk factors for all-cause mortality in patients with coexisting CAD and AF. These findings share common ground with the model proposed by Dong Min et al. However, our model additionally includes prothrombin time, NT-proBNP, blood glucose, diabetes, stroke, chronic obstructive pulmonary disease, and malignant tumors as risk factors influencing all-cause mortality in elderly patients with CAD and AF.\\u003c/p\\u003e \\u003cp\\u003eAmong these, NT-proBNP emerged as the most significant laboratory predictor for all-cause mortality and adverse cardiovascular events. NT-proBNP concentration is associated with left ventricular filling pressure and wall stress [\\u003cspan additionalcitationids=\\\"CR26\\\" citationid=\\\"CR25\\\" class=\\\"CitationRef\\\"\\u003e25\\u003c/span\\u003e\\u0026ndash;\\u003cspan citationid=\\\"CR27\\\" class=\\\"CitationRef\\\"\\u003e27\\u003c/span\\u003e]. Its levels increase during left ventricular dysfunction and acute myocardial infarction, and it counteracts heart failure through diuretic, natriuretic, and antihypertensive effects [\\u003cspan citationid=\\\"CR28\\\" class=\\\"CitationRef\\\"\\u003e28\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR29\\\" class=\\\"CitationRef\\\"\\u003e29\\u003c/span\\u003e]. However, the lack of a specific threshold hinders its application in clinical risk prediction [\\u003cspan citationid=\\\"CR30\\\" class=\\\"CitationRef\\\"\\u003e30\\u003c/span\\u003e]. Therefore, our prediction model provides a more effective tool for utilizing NT-proBNP in assessing all-cause mortality risk in elderly patients with CAD and AF.\\u003c/p\\u003e \\u003cp\\u003eA Swedish study indicated that malignant tumors and renal failure can increase the risk of all-cause mortality in AF patients [\\u003cspan citationid=\\\"CR31\\\" class=\\\"CitationRef\\\"\\u003e31\\u003c/span\\u003e]. Elevated blood glucose has also been reported to be associated with adverse cardiovascular events [\\u003cspan citationid=\\\"CR32\\\" class=\\\"CitationRef\\\"\\u003e32\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR33\\\" class=\\\"CitationRef\\\"\\u003e33\\u003c/span\\u003e]. The present study not only corroborates these established associations but further quantifies their predictive value within a dedicated model for elderly patients with CAD and AF. Specifically, variables representing these conditions: malignant tumor, renal dysfunction, and blood glucose were consistently selected into our LASSO-Cox model (Model B) as significant predictors. It underscores that our model successfully captures and integrates these critical laboratory and comorbidity-based risk dimensions. Consequently, greater attention should be paid to laboratory findings and comorbidities in elderly patients with CAD and AF. Early diagnosis and timely intervention targeting these factors are crucial for improving mortality outcomes and prolonging survival.\\u003c/p\\u003e \\u003cp\\u003eIn a study of 2,174 ACS patients aged 50 and above, Li et al. utilized LASSO regression to select six variables and subsequently developed a Cox regression model to predict 3-year all-cause mortality. Their results showed that this model achieved a predictive performance (AUC) of 0.768 and a calibration of 0.711, both of which were superior to those of the traditional GRACE model (AUC: 0.701; calibration: 0.203) [\\u003cspan citationid=\\\"CR34\\\" class=\\\"CitationRef\\\"\\u003e34\\u003c/span\\u003e]. Similarly, Sun et al., working with 468 heart failure patients over 50, applied LASSO regression to select 26 variables and then employed four machine learning algorithms\\u0026mdash;Extreme Gradient Boosting (XGBoost), Random Survival Forest, Multi-layer Perceptron, and Support Vector Machine\\u0026mdash;to build models predicting the risk of Major Adverse Cardiovascular Events (MACEs) within six months post-discharge. They found that the model built using the XGBoost algorithm performed best, with an AUC of 0.84 and a calibration of 0.77 [\\u003cspan citationid=\\\"CR35\\\" class=\\\"CitationRef\\\"\\u003e35\\u003c/span\\u003e]. These studies collectively affirm the utility of LASSO-based variable selection combined with regression or machine learning modeling in cardiovascular prognosis. Our study extends this methodological approach to a distinct and clinically complex cohort\\u0026mdash;elderly patients with CAD and AF\\u0026mdash;a population at notably high risk yet lacking a dedicated prognostic tool. By applying LASSO-Cox regression, we not only confirmed the feasibility of this approach in a multimorbid setting but also developed a model (Model B) that demonstrated robust predictive performance (AUC up to 0.85 for 1-year mortality), thereby providing a tailored risk assessment instrument for this patient group.\\u003c/p\\u003e \\u003cp\\u003eThe Cox proportional hazards model developed in this study demonstrated significantly superior discriminative ability for all-cause mortality compared to the conventional CHA₂DS₂-VASc score in elderly patients with coexisting CAD and AF. Specifically, our model exhibited robust predictive performance, with an AUC of 0.85 for 1-year mortality in the training set, and maintained stable, clinically relevant accuracy in the independent validation set (1-year AUC: 0.80; sustained AUC\\u0026thinsp;\\u0026gt;\\u0026thinsp;0.75 during long-term follow-up). This performance markedly outperformed the CHA₂DS₂-VASc score, which showed limited discriminative power in the same cohort (AUCs between 0.57 and 0.66). More importantly, decision curve analysis confirmed that our model, which integrates over ten clinical risk factors selected via the LASSO-Cox algorithm, provides a greater net clinical benefit across a wide range of risk thresholds. These findings collectively indicate that a multivariate model incorporating diverse clinical, laboratory, and comorbidity data offers a more accurate and clinically useful risk stratification tool for this high-risk, complex population than the traditional score-based approach.\\u003c/p\\u003e \\u003cdiv id=\\\"Sec14\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003eLimitations\\u003c/h2\\u003e \\u003cp\\u003eThis study has several limitations. First, the types of variables available in the clinical follow-up database for model selection were limited. A lack of structured information, such as details regarding surgical interventions and imaging data, restricted the inclusion of other predictors that might significantly influence clinical outcomes.\\u003c/p\\u003e \\u003cp\\u003eSecondly, the study exclusively employed the LASSO-Cox machine learning algorithm for model construction. The randomness in variable selection can differ across various machine learning algorithms. Therefore, future work should involve comparing models built using other algorithms, including Random Forest, to further validate and refine our approach.\\u003c/p\\u003e \\u003cp\\u003eFurthermore, as this was a single-center, retrospective study, subsequent external validation using cases from other institutions is necessary to verify the model's predictive performance and generalizability.\\u003c/p\\u003e \\u003c/div\\u003e\"},{\"header\":\"Conclusion\",\"content\":\"\\u003cp\\u003eThe Cox proportional hazards regression model established based on the machine learning algorithm demonstrated superior predictive ability for all-cause mortality risk in elderly patients with CAD and AF, showing both higher predictive performance and greater net benefit compared to the CHA₂DS₂-VASc score model.\\u003c/p\\u003e\"},{\"header\":\"Declarations\",\"content\":\"\\u003ch2\\u003eEthics approval and consent to participate\\u003c/h2\\u003e\\n\\u003cp\\u003eThe study was approved by the Ethics Committee of Chinese PLA General Hospital (S2024-442-02). Informed consent was obtained from all subjects.\\u003c/p\\u003e\\n\\u003ch2\\u003eData Avaliability\\u003c/h2\\u003e\\n\\u003cp\\u003eDate is available upon author\\u0026lsquo;s request.\\u003c/p\\u003e\\n\\u003ch2\\u003eConflicts of interest\\u003c/h2\\u003e\\n\\u003cp\\u003eNo potential conflict of interest was reported by the author(s).\\u003c/p\\u003e\\n\\u003ch2\\u003eFunding\\u003c/h2\\u003e\\n\\u003cp\\u003eand acknowledgements\\u003c/p\\u003e\\n\\u003cp\\u003eThis research was financially supported by grants from the National Natural Science Foundation of China (No. 81870262, 82170352).\\u003c/p\\u003e\\n\\u003ch2\\u003eAuthor Contribution\\u003c/h2\\u003e\\n\\u003cp\\u003eYuyan Wang: Conceptualization, Methodology, Formal analysis, Writing \\u0026ndash; Original Draft.Yangxun Wu: Data Curation, Software, Validation.Yuting Zou: Investigation, Resources.Shuai Mao: Visualization, Writing \\u0026ndash; Review \\u0026amp; Editing.Yanzhu Yao: Investigation, Data Curation.Yifan Wang: Methodology, Validation.Yunzhang Zhao: Formal analysis, Software.Chao Lv: Resources, Project Administration.Li Fan: Supervision, Writing \\u0026ndash; Review \\u0026amp; Editing, Funding Acquisition.Tong Yin: Supervision, Writing \\u0026ndash; Review \\u0026amp; Editing, Funding Acquisition, Project Administration.\\u003c/p\\u003e\"},{\"header\":\"References\",\"content\":\"\\u003col\\u003e\\u003cli\\u003e\\u003cspan\\u003eFadah K, Hechanova A, Mukherjee D. Epidemiology, Pathophysiology, and Management of Coronary Artery Disease in the Elderly. Int J Angiol. 2022;31(4):244\\u0026ndash;50.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eZhang J, et al. Epidemiology of Atrial Fibrillation: Geographic/Ecological Risk Factors, Age, Sex, Genetics. Card Electrophysiol Clin. 2021;13(1):1\\u0026ndash;23.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eSteinhubl SR, et al. Early and sustained dual oral antiplatelet therapy following percutaneous coronary intervention: a randomized controlled trial. JAMA. 2002;288(19):2411\\u0026ndash;20.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eSteensig K, et al. CAD Is an Independent Risk Factor for Stroke Among Patients With Atrial Fibrillation. J Am Coll Cardiol. 2018;72(20):2540\\u0026ndash;2.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eSaglietto A, et al. Prognostic implications of atrial fibrillation in patients with stable coronary artery disease: a systematic review and meta-analysis of adjusted observational studies. Rev Cardiovasc Med. 2021;22(2):439\\u0026ndash;44.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003ePetersen JK, et al. Incidence of ischaemic stroke and mortality in patients with acute coronary syndrome and first-time detected atrial fibrillation: a nationwide study. Eur Heart J. 2021;42(44):4553\\u0026ndash;61.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eWu Y, et al. Assessment of the CHA(2)DS(2)-VASc Score for the Prediction of Death in Elderly Patients With Coronary Artery Disease and Atrial Fibrillation. Front Cardiovasc Med. 2021;8:805234.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eHindricks G et al. 2020 \\u003cem\\u003eESC Guidelines for the diagnosis and management of atrial fibrillation developed in collaboration with the European Association for Cardio-Thoracic Surgery (EACTS): The Task Force for the diagnosis and management of atrial fibrillation of the European Society of Cardiology (ESC) Developed with the special contribution of the European Heart Rhythm Association (EHRA) of the ESC.\\u003c/em\\u003e Eur Heart J, 2021. 42(5): pp. 373\\u0026ndash;498.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eKim D, et al. Long-Term Anticoagulation Discontinuation After Catheter Ablation for Atrial Fibrillation: The ALONE-AF Randomized Clinical Trial. JAMA. 2025;334(14):1246\\u0026ndash;54.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eShuvy M, et al. Value of the CHA(2) DS(2) -VASc score for predicting outcome in patients with heart failure. ESC Heart Fail. 2020;7(5):2553\\u0026ndash;60.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eG\\u0026ouml;k M, et al. Relationship Between CHA2DS2-VASc Score and Right Ventricular Dysfunction in Patients With Acute Pulmonary Thromboembolism. Clin Appl Thromb Hemost. 2018;24(9suppl):s56\\u0026ndash;62.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eLi CY, et al. Assessment of CHA2DS2-VASc score for predicting cardiovascular and cerebrovascular outcomes in acute myocardial infarction patients. Med (Baltim). 2018;97(28):e11230.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eTopaz G, et al. CHA(2) DS(2) -VASc score and clinical outcomes of patients with chest pain discharged from internal medicine wards following acute coronary syndrome rule-out. Clin Cardiol. 2018;41(4):539\\u0026ndash;43.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eMa X, et al. Prognostic value of CHADS2 and CHA2DS2-VASc scores for post-discharge outcomes in patients with acute coronary syndrome undergoing percutaneous coronary intervention. Med (Baltim). 2020;99(30):e21321.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eAndreotti F, et al. Acute, periprocedural and longterm antithrombotic therapy in older adults: 2022 Update by the ESC Working Group on Thrombosis. Eur Heart J. 2023;44(4):262\\u0026ndash;79.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eChen Y, et al. Predicting All-Cause Mortality Risk in Atrial Fibrillation Patients: A Novel LASSO-Cox Model Generated From a Prospective Dataset. Front Cardiovasc Med. 2021;8:730453.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eHill NR, et al. Predicting atrial fibrillation in primary care using machine learning. PLoS ONE. 2019;14(11):e0224582.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eLanfear DE, et al. Targeted Metabolomic Profiling of Plasma and Survival in Heart Failure Patients. JACC Heart Fail. 2017;5(11):823\\u0026ndash;32.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eMavaddat N, et al. Polygenic Risk Scores for Prediction of Breast Cancer and Breast Cancer Subtypes. Am J Hum Genet. 2019;104(1):21\\u0026ndash;34.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eVrints C, et al. 2024 ESC Guidelines for the management of chronic coronary syndromes. Eur Heart J. 2024;45(36):3415\\u0026ndash;537.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eLim M, Hastie T. Learning interactions via hierarchical group-lasso regularization. J Comput Graph Stat. 2015;24(3):627\\u0026ndash;54.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eKnuuti J, et al. 2019 ESC Guidelines for the diagnosis and management of chronic coronary syndromes. Eur Heart J. 2020;41(3):407\\u0026ndash;77.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eLip GYH et al. 2018 \\u003cem\\u003eJoint European consensus document on the management of antithrombotic therapy in atrial fibrillation patients presenting with acute coronary syndrome and/or undergoing percutaneous cardiovascular interventions: a joint consensus document of the European Heart Rhythm Association (EHRA), European Society of Cardiology Working Group on Thrombosis, European Association of Percutaneous Cardiovascular Interventions (EAPCI), and European Association of Acute Cardiac Care (ACCA) endorsed by the Heart Rhythm Society (HRS), Asia-Pacific Heart Rhythm Society (APHRS), Latin America Heart Rhythm Society (LAHRS), and Cardiac Arrhythmia Society of Southern Africa (CASSA).\\u003c/em\\u003e Europace, 2019. 21(2): pp. 192\\u0026ndash;193.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eICD-10: there's a code for that. Lancet, 2015. 386(10002): p. 1420.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eNishi H, et al. N-terminal pro-brain natriuretic peptide and cardiorenal outcome in patients with anaemia in chronic kidney disease. ESC Heart Fail. 2025;12(2):848\\u0026ndash;58.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eJanuzzi JL, Mebazaa A, Di Somma S. ST2 and prognosis in acutely decompensated heart failure: the International ST2 Consensus Panel. Am J Cardiol. 2015;115(7 Suppl):b26\\u0026ndash;31.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eJanuzzi JL Jr., et al. Measurement of the interleukin family member ST2 in patients with acute dyspnea: results from the PRIDE (Pro-Brain Natriuretic Peptide Investigation of Dyspnea in the Emergency Department) study. J Am Coll Cardiol. 2007;50(7):607\\u0026ndash;13.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eBooth RA, et al. Performance of BNP and NT-proBNP for diagnosis of heart failure in primary care patients: a systematic review. Heart Fail Rev. 2014;19(4):439\\u0026ndash;51.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eJanuzzi JL, et al. NT-proBNP testing for diagnosis and short-term prognosis in acute destabilized heart failure: an international pooled analysis of 1256 patients: the International Collaborative of NT-proBNP Study. Eur Heart J. 2006;27(3):330\\u0026ndash;7.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eLi X, et al. Plasma NT pro-BNP, hs-CRP and big-ET levels at admission as prognostic markers of survival in hospitalized patients with dilated cardiomyopathy: a single-center cohort study. BMC Cardiovasc Disord. 2014;14:67.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eAndersson T, et al. All-cause mortality in 272,186 patients hospitalized with incident atrial fibrillation 1995\\u0026ndash;2008: a Swedish nationwide long-term case-control study. Eur Heart J. 2013;34(14):1061\\u0026ndash;7.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eYatsuya H, et al. Risk Factor of Cardiovascular Disease Among Older Individuals. J Atheroscler Thromb. 2017;24(3):258\\u0026ndash;61.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eLiu X, et al. Relationship between the triglyceride-glucose index and risk of cardiovascular diseases and mortality in the general population: a systematic review and meta-analysis. Cardiovasc Diabetol. 2022;21(1):124.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eLi YM, et al. A LASSO-derived risk model for long-term mortality in Chinese patients with acute coronary syndrome. J Transl Med. 2020;18(1):157.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eSun Q, et al. A prediction model for major adverse cardiovascular events in patients with heart failure based on high-throughput echocardiographic data. Front Cardiovasc Med. 2022;9:1022658.\\u003c/span\\u003e\\u003c/li\\u003e\\u003c/ol\\u003e\"}],\"fulltextSource\":\"\",\"fullText\":\"\",\"funders\":[],\"hasAdminPriorityOnWorkflow\":false,\"hasManuscriptDocX\":true,\"hasOptedInToPreprint\":true,\"hasPassedJournalQc\":\"\",\"hasAnyPriority\":false,\"hideJournal\":true,\"highlight\":\"\",\"institution\":\"\",\"isAcceptedByJournal\":false,\"isAuthorSuppliedPdf\":false,\"isDeskRejected\":\"\",\"isHiddenFromSearch\":false,\"isInQc\":false,\"isInWorkflow\":false,\"isPdf\":false,\"isPdfUpToDate\":true,\"isWithdrawnOrRetracted\":false,\"journal\":{\"display\":true,\"email\":\"info@researchsquare.com\",\"identity\":\"researchsquare\",\"isNatureJournal\":false,\"hasQc\":true,\"allowDirectSubmit\":true,\"externalIdentity\":\"\",\"sideBox\":\"\",\"snPcode\":\"\",\"submissionUrl\":\"/submission\",\"title\":\"Research Square\",\"twitterHandle\":\"researchsquare\",\"acdcEnabled\":true,\"dfaEnabled\":false,\"editorialSystem\":\"\",\"reportingPortfolio\":\"\",\"inReviewEnabled\":false,\"inReviewRevisionsEnabled\":true},\"keywords\":\"Elderly, Coronary artery disease, Atrial fibrillation, Mortality, Machine learning\",\"lastPublishedDoi\":\"10.21203/rs.3.rs-8337799/v1\",\"lastPublishedDoiUrl\":\"https://doi.org/10.21203/rs.3.rs-8337799/v1\",\"license\":{\"name\":\"CC BY 4.0\",\"url\":\"https://creativecommons.org/licenses/by/4.0/\"},\"manuscriptAbstract\":\"\\u003cp\\u003eBackground: Elderly patients with coexisting coronary artery disease (CAD) and atrial fibrillation (AF) are at significantly increased risk of mortality. Accurate risk stratification is crucial for improving clinical management, yet a dedicated predictive tool for this specific population is lacking. The widely used CHA₂DS₂-VASc score demonstrates limited performance in predicting all-cause mortality in this complex comorbid group.\\u003c/p\\u003e\\n\\u003cp\\u003eMethods: A cohort of elderly inpatients (≥65 years) diagnosed with CAD and AF were retrospectively enrolled from the Department of Cardiology at the Chinese PLA General Hospital between January 2010 and December 2017. Baseline clinical data during hospitalization were collected, and all patients were followed up for all-cause mortality. Patients were randomly divided into a training set (70%) and a validation set (30%). Variable selection was performed using LASSO-Cox regression. Predictive models were established through a Cox proportional hazards model (via LASSO-Cox) and a random survival forest model to predict all-cause mortality risk. Model performance was evaluated using the area under the receiver operating characteristic curve (AUC), calibration curves, and decision curve analysis (DCA), and compared against the CHA₂DS₂-VASc score.\\u003c/p\\u003e\\n\\u003cp\\u003eResults: A total of 1678 elderly patients with CAD and AF were randomly divided into a training set (n=1174) and a validation set (n=504). Through LASSO‑Cox regression, 17 variables were identified as predictors associated with all‑cause mortality. Two distinct models were developed: LASSO‑Cox Model A included factors such as age, left ventricular ejection fraction, blood glucose, plasma fibrinogen, D‑dimer, prothrombin time, NT‑proBNP, hemoglobin, and hematocrit; LASSO‑Cox Model B incorporated all variables from Model A plus acute myocardial infarction, history of myocardial infarction, renal insufficiency, heart failure, diabetes, stroke, chronic obstructive pulmonary disease, and malignant tumor.\\u003c/p\\u003e\\n\\u003cp\\u003eThe analysis of Cox proportional hazards regression models showed that in the training sets, the Area Under Curves (AUC) of model A, model B and CHA₂DS₂-VASc scores for predicting 1-year all-cause death were 0.83, 0.85 and 0.66, respectively; the AUC for 5-year all-cause death were 0.74, 0.76 and 0.62, respectively. In the validation sets, the AUC of model A, model B and the CHA₂DS₂-VASc score for predicting 1-year all-cause mortality were 0.79, 0.78 and 0.57, respectively; the AUC for 5-year all-cause death were 0.74, 0.75 and 0.57, respectively. Model B demonstrated better calibration and provided greater net clinical benefit in DCA.\\u003c/p\\u003e\\n\\u003cp\\u003eConclusion: The LASSO‑Cox machine learning model demonstrates superior predictive performance for all cause mortality relative to the traditional CHA₂DS₂-VASc score in elderly patients with CAD and AF.\\u003c/p\\u003e\",\"manuscriptTitle\":\"Mortality Prediction in the elderly patients with Coronary Artery Disease and Atrial Fibrillation: A Retrospective Machine Learning Approach\",\"msid\":\"\",\"msnumber\":\"\",\"nonDraftVersions\":[{\"code\":1,\"date\":\"2026-01-13 09:16:27\",\"doi\":\"10.21203/rs.3.rs-8337799/v1\",\"editorialEvents\":[{\"type\":\"communityComments\",\"content\":0}],\"status\":\"published\",\"journal\":{\"display\":true,\"email\":\"info@researchsquare.com\",\"identity\":\"researchsquare\",\"isNatureJournal\":false,\"hasQc\":true,\"allowDirectSubmit\":true,\"externalIdentity\":\"\",\"sideBox\":\"\",\"snPcode\":\"\",\"submissionUrl\":\"/submission\",\"title\":\"Research Square\",\"twitterHandle\":\"researchsquare\",\"acdcEnabled\":true,\"dfaEnabled\":false,\"editorialSystem\":\"\",\"reportingPortfolio\":\"\",\"inReviewEnabled\":false,\"inReviewRevisionsEnabled\":true}}],\"origin\":\"\",\"ownerIdentity\":\"8deec779-afcf-4226-8659-da5abefdefd7\",\"owner\":[],\"postedDate\":\"January 13th, 2026\",\"published\":true,\"recentEditorialEvents\":[],\"rejectedJournal\":[],\"revision\":\"\",\"amendment\":\"\",\"status\":\"posted\",\"subjectAreas\":[],\"tags\":[],\"updatedAt\":\"2026-04-22T08:59:23+00:00\",\"versionOfRecord\":[],\"versionCreatedAt\":\"2026-01-13 09:16:27\",\"video\":\"\",\"vorDoi\":\"\",\"vorDoiUrl\":\"\",\"workflowStages\":[]},\"version\":\"v1\",\"identity\":\"rs-8337799\",\"journalConfig\":\"researchsquare\"},\"__N_SSP\":true},\"page\":\"/article/[identity]/[[...version]]\",\"query\":{\"redirect\":\"/article/rs-8337799\",\"identity\":\"rs-8337799\",\"version\":[\"v1\"]},\"buildId\":\"XKTyCvWXoU3ODBz1xrDgd\",\"isFallback\":false,\"isExperimentalCompile\":false,\"dynamicIds\":[84888],\"gssp\":true,\"scriptLoader\":[]}","source_license":"CC-BY-4.0","license_restricted":false}