Improvement in risk prediction for patients with atrial fibrillation and intermediate-risk CHA2DS2-VASc score utilizing highly sensitive cardiac troponin T

Christian Salbach; Mustafa Yildirim; Barbara Ruth Milles; Hauke Hund; Matthias Mueller-Hennessen; Norbert Frey; Hugo Katus; Evangelos Giannitsis

doi:10.1371/journal.pone.0330164

Abstract

Background

Guidelines of the European Society of Cardiology recommend a clinical risk assessment for patients with atrial fibrillation (AF). However, scores such as the CHA₂DS₂-VASc score show only a modest performance for prediction of adverse endpoints.

Methods

This retrospective single-center all-comer study uses data from the Heidelberg Registry of Atrial Fibrillation of 9,995 patients with non-valvular AF presenting to the emergency department (ED) of the University Hospital of Heidelberg from June 2009 until March 2020. Per CHA₂DS₂-VASc, risk was classified as low (0 point in men, ≤ 1 point in females), intermediate, or high (≥2 points in men and ≥3 points in females). The predictive performance of the CHA₂DS₂-VASc score, with and without highly sensitive cardiac troponin T (hs-cTnT), was evaluated for a composite endpoint comprising stroke, myocardial infarction (MI) or all-cause mortality.

Results

Performance of the CHA₂DS₂-VASc score for the prediction of the composite endpoint was poor Area under the curve (AUC): 0.648 (95%CI: 0.638–0.657) particularly in patients at intermediate-risk AUC: 0.542 (95%CI: 0.508–0.575). Adding hs-cTnT improved discrimination substantially in intermediate-risk patients (AUC: 0.778, 95% CI: 0.748–0.805). Notably, no events occurred in intermediate-risk patients with undetectable hs-cTnT (<5 ng/L).

Conclusion

In patients with AF at intermediate thromboembolic risk, the addition of hs-cTnT to the CHA₂DS₂-VASc score enhances prediction of adverse cardiovascular outcomes. Hs-cTnT may help identify patients who could benefit from anticoagulation, while also identifying a low-risk subgroup unlikely to experience events.

Citation: Salbach C, Yildirim M, Milles BR, Hund H, Mueller-Hennessen M, Frey N, et al. (2025) Improvement in risk prediction for patients with atrial fibrillation and intermediate-risk CHA₂DS₂-VASc score utilizing highly sensitive cardiac troponin T. PLoS One 20(8): e0330164. https://doi.org/10.1371/journal.pone.0330164

Editor: Seunghwa Lee, Wiltse Memorial Hospital, KOREA, REPUBLIC OF

Received: November 19, 2024; Accepted: July 28, 2025; Published: August 21, 2025

Copyright: © 2025 Salbach et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: The data for this study was retrospectively collected from routine care of consecutive patients admitted to the emergency department. The study was approved by the Ethics Committee of the Medical Faculty of Heidelberg without requiring informed consent. However, patient-level data cannot be shared publicly as informed consent is required for such sharing, which is not feasible in this context. For inquiries regarding the data, the study group of cardiac biomarkers at the Department of Cardiology, Angiology, and Pulmonology, Heidelberg University Hospital, can provide assistance. Requests will be handled in accordance with the study protocol and relevant regulations (e.g. the Declaration of Helsinki (2008) and General Data Protection Regulation (EU) 2016/679). Address for correspondence: Study Group Cardiac Biomarkers, Department of Cardiology, Angiology, and Pulmonology, Heidelberg University Hospital, Im Neuenheimer Feld 410, 69120 Heidelberg, Germany; E-mail: Giannitsis.Studienbuero@med.uni-heidelberg.de.

Funding: The study was supported by a research grant from Daiichi Sankyo and Bayer diagnostics. The sponsor had no influence on the study concept, data collection, or interpretation. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing interests: MMH received research funding from BRAHMS, Thermo Fisher Scientific and Roche Diagnostics. EG received honoraria for lectures from Roche Diagnostics, AstraZeneca, Bayer, Daiichi-Sankyo, Lilly Eli Deutschland. He serves as a consultant for Roche Diagnostics, BRAHMS, Thermo Fisher Scientific, Boehringer Ingelheim and has received research funding from BRAHMS, Thermo Fisher Scientific, Roche Diagnostics, Bayer Vital and Daiichi Sankyo. NF has received speaker honoraria from Daiichi Sankyo, Astra Zeneca, Boehringer Ingelheim and Bayer Vital. He serves as a consultant for CMS. BRM, received research funding from Daiichi Sankyo. There are no disclosures for MY, HH, HK and CS. This does not alter our adherence to PLOS ONE policies on sharing data and materials.

Introduction

Patients with atrial fibrillation (AF) have a manifold higher risk of stroke and suffer from a substantial burden of the disease [1–3]. Recent improvements for stroke prevention in AF patients include the replacement of vitamin-K-antagonist (VKA) by direct oral anticoagulant (DOAC) agents such as thrombin or factor Xa-inhibitors [4]. In clinical reality, AF is the most frequent arrhythmia in the setting of an emergency department (ED), causing rising costs and health care burden [5,6]. For the reduction of stroke risk in AF patients, initiation of adequate anticoagulation has been shown to be valuable but must be carefully balanced against the risk of higher bleeding rates [7,8]. Thus, guidelines of the European Society of Cardiology (ESC) recommend a clinical risk-factor based scoring system for the prediction of a patient individual stroke and bleeding risk [7]. Herein, several scores exists including the CHA₂DS₂-, CHA₂DS₂-VA- and the CHA₂DS₂-VASc score for evaluating stroke risk and guide decision of oral anticoagulation. [7] Unfortunately the ability of the CHA₂DS₂-VASc score to predict stroke is modest which has prompted the introduction of other validated scores such as the ABC (age, biomarkers, clinical history) stroke risk score, ATRIA (Anticoagulation and Risk Factors in Atrial Fibrillation) score, ORBIT (Outcomes Registry for Better Informed Treatment of Atrial Fibrillation) score, and the GARFIELD-AF (Global Anticoagulant Registry in the FIELD – Atrial Fibrillation) tool [9–12]. Current guideline recommendations regarding initiation of oral anticoagulation are clear for patients at high and low risk, whereas decision-making and risk prediction remains uncertain for those at intermediate risk according to CHA₂DS₂-VASc or CHA₂DS₂-VA score. [7,13] No specific guideline recommendations definitively guides management in this subgroup, making individualized assessment particularly important. Given this clinical uncertainty, we focused on patients at intermediate risk according to CHA₂DS₂-VASc score to evaluate whether biomarker-based risk stratification could provide more precise risk stratification for adverse outcomes. Several studies have been prompted to improve risk prediction of risk-factor based scores by using biomarkers [14–18]. For Asian patients with clinically low-risk CHA₂DS₂-VASc score and non-valvular AF, current evidence suggests that biomarkers such as NT-proBNP and creatinine could refine stroke risk stratification and prediction of adverse outcomes [19]. Thus, the current study aimed to evaluate an improvement in risk prediction of the CHA₂DS₂-VASc score by using highly sensitive cardiac troponin T (hs-cTnT) and retrospective data from the Heidelberg Registry of Atrial Fibrillation (HERA-FIB).

Materials and methods

Study population, design and follow-up

HERA-FIB consecutively included AF patients presenting between June 2009 and March 2020 to the ED of University Hospital of Heidelberg. Design and rationale of HERA-FIB, such as inclusion and exclusion criteria were published earlier [20]. Inclusion criteria were comprehensive including age ≥ 18 years and AF either as primary reason for admission to the ED or as a comorbidity. Patients were excluded from the study for nonavailability of at least one hs-cTnT value or when lost to follow-up for all-cause mortality. For the current study, patients with valvular AF, defined as patients with mechanical prosthetic heart valve(s) or moderate/severe mitral stenosis were excluded [21]. A flow diagram for in- and excluded patients within this study is shown in Fig 1. For outcome parameters, a sequential follow-up method was performed as previously described including screening of electronical patient data, phone calls and postal questionnaires and at least a contact to registrational offices for vital status [20]. This study was conducted according to ethical principles stated in the Declaration of Helsinki (2008) and was reviewed and approved by the local ethics committee of the Medical Faculty of Heidelberg. Data was accessed for research purposes 1^st April 2024. No written informed consent was required. Patient identifiable data was pseudonymized to ensure data confidentiality and was not passed on to third parties. HERA-FIB is registered at ClinicalTrials.gov. Clinical Trials.gov identifier: NCT05995561.

Download:

Fig 1. Flow-diagram for included AF patients admitted to the ED. The diagram shows patient recruitment after in- and exclusion of AF patients.

Abbreviations: AF, atrial fibrillation; ED, emergency departement; hs-cTnT, highly sensitive cardiac troponin T.

https://doi.org/10.1371/journal.pone.0330164.g001

Definitions

For this study, a composite endpoint including stroke, myocardial infarction (MI) or all-cause mortality was chosen. Stroke events included ischemic or unknown cause of stroke but excluded hemorrhagic strokes, which were allocated as a major bleeding event according to International Society on Thrombosis and Hemostasis (ISTH) major bleeding criteria. The outcome of major bleeding events was defined according to ISTH major bleeding criteria. The CHA₂DS₂-VASc score was calculated retrospectively by adding two points for age ≥ 75 years and prior stroke, transient ischemic attack (TIA) or thromboembolism, as well as one point for congestive heart failure (defined by left ventricular ejection fraction <50%), arterial hypertension, age between 65–74 years, known diabetes mellitus, prior MI or peripheral artery disease, and female sex, respectively [7,22]. The CHA₂DS₂-VA score was calculated as described [13]. Risk categories comprised of a high risk category according to CHA₂DS₂-VASc (≥2 points in men, ≥ 3 points in women or CHA₂DS₂-VASc ≥ 2 points regardless of sex) an intermediate risk category (CHA₂DS₂-VASc = 1 point in men, 2 points in women or CHA₂DS₂-VASc = 1 point regardless of sex) and a low risk category (CHA₂DS₂-VASc = 0 points in men, ≤ 1 point in women, or CHA₂DS₂-VASc = 0 points regardless of sex). For calculation of the hs-cTnCHA₂DS₂VASc-score, the original CHA₂DS₂-VASc score was extended by adding 0 points for hs-cTnT < 5 ng/L, 1 point for hs-cTnT 5–14 ng/L and 2 points for hs-cTnT > 14 ng/L. Throughout the study period, hs-cTnT was measured using the Roche Cobas E411 platform (Roche Diagnostics Ltd., Rotkreuz, Switzerland), with no changes in vendor. Test kits were updated as per manufacturer recommendations over time, but the fundamental analytical platform remained unchanged.

Statistical analysis

Data is presented as means (standard deviations, SD), medians (25^th, 75^th percentiles, IQR), Kaplan-Meier estimates, as well as counts or percentages. For continuous variables the Kolmogorov-Smirnov test was used to test for normal distribution. Subgroups of categorical variables were tested against each other using chi-squared test or Fisher’s exact test. For continuous variables an unpaired Student’s t-test or Wilcoxon rank-sum test was performed. For Kaplan-Meier analysis, subgroups were compared by Log-rank test. The proportional hazards assumption was analyzed by the Grambsch and Therneau method [23]. We used time dependent receiver-operating-characteristic (ROC) curves and calculated the area under the ROC curves (AUC) on censored survival data. Here, the 95% confidence interval (CI) was calculated according to Hanley and McNeil [24]. For comparison of ROC curves, method by Hanley and McNeil was used as described [25]. The incremental prognostic value of adding hs-cTnT to the CHA₂DS₂-VASc score was assessed using the Net Reclassification Index (NRI), following established methodology for survival data [26]. A two tailed P-value of <0.05 was considered to indicate statistical significance. Statistical analyses were performed using R software (version 4.3.0, R Foundation for Statistical Computing, Vienna, Austria) as well as MedCalc (Version 20.105. MedCalc Software Ltd, Ostend Belgium).

Results

In this study, 227 (2.2%) of initially 10,222 included AF patients within HERA-FIB were excluded for the presence of mechanical prosthetic heart valve(s) or moderate/severe mitral stenosis, resulting in a total of 9,995 included patients with non-valvular AF (Fig 1). Baseline characteristics are reported, stratified by clinical risk category of CHA₂DS₂-VASc score. A total of 455 patients (4.6%) were classified as low-risk, 871 patients (8.7%) as intermediate-risk and 8,669 patients (86.7%) as high-risk (Table 1). A figure for the distribution of the CHA₂DS₂-VASc score among the whole cohort is shown in S1 Fig. Patients subcategorized using CHA₂DS₂-VASc score differed significantly with respect to the baseline variables. The majority of the patients in the intermediate-risk group were male (65.7%), median age was 59 years (IQR 52–64). Within intermediate-risk group, the most frequent comorbidity was arterial hypertension (54.9%), followed by coronary artery disease (6.3%) and former MI (2.3%). A total of 61.1% of patients categorized intermediate-risk by CHA₂DS₂-VASc score received an oral anticoagulation (OAC) at discharge. Heart rate at admission was the highest in intermediate-risk category. Cardiac biomarkers such as hs-cTnT showed a steady increase among risk all risk categories.

Download:

Table 1. Baseline characteristics classified by CHA₂DS₂-VASc risk category.

https://doi.org/10.1371/journal.pone.0330164.t001

Deterioration of adverse events in higher CHA₂DS₂-VASc -risk categories

Outcomes and hazard ratios (HR) classified by risk category of the CHA₂DS₂-VASc score are shown in Table 2. During a median follow-up of 23 months (IQR 12–35), the composite endpoint defined as stroke, MI or all-cause mortality event occurred in 5.3% for low-risk patients, 7.3% for intermediate-risk patients and 28.6% for patients within the high-risk category. A Kaplan-Meier analysis classified by CHA₂DS₂-VASc-risk category for the composite endpoint is shown in Fig 2. Here, patients within the intermediate-risk category showed an HR of 1.43 (95%CI: 1.16–1.76, p < 0.0001) for an adverse outcome compared to low-risk category and an HR of 4.53 (95%CI: 3.97–5.15, p < 0.0001) for an adverse outcome compared to high-risk category. Compared to patients with low-risk, patients within the intermediate CHA₂DS₂-VASc-risk category had an HR of 2.39 (95% CI: 1.28–4.47, p < 0.0001) for stroke and an HR of 3.73 (95%CI: 2.31–6.03, p < 0.0001) for a major bleeding event. Kaplan-Meier analyses for separate outcome parameters and other outcomes recorded within HERA-FIB, such as all-cause mortality, stroke, major bleeding and myocardial infarction (MI) stratified by CHA₂DS₂-VASc score risk category is shown in S2 Fig.

Download:

Table 2. Outcomes and HR classified by CHA₂DS₂-VASc risk category.

https://doi.org/10.1371/journal.pone.0330164.t002

Download:

Fig 2. Kaplan Meier analysis for composite endpoint consisting of stroke, MI or all-cause mortality showed an increase in outcomes stratified by CHA₂DS₂-VASc risk category.

Patients within higher risk categories of the CHA₂DS₂-VASc score showed a higher probability for the composite endpoint. Abbreviations: EP, endpoint.

https://doi.org/10.1371/journal.pone.0330164.g002

Modest prognostic performance of CHA₂DS₂-VASc score

For the entire cohort, the AUC for the CHA₂DS₂-VASc score in prediction of the composite endpoint was 0.648 (95%CI: 0.638–0.657, p < 0.0001). For the individual components of the composite endpoint, CHA₂DS₂-VASc score showed only a modest performance with an AUC of 0.623 (95%CI: 0.612–0.633, p < 0.0001) for stroke, an AUC of 0.632 (95%CI: 0.621–0.642, p < 0.0001) for MI and an AUC of 0.643 (95%CI: 0.634–0.652, p < 0.0001) for all-cause mortality. For major bleeding events, AUC of the CHA₂DS₂-VASc score was 0.604 (95%CI: 0.594–0.615, p < 0.0001) for the entire cohort.

Improved prediction of adverse outcomes adding hs-cTnT

Since the CHA₂DS₂-VASc score showed only a modest performance in prediction of the composite endpoint, we thought to evaluate the role of a biomarker enhanced risk score utilizing hs-cTnT. For the prediction of the composite endpoint, hs-cTnT alone showed an AUC of 0.745 (95%CI: 0.737–0.754, p < 0.0001) within the entire study cohort. AUCs for hs-cTnT separated for the individual endpoints are shown in S1 Table. A Kaplan-Meier analysis for the composite endpoint and hs-cTnT categories is shown in Fig 3A. HRs for the composite endpoint increased among hs-cTnT categories. Compared to patients with hs-cTnT serum levels <5 ng/L, patients with a hs-cTnT serum level between 5–14 ng/L had a HR of 4.06 (95%CI: 3.45–4.79) for the composite endpoint. In patients with hs-cTnT serum level >14 ng/L, HR for the composite endpoint was 17.19 (95%CI: 14.65–20.18), respectively. Kaplan-Meier analyses separated for available outcomes classified by hs-cTnT levels (hs-cTnT < 5 ng/l, hs-cTnT 5–14 ng/L and hs-cTnT > 14 ng/L) are shown in S3 Fig. To quantify the incremental predictive value of hs-cTnT when added to the CHA₂DS₂-VASc score, we calculated the NRI, which was 0.83 (95% CI: 0.56–0.98; p = 0.0099), indicating substantial improvement in risk stratification for adverse outcomes.

Download:

Fig 3. Kaplan Meier analysis for the composite endpoint stratified by hs-cTnT categories for the entire cohort (A) and patients in intermediate-risk category according to CHA₂DS₂-VASc score (B).

Abbreviations: EP, endpoint; hs-cTnT, highly sensitive cardiac troponin T.

https://doi.org/10.1371/journal.pone.0330164.g003

Improvement of risk prediction by adding hs-cTnT in intermediate-risk AF patients

In AF patients with intermediate-risk CHA₂DS₂-VASc, AUC of the CHA₂DS₂-VASc score in prediction of the composite endpoint was 0.542 (95%CI: 0.508–0.575, p = 0.2487). Instrumenting hs-cTnT serum levels for risk prediction in patients with intermediate CHA₂DS₂-VASc score showed a significantly higher AUC for the prediction of the composite endpoint of 0.778 (95%CI: 0.748–0.805), ∆AUC 0.236 (95%CI 0.146–0.325, p < 0.001). Kaplan Meier analysis for patients within the intermediate-risk CHA₂DS₂-VASc category, stratified by hs-cTnT categories (hs-cTnT < 5 ng/l, hs-cTnT 5–14 ng/L and hs-cTnT > 14 ng/L) is shown in Fig 3B. In patients with undetectable hs-cTnT serum levels, i.e., values below the limit of detection (LoD) <5 ng/L, no composite endpoint event occurred, whereas rates of the composite endpoint increased gradually from 4.87% in patients with hs-cTnT of 5–14 ng/L to 17.41% in those with hs-cTnT serum levels >14 ng/L. Accordingly, HR for patients with hs-cTnT serum levels of >14 ng/L compared to patients with hs-cTnT serum levels 5–14 ng/L was 3.92 (95%CI: 2.16–7.09, p < 0.0001). Consistently, the addition of hs-cTnT to the CHA₂DS₂-VASc score in intermediate-risk patients significantly improved risk classification, as reflected by a NRI of 0.24 (95% CI: 0.12–0.47; p < 0.0001).

Prediction of the composite endpoint in patients with intermediate CHA₂DS₂-VASc score treated with oral anticoagulants

Among 871 patients within the intermediate CHA₂DS₂-VASc category, a total of 532 (61.1%) patients were treated with oral anticoagulants. In OAC treated patients, AUC CHA₂DS₂-VASc score in prediction of the composite endpoint was 0.515 (95%CI: 0.472–0.558, p = 0.7958). For stroke prediction AUC was 0.572 (95%CI: 0.526–0.617, p = 0.5974) and for the prediction of a major bleeding event AUC was 0.641 (95%CI: 0.596–0.684, p = 0.0782). However, when hs-cTnT was added, AUC improved to 0.755 (95%CI: 0.716–0.791), p < 0.0001 in prediction of the composite endpoint.

Prediction of the composite endpoint in patients with intermediate CHA₂DS₂-VASc score not treated with oral anticoagulants

In 38.9% of AF patients not treated with oral anticoagulants and categorized as intermediate risk per CHA₂DS₂-VASc, AUC for the CHA₂DS₂-VASc score in prediction of the composite endpoint was 0.559 (95%CI: 0.504–0.613, p = 0.2076). However, when hs-cTnT was added, AUC significantly improved to 0.808 (95%CI: 0.62–0.848, p < 0.0001) for prediction of the composite endpoint.

Comparison of CHA₂DS₂-VASc to CHA₂DS₂-VA in risk assessment of patients within the intermediate risk group

According to the CHA₂DS₂-VASc score, 871 patients were classified as intermediate risk (1 point in men, 2 points in women). However, reclassification using the CHA₂DS₂-VA score resulted in a different distribution of risk groups, with 564 patients (5.6%) classified as low risk, 967 patients (9.7%) as intermediate risk, and 8,464 patients (84.7%) as high risk within the HERA-FIB cohort. As a result, the distribution of outcome parameters across risk categories differed between the two scoring models, as detailed in S2 Table. For the entire cohort, the CHA₂DS₂-VA score showed limited discriminatory ability for predicting the composite endpoint (AUC: 0.650, 95% CI: 0.641–0.660, p < 0.0001). A summary of AUC values for all outcome parameters is provided in S3 Table. Among patients with an intermediate CHA₂DS₂-VA score, the predictive value for the composite endpoint was poor (AUC: 0.509, 95% CI: 0.475–0.542, p = 0.8185). However, when hs-cTnT was incorporated within patients with intermediate CHA₂DS₂-VA score, the predictive performance significantly improved, with an AUC of 0.778 (95% CI: 0.750–0.803, p < 0.0001). A Kaplan-Meier analysis for the composite endpoint in intermediate-risk patients per CHA₂DS₂-VA score is shown in S4 Fig. Among the 967 patients classified as intermediate risk by CHA₂DS₂-VA, 611 (63.2%) received OAC, while 356 (36.8%) remained without OAC treatment. For the prediction of the composite endpoint, the CHA₂DS₂-VA score alone showed poor discrimination in both OAC-treated patients (AUC: 0.500, 95% CI: 0.460–0.540) and non-OAC patients (AUC: 0.500, 95% CI: 0.447–0.553). However, the addition of hs-cTnT significantly improved risk prediction, showing an AUC of 0.733 (95% CI: 0.696–0.767, p < 0.0001) in OAC-treated patients and 0.825 (95% CI: 0.782–0.863, p < 0.0001) in non-OAC patients.

Risk assessment using a novel hs-cTnCHA₂DS₂VASc-score

Considering that serum levels of hs-cTnT > 5 ng/L and hs-cTnT > 14 ng/L remained independent predictors for the composite endpoint in patients with intermediate-risk CHA₂DS₂-VASc score, we evaluated an extension of the original score by adding 0 points for hs-cTnT < 5 ng/L, 1 point for hs-cTnT 5–14 ng/L and 2 points for hs-cTnT > 14 ng/L. The median for hs-cTnCHA₂DS₂VASc-score in patients within the intermediate risk category was 2 points (IQR 2–3).

Importantly, no events were observed in intermediate risk patients with hs-cTnT < 5 ng/L, while the event rate increased to 5.0% in the 5–14 ng/L group and reached 17.5% in those with hs-cTnT > 14 ng/L, indicating a clear gradient of risk across hs-cTnT strata.

In all patients, a comparison between the original CHA₂DS₂VASc score and the hs-cTnCHA₂DS₂VASc-score in prediction of the composite endpoint revealed an AUC of 0.647 (95%CI 0.638–0.657) for the original CHA₂DS₂VASc-score and an AUC of 0.683 (95%CI 0.674–0.692) for the hs-cTnCHA₂DS₂VASc-score, ∆AUC 0.0358 (95%CI 0.0325–0.0390, p < 0.0001). In patients with oral anticoagulants and an intermediate CHA₂DS₂-VASc score, AUC for hs-cTnCHA2DS2VASc-score in prediction of the composite endpoint was 0.683 (95%CI: 0.642–0.723) showing a ∆AUC of 0.168 (95%CI: 0.0767–0.260, p = 0.0003) in comparison to the original CHA₂DS₂-VASc score. In intermediate risk patients without oral anticoagulants AUC for hs-cTnCHA2DS2VASc-score in prediction of the composite endpoint was 0.686 (95%CI: 0.631–0.732, ∆AUC in comparison to the original CHA₂DS₂-VASc score was 0.124 (95%CI: 0.0367–0.212, p = 0.0054).

Discussion

This study sought to investigate whether hs-cTnT might improve prediction of a composite endpoint including stroke, MI and all-cause mortality in AF patients at an intermediate-risk according to the CHA₂DS₂-VASc score. The later has only modest performance for prediction of the composite endpoint and does not help for the decision whether AF patients at intermediate risk should receive an anticoagulation. Our findings show that the addition of hs-cTnT to the CHA₂DS₂-VASc score improves the prediction of the composite endpoint in AF patients presenting in the setting of an ED. A unique finding of our study is that patients with undetectable hs-cTnT represent a very low risk cohort in respective to stroke, MI or all-cause mortality.

In order to identify appropriate candidates for oral anticoagulation, guidelines on management of AF recommend a clinical risk-factor based assessment for prediction of stroke and major bleeding events [7]. However, scoring systems recommended by guidelines show only a modest performance and fail to provide clear recommendations on important subgroups of patients, such as patients with intermediate-risk CHA₂DS₂-VASc score (men: 1 point, women: 2 points) [7,22,27]. Beyond its use in AF-related thromboembolic risk stratification, emerging evidence highlights that the CHA₂DS₂-VASc score has broader prognostic utility in cardiovascular conditions. Studies have demonstrated its predictive value for major adverse cardiovascular events (MACE) in patients with coronary artery disease (CAD), acute coronary syndrome (ACS), and even in individuals without AF. For example, Hioki et al. [28] showed that the CHA₂DS₂-VASc score effectively stratifies risk in patients undergoing percutaneous coronary intervention (PCI), while Sen et al. [29] found its association with cardiovascular mortality and MACE in ACS patients. Furthermore, it has been applied to non-thromboembolic cardiovascular risks, including mortality in hospitalized cardiovascular patients and even pneumonia-related cardiovascular events. Lee et al. [30] demonstrated its relevance in predicting hospital readmission rates among patients receiving tele-health monitoring, while Wang et al. [31] showed that it stratifies MACE risk in pneumonia patients. Additionally, Cheng et al. [32] validated its prognostic role in PAD patients undergoing angioplasty.

The 2024 ESC Guidelines now recommend the use of the CHA₂DS₂-VA score instead of CHA₂DS₂-VASc for thromboembolic risk stratification in AF patients [13]. However, the role of sex as a risk factor in anticoagulation decision-making remains controversial. Large population-based studies have yielded conflicting results regarding the predictive superiority of CHA₂DS₂-VA over CHA₂DS₂-VASc [33]. For example, an analysis of UK primary and secondary care data from 195,719 AF patients found that removing female sex from the CHA₂DS₂-VASc score did not reduce its ability to discriminate thromboembolic events, challenging the necessity of this adjustment [34]. In our cohort, we compared both risk scores in patients classified as intermediate risk. We found that the predictive ability of CHA₂DS₂-VA for stroke, MI and all-cause mortality remained poor in patients at intermediate risk (AUC: 0.509, 95% CI: 0.475–0.542 p = 1.00), similar to CHA₂DS₂-VASc. However, when hs-cTnT was incorporated, predictive performance significantly improved for both scores, irrespective of oral anticoagulation. Specifically, the AUC for the composite endpoint increased to 0.778 (95% CI: 0.750–0.803, p < 0.0001). The improvement was observed both in patients receiving OAC and in those not treated with anticoagulation, with a more pronounced effect in non-OAC patients. These findings highlight the limitations of both CHA₂DS₂-VASc and CHA₂DS₂-VA in risk stratification for intermediate-risk AF patients and suggest that biomarker-enhanced models such as hs-cTnT could improve risk assessment in clinical routine. Given that guideline recommendations remain uncertain in this subgroup, the integration of hs-cTnT may help individualize anticoagulation strategies in patients who otherwise fall into a risk “gray zone”. Our results reveal that patients classified as intermediate-risk based on the CHA₂DS₂-VASc score had a significantly higher HR for adverse outcomes when compared to low-risk patients. Limitations of the CHA₂DS₂-VASc score prevail across all risk categories but individualization of therapy has the highest clinical relevance among patients at intermediate risk, where clear recommendations for the initiation of anticoagulation are missing. Here, we could show a more accurate discrimination of risk assessment by integrating hs-cTnT serum levels in risk prediction for stroke, MI and all-cause mortality. The significant increase in HR, as hs-cTnT levels rose indicates its potential as a valuable biomarker for refining risk stratification in AF patients. The most relevant finding is that patients with undetectable hs-cTnT represent a cohort at very low risk experiencing no adverse outcome events of stroke, MI or all-cause mortality during follow-up. In fact, within the intermediate-risk group, no events occurred in patients with hs-cTnT < 5 ng/L, whereas the event rate for the composite outcome (stroke, MI, and all-cause mortality) increased to 5.0% in those with hs-cTnT between 5–14 ng/L and reached 17.5% in those with hs-cTnT > 14 ng/L, showing a clear, step wise increase in adverse outcomes. These findings suggest that intermediate-risk patients with hs-cTnT < 5 ng/L represent a cohort of extremely low risk for adverse events. The present study contributes to the continuous evolution of risk stratification in AF patients presenting to an emergency department. Historically, stroke risk prediction models, including CHA₂DS₂-VASc, were developed in cohorts of untreated AF patients. However, in contemporary practice, stroke risk in AF represents a mixture of residual risk despite OAC and the true risk in untreated patients. As Lip et al. (2023) emphasize, interpreting stroke risk in modern AF cohorts must account for the impact of anticoagulation use. [35] In our study, 61% of intermediate-risk patients received OAC, while 39% remained untreated, reflecting this dual-risk profile. Thus, even when separated by OAC or non-OAC therapy, the addition of hs-cTnT to the CHA₂DS₂-VASc score added relevant predictive information for the composite endpoint of stroke, MI and all-cause mortality

Given the broad cardiovascular implications of the CHA₂DS₂-VASc score, the use of a composite endpoint consisting of stroke, MI and all-cause mortality our study is justified. Previous studies have shown that ischemic and hemorrhagic risks are closely interlinked, as seen in patients with CAD, heart failure, and peripheral artery disease. Thus, incorporating both stroke and major bleeding into the risk assessment provides a more holistic approach to patient management.

The reasons why hs-cTn allows prediction of adverse outcomes in AF patients is not fully understood but most likely multifactorial as cardiac troponin summarizes the cumulative impact of comorbidities that trigger myocardial injury. In agreement with previous observations bleeding risk is closely linked to ischemic risk [36–39]. A subanalysis from the RE-LY study (Randomized Evaluation of Long-Term Anticoagulation Therapy), which was initially designed to investigate the appropriateness of dabigatran in comparison of warfarin, investigated the role of elevated cardiac troponin I in a preselected cohort of patients with new onset AF and found that cTnI serum levels were indepenently related to stroke [40]. However, the authors could not provide data on cTnT. Moreover, the RE-LY trial did not provide information for patients in real-world and particularly in the setting of admission to an ED. The hypothesis that biomarkers might improve prediction of adverse outcomes including stroke and adverse outcomes is not novel. Previously, Hijazi et al. introduced and validated the ABC score, a risk assessment tool that incorporates clinical information and biomarkers [9]. While prediction of adverse outcomes was improved, the implementation into clinical routine lags behind expectations outside randomized trials. The combination of the established CHA₂DS₂VASc-score with the single broadly available hs-cTnT or hs-cTnI which is used as a dichotomous variable is more convenient and more likely to be adopted in routine. However, future external validation of this combined risk assessment strategy is required before our findings will help to individualize anticoagulation in AF patients at intermediate stroke risk.

Our study introduces the hs-cTnCHA₂DS₂VASc-score as a novel predictive tool for assessing the combined endpoint of stroke, MI and all-cause mortality in AF patients with intermediate-risk in CHA₂DS₂-VASc score. Despite being widely used in clinical routine, our study underscores the modest performance of CHA₂DS₂-VASc score emphasizing the need for innovative and simple approaches to improve risk prediction in AF patients presenting in the acute setting of an ED. Particularly, the enhanced predictive value of hs-cTnT aligns with previous research suggesting that biomarkers – especially cardiac biomarkers – as demonstrated by studies [14–16,26,41], may significantly contribute to risk stratification for AF patients. Notably, the inclusion of hs-cTnT significantly enhances the discriminatory power of the risk assessment tool, within the intermediate-risk CHA₂DS₂-VASc-category. Beyond improvements in discrimination, we also evaluated reclassification performance. For the whole cohort, adding hs-cTnT to the CHA₂DS₂-VASc score resulted in an NRI of 0.83 (95% CI: 0.56–0.98; p = 0.0099), indicating substantial improvement in risk categorization. Among intermediate-risk patients, the NRI was 0.24 (95% CI: 0.12–0.47; p < 0.0001), reflecting clinically meaningful reclassification within this “gray-zone” population. This underscores the clinical value of hs-cTnT in refining risk assessment for AF patients, particularly in the intermediate-risk group, potentially offering a more precise basis for decisions regarding anticoagulation therapy. Moreover, given the increasing use of CHA₂DS₂-VASc in cardiovascular risk assessment beyond AF, our study provides further insights into how this score can be adapted for broader clinical applications.

Limitations and future directions

It is important to acknowledge the limitations of our study. First, our findings were derived from a single tertiary referral center at an academic center in Germany and may not be generalized to other geographic regions or health care systems. Second, our findings on the predictive role of hs-cTnT alone and particularly the novel hs-cTnCHA₂DS₂VASc-score still requires prospective validation in independent populations. Third, our findings are based exclusively on hs-cTnT and no blood specimen are available for evaluation of other hs-cTnI assays. Fourth, the relatively short median follow-up duration of 23 months may not fully capture the long-term effects of risk assessment using the hs-cTnCHA₂DS₂VASc-score. Fifth, we studied the added value of hs-cTnT when combined with the CHA₂DS₂-VASc score but did not evaluate other risk prediction models such as the ABC score which already contains biomarkers. Finally, temporal changes in patient management and healthcare practices have changed over the study period, spanning from June 2009 to March 2020. In particular, we cannot exclude that changes in demographic characteristics such as age, as well as the increasing preference of DOAC versus VKA may have an impact on risk of stroke and other outcomes in contemporary clinical practice. Within HERA-FIB, the majority of patients were stratified as high risk per CHA₂DS₂-VASc. However, the current analysis focused exclusively on patients with intermediate risk, as they represent the subgroup where clinical decision-making regarding anticoagulation therapy and risk stratification remains most uncertain. While our findings show that hs-cTnT improves risk stratification in this population, an important area for future research will be the refinement of risk assessment strategies in high-risk patients. Additional studies should evaluate whether biomarker-based models, such as the hs-cTnT-enhanced CHA₂DS₂-VASc score, could provide further prognostic discrimination within the high-risk category and identify subgroups with particularly elevated or lower risk within this population.

Conclusions

In conclusion, in patients with AF and intermediate thromboembolic risk, the decision to initiate oral anticoagulation remains uncertain due to the modest predictive performance of current risk scores. Our study shows that adding hs-cTnT to the CHA₂DS₂-VASc score substantially improves risk stratification for adverse cardiovascular outcomes, including stroke, myocardial infarction, and all-cause mortality, particularly within this intermediate-risk population. Importantly, no events occurred in intermediate-risk patients with hs-cTnT levels below the limit of detection, whereas the event rate increased stepwise in patients with elevated hs-cTnT. The addition of hs-cTnT yielded a significant net reclassification improvement in this group, supporting its potential role in risk stratification. In conclusion, our study shows a strong value of hs-cTnT when added to the CHA₂DS₂-VASc score These findings highlight the promise of a biomarker-enhanced CHA₂DS₂-VASc score for individualized risk assessment in AF patients at intermediate risk. However, prospective validation in independent cohorts is needed before clinical implementation.

Supporting information

S1 Fig. Distribution of CHA₂DS₂-VASc score across the entire cohort.

https://doi.org/10.1371/journal.pone.0330164.s001

(TIF)

S2 Fig. Kaplan-Meier analysis for all-cause mortality, stroke, major bleeding and myocardial infarction classified by CHA₂DS₂-VASc score risk category.

Patients with a high-risk category according to CHA₂DS₂VASc-score had a higher all-cause mortality (A), a higher probability of ischemic stroke rate (B), a higher probability of major bleeding events (C) and higher risk for myocardial infarction (D). Abbreviatons: MI, myocardial infarction.

https://doi.org/10.1371/journal.pone.0330164.s002

(TIF)

S3 Fig. Kaplan-Meier analysis for all-cause mortality, stroke, major bleeding and myocardial infarction classified by hs-cTnT category.

Patients with a high hs-cTnT had a higher all-cause mortality (A), a higher probability of ischemic stroke (B), a higher probability of major bleeding (C) and higher risk for myocardial infarction (D). Abbreviations: MI, myocardial infarction, hs-cTnT, highly sensitive cardiac troponin T.

https://doi.org/10.1371/journal.pone.0330164.s003

(TIF)

S4 Fig. Kaplan Meier analysis for the composite endpoint stratified by hs-cTnT categories for patients in intermediate-risk category according to CHA₂DS₂-VA score.

Abbreviations: EP, endpoint; hs-cTnT, highly sensitive cardiac troponin T.

https://doi.org/10.1371/journal.pone.0330164.s004

(TIF)

S1 Table. AUC analysis for hs-cTnT in predicting outcome variables. Abbreviations: AUC, area under the curve, CI, confidence interval.

https://doi.org/10.1371/journal.pone.0330164.s005

(DOCX)

S2 Table. Outcomes and HR classified by CHA₂DS₂-VA risk category.

*The composite EP consisted of stroke or major bleeding, p-value was calculated as p-value for trend and log rank test for HRs. Abbreviations: CI, confidence interval; HR, hazard ratio; MI, myocardial infarction.

https://doi.org/10.1371/journal.pone.0330164.s006

(DOCX)

S3 Table. AUCs of the CHA₂DS₂VA score separated for outcome variables.

*The composite EP consisted of stroke and major bleeding events. Abbreviations: AUC, area under the curve, CI, confidence interval, EP endpoint.

https://doi.org/10.1371/journal.pone.0330164.s007

(DOCX)

References

1. Wolf PA, Abbott RD, Kannel WB. Atrial fibrillation as an independent risk factor for stroke: the Framingham Study. Stroke. 1991;22(8):983–8. pmid:1866765
- View Article
- PubMed/NCBI
- Google Scholar
2. Benjamin EJ, Muntner P, Alonso A, Bittencourt MS, Callaway CW, Carson AP, et al. Heart Disease and Stroke Statistics-2019 Update: A Report From the American Heart Association. Circulation. 2019;139(10):e56–528. pmid:30700139
- View Article
- PubMed/NCBI
- Google Scholar
3. Hylek EM, Go AS, Chang Y, Jensvold NG, Henault LE, Selby JV, et al. Effect of intensity of oral anticoagulation on stroke severity and mortality in atrial fibrillation. N Engl J Med. 2003;349(11):1019–26. pmid:12968085
- View Article
- PubMed/NCBI
- Google Scholar
4. Bode W, Ptaszek LM. Management of Atrial Fibrillation in the Emergency Department. Curr Cardiol Rep. 2021;23(12):179. pmid:34657210
- View Article
- PubMed/NCBI
- Google Scholar
5. Deshmukh A, Iglesias M, Khanna R, Beaulieu T. Healthcare utilization and costs associated with a diagnosis of incident atrial fibrillation. Heart Rhythm O2. 2022;3(5):577–86. pmid:36340482
- View Article
- PubMed/NCBI
- Google Scholar
6. Milman B, Burns BD. Atrial fibrillation: an approach to diagnosis and management in the emergency department. Emerg Med Pract. 2021;23(5):1–28. pmid:33885254
- View Article
- PubMed/NCBI
- Google Scholar
7. Hindricks G, Potpara T, Dagres N, Arbelo E, Bax JJ, Blomström-Lundqvist C, 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):373–498. pmid:32860505
- View Article
- PubMed/NCBI
- Google Scholar
8. Gorog DA, Gue YX, Chao T-F, Fauchier L, Ferreiro JL, Huber K, et al. Assessment and mitigation of bleeding risk in atrial fibrillation and venous thromboembolism: A Position Paper from the ESC Working Group on Thrombosis, in collaboration with the European Heart Rhythm Association, the Association for Acute CardioVascular Care and the Asia-Pacific Heart Rhythm Society. Europace. 2022;24(11):1844–71. pmid:35323922
- View Article
- PubMed/NCBI
- Google Scholar
9. Hijazi Z, Oldgren J, Lindbäck J, Alexander JH, Connolly SJ, Eikelboom JW, et al. The novel biomarker-based ABC (age, biomarkers, clinical history)-bleeding risk score for patients with atrial fibrillation: a derivation and validation study. Lancet. 2016;387(10035):2302–11. pmid:27056738
- View Article
- PubMed/NCBI
- Google Scholar
10. Fang MC, Go AS, Chang Y, Borowsky LH, Pomernacki NK, Udaltsova N, et al. A new risk scheme to predict warfarin-associated hemorrhage: The ATRIA (Anticoagulation and Risk Factors in Atrial Fibrillation) Study. J Am Coll Cardiol. 2011;58(4):395–401. pmid:21757117
- View Article
- PubMed/NCBI
- Google Scholar
11. O’Brien EC, Simon DN, Thomas LE, Hylek EM, Gersh BJ, Ansell JE, et al. The ORBIT bleeding score: a simple bedside score to assess bleeding risk in atrial fibrillation. Eur Heart J. 2015;36(46):3258–64. pmid:26424865
- View Article
- PubMed/NCBI
- Google Scholar
12. Fox KAA, Lucas JE, Pieper KS, Bassand J-P, Camm AJ, Fitzmaurice DA, et al. Improved risk stratification of patients with atrial fibrillation: an integrated GARFIELD-AF tool for the prediction of mortality, stroke and bleed in patients with and without anticoagulation. BMJ Open. 2017;7(12):e017157. pmid:29273652
- View Article
- PubMed/NCBI
- Google Scholar
13. Van Gelder IC, Rienstra M, Bunting KV, Casado-Arroyo R, Caso V, Crijns HJGM, et al. 2024 ESC Guidelines for the management of atrial fibrillation developed in collaboration with the European Association for Cardio-Thoracic Surgery (EACTS). Eur Heart J. 2024;45(36):3314–414. pmid:39210723
- View Article
- PubMed/NCBI
- Google Scholar
14. Berg DD, Ruff CT, Jarolim P, Giugliano RP, Nordio F, Lanz HJ, et al. Performance of the ABC Scores for Assessing the Risk of Stroke or Systemic Embolism and Bleeding in Patients With Atrial Fibrillation in ENGAGE AF-TIMI 48. Circulation. 2019;139(6):760–71. pmid:30586727
- View Article
- PubMed/NCBI
- Google Scholar
15. Berg DD, Ruff CT, Morrow DA. Biomarkers for Risk Assessment in Atrial Fibrillation. Clin Chem. 2021;67(1):87–95. pmid:33313695
- View Article
- PubMed/NCBI
- Google Scholar
16. Hijazi Z, Oldgren J, Siegbahn A, Wallentin L. Application of Biomarkers for Risk Stratification in Patients with Atrial Fibrillation. Clin Chem. 2017;63(1):152–64. pmid:27811208
- View Article
- PubMed/NCBI
- Google Scholar
17. Yildirim M, Salbach C, Reich C, Milles BR, Biener M, Frey N, et al. Comparison of the clinical chemistry score to other biomarker algorithms for rapid rule-out of acute myocardial infarction and risk stratification in patients with suspected acute coronary syndrome. Int J Cardiol. 2024;400:131815. pmid:38278492
- View Article
- PubMed/NCBI
- Google Scholar
18. Yildirim M, Reich C, Salbach C, Biener M, Mueller-Hennessen M, Sörensen NA, et al. Identification of patients with suspected NSTE-ACS in the observe zone: evaluating GRACE 1.0 score and a biomarker panel for risk stratification and management optimization. Clin Res Cardiol. 2025;114(6):783–95. pmid:40227426
- View Article
- PubMed/NCBI
- Google Scholar
19. Shin SY, Han S-J, Kim J-S, Im SI, Shim J, Ahn J, et al. Identification of Markers Associated With Development of Stroke in “Clinically Low-Risk” Atrial Fibrillation Patients. J Am Heart Assoc. 2019;8(21):e012697. pmid:31668140
- View Article
- PubMed/NCBI
- Google Scholar
20. Salbach C, Yildirim M, Hund H, Biener M, Mueller-Hennessen M, Frey N, et al. Design, rationale and initial findings from the Heidelberg Registry of Atrial Fibrillation on 10,222 patients with atrial fibrillation presenting to an emergency department over an 11-year period. Manuscript submitted for publication2024.
21. Lip GYH, Collet JP, Caterina R de, Fauchier L, Lane DA, Larsen TB, et al. Antithrombotic therapy in atrial fibrillation associated with valvular heart disease: a joint consensus document from the European Heart Rhythm Association (EHRA) and European Society of Cardiology Working Group on Thrombosis, endorsed by the ESC Working Group on Valvular Heart Disease, Cardiac Arrhythmia Society of Southern Africa (CASSA), Heart Rhythm Society (HRS), Asia Pacific Heart Rhythm Society (APHRS), South African Heart (SA Heart) Association and Sociedad Latinoamericana de Estimulación Cardíaca y Electrofisiología (SOLEACE). Europace. 2017;19(11):1757–8. pmid:29096024
- View Article
- PubMed/NCBI
- Google Scholar
22. Lip GYH, Nieuwlaat R, Pisters R, Lane DA, Crijns HJGM. Refining clinical risk stratification for predicting stroke and thromboembolism in atrial fibrillation using a novel risk factor-based approach: the euro heart survey on atrial fibrillation. Chest. 2010;137(2):263–72. pmid:19762550
- View Article
- PubMed/NCBI
- Google Scholar
23. Grambsch PM, Therneau TM, Fleming TR. Diagnostic plots to reveal functional form for covariates in multiplicative intensity models. Biometrics. 1995;51(4):1469–82. pmid:8589234
- View Article
- PubMed/NCBI
- Google Scholar
24. Hanley JA, McNeil BJ. The meaning and use of the area under a receiver operating characteristic (ROC) curve. Radiology. 1982;143(1):29–36. pmid:7063747
- View Article
- PubMed/NCBI
- Google Scholar
25. Hanley JA, McNeil BJ. A method of comparing the areas under receiver operating characteristic curves derived from the same cases. Radiology. 1983;148(3):839–43. pmid:6878708
- View Article
- PubMed/NCBI
- Google Scholar
26. Pencina MJ, D’Agostino RB Sr, D’Agostino RB, Vasan RS. Evaluating the added predictive ability of a new marker: from area under the ROC curve to reclassification and beyond. Stat Med. 2008;27(2):157–72; discussion 207-12. pmid:17569110
- View Article
- PubMed/NCBI
- Google Scholar
27. Chen J-Y, Zhang A-D, Lu H-Y, Guo J, Wang F-F, Li Z-C. CHADS2 versus CHA2DS2-VASc score in assessing the stroke and thromboembolism risk stratification in patients with atrial fibrillation: a systematic review and meta-analysis. J Geriatr Cardiol. 2013;10(3):258–66. pmid:24133514
- View Article
- PubMed/NCBI
- Google Scholar
28. Hioki H, Miura T, Miyashita Y, Motoki H, Shimada K, Kobayashi M, et al. Risk stratification using the CHA2DS2-VASc score in patients with coronary heart disease undergoing percutaneous coronary intervention; sub-analysis of SHINANO registry. Int J Cardiol Heart Vasc. 2015;7:76–81. pmid:28785649
- View Article
- PubMed/NCBI
- Google Scholar
29. Sen J, Tonkin A, Varigos J, Fonguh S, Berkowitz SD, Yusuf S, et al. CHA2DS2-VASc and CHADS2 scores for risk stratification of major adverse cardiovascular events in the COMPASS trial. European Heart Journal. 2020;41(Supplement_2).
- View Article
- Google Scholar
30. Lee J-K, Hung C-S, Huang C-C, Chen Y-H, Chuang P-Y, Yu J-Y, et al. Use of the CHA2DS2-VASc Score for Risk Stratification of Hospital Admissions Among Patients With Cardiovascular Diseases Receiving a Fourth-Generation Synchronous Telehealth Program: Retrospective Cohort Study. J Med Internet Res. 2019;21(1):e12790. pmid:30702437
- View Article
- PubMed/NCBI
- Google Scholar
31. Wang B-Y, Lin F-Y, Ku M-S, Wang Y-H, Lee K-Y, Ho S-W. CHA2DS2-VASc Score for Major Adverse Cardiovascular Events Stratification in Patients with Pneumonia with and without Atrial Fibrillation. J Clin Med. 2021;10(18):4093. pmid:34575202
- View Article
- PubMed/NCBI
- Google Scholar
32. Cheng Y-T, Chang F-L, Li P-H, Lu W-C, Chiu C-S. Assessing the Suitability of CHA2DS2-VASc for Predicting Adverse Limb Events and Cardiovascular Outcomes in Peripheral Artery Disease Patients with Percutaneous Transluminal Angioplasty: A Retrospective Cohort Study. Biomedicines. 2024;12(6):1374. pmid:38927581
- View Article
- PubMed/NCBI
- Google Scholar
33. Champsi A, Mobley AR, Subramanian A, Nirantharakumar K, Wang X, Shukla D, et al. Gender and contemporary risk of adverse events in atrial fibrillation. Eur Heart J. 2024;45(36):3707–17. pmid:39217497
- View Article
- PubMed/NCBI
- Google Scholar
34. Yoshimura H, Providencia R, Finan C, Schmidt AF, Lip GYH. Refining the CHA2DS2VASc risk stratification scheme: shall we drop the sex category criterion?. Europace. 2024;26(11):euae280. pmid:39522169
- View Article
- PubMed/NCBI
- Google Scholar
35. Lip GYH, Proietti M, Potpara T, Mansour M, Savelieva I, Tse HF, et al. Atrial fibrillation and stroke prevention: 25 years of research at EP Europace journal. Europace. 2023;25(9):euad226. pmid:37622590
- View Article
- PubMed/NCBI
- Google Scholar
36. Yeh RW, Secemsky EA, Kereiakes DJ, Normand S-LT, Gershlick AH, Cohen DJ, et al. Development and Validation of a Prediction Rule for Benefit and Harm of Dual Antiplatelet Therapy Beyond 1 Year After Percutaneous Coronary Intervention. JAMA. 2016;315(16):1735–49. pmid:27022822
- View Article
- PubMed/NCBI
- Google Scholar
37. Costa F, van Klaveren D, James S, Heg D, Räber L, Feres F, et al. Derivation and validation of the predicting bleeding complications in patients undergoing stent implantation and subsequent dual antiplatelet therapy (PRECISE-DAPT) score: a pooled analysis of individual-patient datasets from clinical trials. Lancet. 2017;389(10073):1025–34. pmid:28290994
- View Article
- PubMed/NCBI
- Google Scholar
38. Ueki Y, Bär S, Losdat S, Otsuka T, Zanchin C, Zanchin T, et al. Validation of the Academic Research Consortium for High Bleeding Risk (ARC-HBR) criteria in patients undergoing percutaneous coronary intervention and comparison with contemporary bleeding risk scores. EuroIntervention. 2020;16(5):371–9. pmid:32065586
- View Article
- PubMed/NCBI
- Google Scholar
39. Brabrand M, Lassen AT, Knudsen T, Hallas J. Seven-day mortality can be predicted in medical patients by blood pressure, age, respiratory rate, loss of independence, and peripheral oxygen saturation (the PARIS score): a prospective cohort study with external validation. PLoS One. 2015;10(4):e0122480. pmid:25867881
- View Article
- PubMed/NCBI
- Google Scholar
40. Hijazi Z, Oldgren J, Andersson U, Connolly SJ, Ezekowitz MD, Hohnloser SH, et al. Cardiac biomarkers are associated with an increased risk of stroke and death in patients with atrial fibrillation: a Randomized Evaluation of Long-term Anticoagulation Therapy (RE-LY) substudy. Circulation. 2012;125(13):1605–16. pmid:22374183
- View Article
- PubMed/NCBI
- Google Scholar
41. Providência R, Paiva L, Barra S. Risk stratification of patients with atrial fibrillation: Biomarkers and other future perspectives. World J Cardiol. 2012;4(6):195–200. pmid:22761972
- View Article
- PubMed/NCBI
- Google Scholar

[ref1] 1. Wolf PA, Abbott RD, Kannel WB. Atrial fibrillation as an independent risk factor for stroke: the Framingham Study. Stroke. 1991;22(8):983–8. pmid:1866765
View Article
PubMed/NCBI
Google Scholar

[2] View Article

[3] PubMed/NCBI

[4] Google Scholar

[ref2] 2. Benjamin EJ, Muntner P, Alonso A, Bittencourt MS, Callaway CW, Carson AP, et al. Heart Disease and Stroke Statistics-2019 Update: A Report From the American Heart Association. Circulation. 2019;139(10):e56–528. pmid:30700139
View Article
PubMed/NCBI
Google Scholar

[6] View Article

[7] PubMed/NCBI

[8] Google Scholar

[ref3] 3. Hylek EM, Go AS, Chang Y, Jensvold NG, Henault LE, Selby JV, et al. Effect of intensity of oral anticoagulation on stroke severity and mortality in atrial fibrillation. N Engl J Med. 2003;349(11):1019–26. pmid:12968085
View Article
PubMed/NCBI
Google Scholar

[10] View Article

[11] PubMed/NCBI

[12] Google Scholar

[ref4] 4. Bode W, Ptaszek LM. Management of Atrial Fibrillation in the Emergency Department. Curr Cardiol Rep. 2021;23(12):179. pmid:34657210
View Article
PubMed/NCBI
Google Scholar

[14] View Article

[15] PubMed/NCBI

[16] Google Scholar

[ref5] 5. Deshmukh A, Iglesias M, Khanna R, Beaulieu T. Healthcare utilization and costs associated with a diagnosis of incident atrial fibrillation. Heart Rhythm O2. 2022;3(5):577–86. pmid:36340482
View Article
PubMed/NCBI
Google Scholar

[18] View Article

[19] PubMed/NCBI

[20] Google Scholar

[ref6] 6. Milman B, Burns BD. Atrial fibrillation: an approach to diagnosis and management in the emergency department. Emerg Med Pract. 2021;23(5):1–28. pmid:33885254
View Article
PubMed/NCBI
Google Scholar

[22] View Article

[23] PubMed/NCBI

[24] Google Scholar

[ref7] 7. Hindricks G, Potpara T, Dagres N, Arbelo E, Bax JJ, Blomström-Lundqvist C, 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):373–498. pmid:32860505
View Article
PubMed/NCBI
Google Scholar

[26] View Article

[27] PubMed/NCBI

[28] Google Scholar

[ref8] 8. Gorog DA, Gue YX, Chao T-F, Fauchier L, Ferreiro JL, Huber K, et al. Assessment and mitigation of bleeding risk in atrial fibrillation and venous thromboembolism: A Position Paper from the ESC Working Group on Thrombosis, in collaboration with the European Heart Rhythm Association, the Association for Acute CardioVascular Care and the Asia-Pacific Heart Rhythm Society. Europace. 2022;24(11):1844–71. pmid:35323922
View Article
PubMed/NCBI
Google Scholar

[30] View Article

[31] PubMed/NCBI

[32] Google Scholar

[ref9] 9. Hijazi Z, Oldgren J, Lindbäck J, Alexander JH, Connolly SJ, Eikelboom JW, et al. The novel biomarker-based ABC (age, biomarkers, clinical history)-bleeding risk score for patients with atrial fibrillation: a derivation and validation study. Lancet. 2016;387(10035):2302–11. pmid:27056738
View Article
PubMed/NCBI
Google Scholar

[34] View Article

[35] PubMed/NCBI

[36] Google Scholar

[ref10] 10. Fang MC, Go AS, Chang Y, Borowsky LH, Pomernacki NK, Udaltsova N, et al. A new risk scheme to predict warfarin-associated hemorrhage: The ATRIA (Anticoagulation and Risk Factors in Atrial Fibrillation) Study. J Am Coll Cardiol. 2011;58(4):395–401. pmid:21757117
View Article
PubMed/NCBI
Google Scholar

[38] View Article

[39] PubMed/NCBI

[40] Google Scholar

[ref11] 11. O’Brien EC, Simon DN, Thomas LE, Hylek EM, Gersh BJ, Ansell JE, et al. The ORBIT bleeding score: a simple bedside score to assess bleeding risk in atrial fibrillation. Eur Heart J. 2015;36(46):3258–64. pmid:26424865
View Article
PubMed/NCBI
Google Scholar

[42] View Article

[43] PubMed/NCBI

[44] Google Scholar

[ref12] 12. Fox KAA, Lucas JE, Pieper KS, Bassand J-P, Camm AJ, Fitzmaurice DA, et al. Improved risk stratification of patients with atrial fibrillation: an integrated GARFIELD-AF tool for the prediction of mortality, stroke and bleed in patients with and without anticoagulation. BMJ Open. 2017;7(12):e017157. pmid:29273652
View Article
PubMed/NCBI
Google Scholar

[46] View Article

[47] PubMed/NCBI

[48] Google Scholar

[ref13] 13. Van Gelder IC, Rienstra M, Bunting KV, Casado-Arroyo R, Caso V, Crijns HJGM, et al. 2024 ESC Guidelines for the management of atrial fibrillation developed in collaboration with the European Association for Cardio-Thoracic Surgery (EACTS). Eur Heart J. 2024;45(36):3314–414. pmid:39210723
View Article
PubMed/NCBI
Google Scholar

[50] View Article

[51] PubMed/NCBI

[52] Google Scholar

[ref14] 14. Berg DD, Ruff CT, Jarolim P, Giugliano RP, Nordio F, Lanz HJ, et al. Performance of the ABC Scores for Assessing the Risk of Stroke or Systemic Embolism and Bleeding in Patients With Atrial Fibrillation in ENGAGE AF-TIMI 48. Circulation. 2019;139(6):760–71. pmid:30586727
View Article
PubMed/NCBI
Google Scholar

[54] View Article

[55] PubMed/NCBI

[56] Google Scholar

[ref15] 15. Berg DD, Ruff CT, Morrow DA. Biomarkers for Risk Assessment in Atrial Fibrillation. Clin Chem. 2021;67(1):87–95. pmid:33313695
View Article
PubMed/NCBI
Google Scholar

[58] View Article

[59] PubMed/NCBI

[60] Google Scholar

[ref16] 16. Hijazi Z, Oldgren J, Siegbahn A, Wallentin L. Application of Biomarkers for Risk Stratification in Patients with Atrial Fibrillation. Clin Chem. 2017;63(1):152–64. pmid:27811208
View Article
PubMed/NCBI
Google Scholar

[62] View Article

[63] PubMed/NCBI

[64] Google Scholar

[ref17] 17. Yildirim M, Salbach C, Reich C, Milles BR, Biener M, Frey N, et al. Comparison of the clinical chemistry score to other biomarker algorithms for rapid rule-out of acute myocardial infarction and risk stratification in patients with suspected acute coronary syndrome. Int J Cardiol. 2024;400:131815. pmid:38278492
View Article
PubMed/NCBI
Google Scholar

[66] View Article

[67] PubMed/NCBI

[68] Google Scholar

[ref18] 18. Yildirim M, Reich C, Salbach C, Biener M, Mueller-Hennessen M, Sörensen NA, et al. Identification of patients with suspected NSTE-ACS in the observe zone: evaluating GRACE 1.0 score and a biomarker panel for risk stratification and management optimization. Clin Res Cardiol. 2025;114(6):783–95. pmid:40227426
View Article
PubMed/NCBI
Google Scholar

[70] View Article

[71] PubMed/NCBI

[72] Google Scholar

[ref19] 19. Shin SY, Han S-J, Kim J-S, Im SI, Shim J, Ahn J, et al. Identification of Markers Associated With Development of Stroke in “Clinically Low-Risk” Atrial Fibrillation Patients. J Am Heart Assoc. 2019;8(21):e012697. pmid:31668140
View Article
PubMed/NCBI
Google Scholar

[74] View Article

[75] PubMed/NCBI

[76] Google Scholar

[ref20] 20. Salbach C, Yildirim M, Hund H, Biener M, Mueller-Hennessen M, Frey N, et al. Design, rationale and initial findings from the Heidelberg Registry of Atrial Fibrillation on 10,222 patients with atrial fibrillation presenting to an emergency department over an 11-year period. Manuscript submitted for publication2024.

[ref21] 21. Lip GYH, Collet JP, Caterina R de, Fauchier L, Lane DA, Larsen TB, et al. Antithrombotic therapy in atrial fibrillation associated with valvular heart disease: a joint consensus document from the European Heart Rhythm Association (EHRA) and European Society of Cardiology Working Group on Thrombosis, endorsed by the ESC Working Group on Valvular Heart Disease, Cardiac Arrhythmia Society of Southern Africa (CASSA), Heart Rhythm Society (HRS), Asia Pacific Heart Rhythm Society (APHRS), South African Heart (SA Heart) Association and Sociedad Latinoamericana de Estimulación Cardíaca y Electrofisiología (SOLEACE). Europace. 2017;19(11):1757–8. pmid:29096024
View Article
PubMed/NCBI
Google Scholar

[79] View Article

[80] PubMed/NCBI

[81] Google Scholar

[ref22] 22. Lip GYH, Nieuwlaat R, Pisters R, Lane DA, Crijns HJGM. Refining clinical risk stratification for predicting stroke and thromboembolism in atrial fibrillation using a novel risk factor-based approach: the euro heart survey on atrial fibrillation. Chest. 2010;137(2):263–72. pmid:19762550
View Article
PubMed/NCBI
Google Scholar

[83] View Article

[84] PubMed/NCBI

[85] Google Scholar

[ref23] 23. Grambsch PM, Therneau TM, Fleming TR. Diagnostic plots to reveal functional form for covariates in multiplicative intensity models. Biometrics. 1995;51(4):1469–82. pmid:8589234
View Article
PubMed/NCBI
Google Scholar

[87] View Article

[88] PubMed/NCBI

[89] Google Scholar

[ref24] 24. Hanley JA, McNeil BJ. The meaning and use of the area under a receiver operating characteristic (ROC) curve. Radiology. 1982;143(1):29–36. pmid:7063747
View Article
PubMed/NCBI
Google Scholar

[91] View Article

[92] PubMed/NCBI

[93] Google Scholar

[ref25] 25. Hanley JA, McNeil BJ. A method of comparing the areas under receiver operating characteristic curves derived from the same cases. Radiology. 1983;148(3):839–43. pmid:6878708
View Article
PubMed/NCBI
Google Scholar

[95] View Article

[96] PubMed/NCBI

[97] Google Scholar

[ref26] 26. Pencina MJ, D’Agostino RB Sr, D’Agostino RB, Vasan RS. Evaluating the added predictive ability of a new marker: from area under the ROC curve to reclassification and beyond. Stat Med. 2008;27(2):157–72; discussion 207-12. pmid:17569110
View Article
PubMed/NCBI
Google Scholar

[99] View Article

[100] PubMed/NCBI

[101] Google Scholar

[ref27] 27. Chen J-Y, Zhang A-D, Lu H-Y, Guo J, Wang F-F, Li Z-C. CHADS2 versus CHA2DS2-VASc score in assessing the stroke and thromboembolism risk stratification in patients with atrial fibrillation: a systematic review and meta-analysis. J Geriatr Cardiol. 2013;10(3):258–66. pmid:24133514
View Article
PubMed/NCBI
Google Scholar

[103] View Article

[104] PubMed/NCBI

[105] Google Scholar

[ref28] 28. Hioki H, Miura T, Miyashita Y, Motoki H, Shimada K, Kobayashi M, et al. Risk stratification using the CHA2DS2-VASc score in patients with coronary heart disease undergoing percutaneous coronary intervention; sub-analysis of SHINANO registry. Int J Cardiol Heart Vasc. 2015;7:76–81. pmid:28785649
View Article
PubMed/NCBI
Google Scholar

[107] View Article

[108] PubMed/NCBI

[109] Google Scholar

[ref29] 29. Sen J, Tonkin A, Varigos J, Fonguh S, Berkowitz SD, Yusuf S, et al. CHA2DS2-VASc and CHADS2 scores for risk stratification of major adverse cardiovascular events in the COMPASS trial. European Heart Journal. 2020;41(Supplement_2).
View Article
Google Scholar

[111] View Article

[112] Google Scholar

[ref30] 30. Lee J-K, Hung C-S, Huang C-C, Chen Y-H, Chuang P-Y, Yu J-Y, et al. Use of the CHA2DS2-VASc Score for Risk Stratification of Hospital Admissions Among Patients With Cardiovascular Diseases Receiving a Fourth-Generation Synchronous Telehealth Program: Retrospective Cohort Study. J Med Internet Res. 2019;21(1):e12790. pmid:30702437
View Article
PubMed/NCBI
Google Scholar

[114] View Article

[115] PubMed/NCBI

[116] Google Scholar

[ref31] 31. Wang B-Y, Lin F-Y, Ku M-S, Wang Y-H, Lee K-Y, Ho S-W. CHA2DS2-VASc Score for Major Adverse Cardiovascular Events Stratification in Patients with Pneumonia with and without Atrial Fibrillation. J Clin Med. 2021;10(18):4093. pmid:34575202
View Article
PubMed/NCBI
Google Scholar

[118] View Article

[119] PubMed/NCBI

[120] Google Scholar

[ref32] 32. Cheng Y-T, Chang F-L, Li P-H, Lu W-C, Chiu C-S. Assessing the Suitability of CHA2DS2-VASc for Predicting Adverse Limb Events and Cardiovascular Outcomes in Peripheral Artery Disease Patients with Percutaneous Transluminal Angioplasty: A Retrospective Cohort Study. Biomedicines. 2024;12(6):1374. pmid:38927581
View Article
PubMed/NCBI
Google Scholar

[122] View Article

[123] PubMed/NCBI

[124] Google Scholar

[ref33] 33. Champsi A, Mobley AR, Subramanian A, Nirantharakumar K, Wang X, Shukla D, et al. Gender and contemporary risk of adverse events in atrial fibrillation. Eur Heart J. 2024;45(36):3707–17. pmid:39217497
View Article
PubMed/NCBI
Google Scholar

[126] View Article

[127] PubMed/NCBI

[128] Google Scholar

[ref34] 34. Yoshimura H, Providencia R, Finan C, Schmidt AF, Lip GYH. Refining the CHA2DS2VASc risk stratification scheme: shall we drop the sex category criterion?. Europace. 2024;26(11):euae280. pmid:39522169
View Article
PubMed/NCBI
Google Scholar

[130] View Article

[131] PubMed/NCBI

[132] Google Scholar

[ref35] 35. Lip GYH, Proietti M, Potpara T, Mansour M, Savelieva I, Tse HF, et al. Atrial fibrillation and stroke prevention: 25 years of research at EP Europace journal. Europace. 2023;25(9):euad226. pmid:37622590
View Article
PubMed/NCBI
Google Scholar

[134] View Article

[135] PubMed/NCBI

[136] Google Scholar

[ref36] 36. Yeh RW, Secemsky EA, Kereiakes DJ, Normand S-LT, Gershlick AH, Cohen DJ, et al. Development and Validation of a Prediction Rule for Benefit and Harm of Dual Antiplatelet Therapy Beyond 1 Year After Percutaneous Coronary Intervention. JAMA. 2016;315(16):1735–49. pmid:27022822
View Article
PubMed/NCBI
Google Scholar

[138] View Article

[139] PubMed/NCBI

[140] Google Scholar

[ref37] 37. Costa F, van Klaveren D, James S, Heg D, Räber L, Feres F, et al. Derivation and validation of the predicting bleeding complications in patients undergoing stent implantation and subsequent dual antiplatelet therapy (PRECISE-DAPT) score: a pooled analysis of individual-patient datasets from clinical trials. Lancet. 2017;389(10073):1025–34. pmid:28290994
View Article
PubMed/NCBI
Google Scholar

[142] View Article

[143] PubMed/NCBI

[144] Google Scholar

[ref38] 38. Ueki Y, Bär S, Losdat S, Otsuka T, Zanchin C, Zanchin T, et al. Validation of the Academic Research Consortium for High Bleeding Risk (ARC-HBR) criteria in patients undergoing percutaneous coronary intervention and comparison with contemporary bleeding risk scores. EuroIntervention. 2020;16(5):371–9. pmid:32065586
View Article
PubMed/NCBI
Google Scholar

[146] View Article

[147] PubMed/NCBI

[148] Google Scholar

[ref39] 39. Brabrand M, Lassen AT, Knudsen T, Hallas J. Seven-day mortality can be predicted in medical patients by blood pressure, age, respiratory rate, loss of independence, and peripheral oxygen saturation (the PARIS score): a prospective cohort study with external validation. PLoS One. 2015;10(4):e0122480. pmid:25867881
View Article
PubMed/NCBI
Google Scholar

[150] View Article

[151] PubMed/NCBI

[152] Google Scholar

[ref40] 40. Hijazi Z, Oldgren J, Andersson U, Connolly SJ, Ezekowitz MD, Hohnloser SH, et al. Cardiac biomarkers are associated with an increased risk of stroke and death in patients with atrial fibrillation: a Randomized Evaluation of Long-term Anticoagulation Therapy (RE-LY) substudy. Circulation. 2012;125(13):1605–16. pmid:22374183
View Article
PubMed/NCBI
Google Scholar

[154] View Article

[155] PubMed/NCBI

[156] Google Scholar

[ref41] 41. Providência R, Paiva L, Barra S. Risk stratification of patients with atrial fibrillation: Biomarkers and other future perspectives. World J Cardiol. 2012;4(6):195–200. pmid:22761972
View Article
PubMed/NCBI
Google Scholar

[158] View Article

[159] PubMed/NCBI

[160] Google Scholar

Figures

Abstract

Background

Methods

Results

Conclusion

Introduction

Materials and methods

Study population, design and follow-up

Definitions

Statistical analysis

Results

Deterioration of adverse events in higher CHA2DS2-VASc -risk categories

Modest prognostic performance of CHA2DS2-VASc score

Improved prediction of adverse outcomes adding hs-cTnT

Improvement of risk prediction by adding hs-cTnT in intermediate-risk AF patients

Prediction of the composite endpoint in patients with intermediate CHA2DS2-VASc score treated with oral anticoagulants

Prediction of the composite endpoint in patients with intermediate CHA2DS2-VASc score not treated with oral anticoagulants

Comparison of CHA₂DS₂-VASc to CHA₂DS₂-VA in risk assessment of patients within the intermediate risk group

Risk assessment using a novel hs-cTnCHA2DS2VASc-score

Discussion

Limitations and future directions

Conclusions

Supporting information

S1 Fig. Distribution of CHA2DS2-VASc score across the entire cohort.

S2 Fig. Kaplan-Meier analysis for all-cause mortality, stroke, major bleeding and myocardial infarction classified by CHA2DS2-VASc score risk category.

S3 Fig. Kaplan-Meier analysis for all-cause mortality, stroke, major bleeding and myocardial infarction classified by hs-cTnT category.

S4 Fig. Kaplan Meier analysis for the composite endpoint stratified by hs-cTnT categories for patients in intermediate-risk category according to CHA2DS2-VA score.

S1 Table. AUC analysis for hs-cTnT in predicting outcome variables. Abbreviations: AUC, area under the curve, CI, confidence interval.

S2 Table. Outcomes and HR classified by CHA₂DS₂-VA risk category.

S3 Table. AUCs of the CHA2DS2VA score separated for outcome variables.

References

Deterioration of adverse events in higher CHA₂DS₂-VASc -risk categories

Modest prognostic performance of CHA₂DS₂-VASc score

Prediction of the composite endpoint in patients with intermediate CHA₂DS₂-VASc score treated with oral anticoagulants

Prediction of the composite endpoint in patients with intermediate CHA₂DS₂-VASc score not treated with oral anticoagulants

Risk assessment using a novel hs-cTnCHA₂DS₂VASc-score

S1 Fig. Distribution of CHA₂DS₂-VASc score across the entire cohort.

S2 Fig. Kaplan-Meier analysis for all-cause mortality, stroke, major bleeding and myocardial infarction classified by CHA₂DS₂-VASc score risk category.

S4 Fig. Kaplan Meier analysis for the composite endpoint stratified by hs-cTnT categories for patients in intermediate-risk category according to CHA₂DS₂-VA score.

S3 Table. AUCs of the CHA₂DS₂VA score separated for outcome variables.