Assessment of Novel Biomarkers and CHA2DS2-VASc Score in Predicting Stroke and All-cause Death in Patients With and Without Atrial Fibrillation Open Access

Chen, Yunyun (2016)

Permanent URL:


Objective: Given that patients with atrial fibrillation (AF) have an increase in stroke risk, CHA2DS2-VASc has been recommended by the American College of Cardiology/American Heart Association Task Force on Practice Guidelines to manage AF. We considered five novel biomarkers, C-reactive protein (CRP), heat shock protein (HSP), fibrin degradation product (FDP), high sensitivity troponin, and soluble urokinase-type plasminogen activator receptor (suPAR) used in cardiovascular disease outcome prediction, to evaluate whether including these biomarkers improves the prediction performance for stroke and all-cause death.

Methods: A total of 383 adult patients with AF and 2,729 without AF were included in analysis, and all five biomarkers were collected at baseline. Proportional sub-distribution hazards models and Cox proportional hazards models were used to analyze the relationship between stroke, death and CHA2DS2-VASc score and five biomarkers for patients with and without AF. C-statistics, continuous net reclassification improvement (NRI) and integrated discrimination improvement (IDI) were used to compare model prediction performance.

Results: When CHA2DS2-VASc changed from 0-1 to 6 or more, the cumulative incidence of stroke increased by 0.40 for those without AF at 2,500 days of follow-up. When predicting death, for one unit increase in CHA2DS2-VASc score, suPAR and FDP, the estimated hazard ratios were 1.28(95% CI=[1.16, 1.41]), 1.16(95% CI=[1.11, 1.21]) and 1.01(95% CI=[1.00, 1.02]). For patients without AF, with one unit increase in CHA2DS2-VASc score, suPAR and CRP, the estimated hazard ratios were 1.30(95% CI=[1.22, 1.35]), 1.20(95% CI=[1.17, 1.23]) and 1.01(95% CI=[1.00, 1.01]). The c-statistic improved significantly from 0.65(95% CI=[0.60,0.70]) to 0.71(95% CI=[0.66,0.76]) after adding suPAR and FDP for patients with AF, and improved from 0.63(95% CI=[0.58,0.69]) to 0.71(95% CI=[0.65,0.76]) after adding suPAR and CRP for patients without AF.

Conclusions: Our results suggest that CHA2DS2-VASc score, suPAR and FDP were significantly associated with all-cause death in patients with AF and CHA2DS2-VASc score, suPAR and CRP were significantly associated with all cause death in patients without AF. The models incorporating these biomarkers can refine the prediction probability. More research is needed to validate our results.

Table of Contents

1. Introduction 1

1.1 Atrial Fibrillation 1

1.2 Connection with Stroke 1

1.3 Existing Risk Scores for Predicting Stroke after Atrial Fibrillation 2

1.4 Limitations of Existing Risk Scores 3

1.5 Novel Biomarkers for Cardiovascular Disease 4

1.6 Study Goals 5

1.7 Significance 5

2. Methods 6

2.1 Study Population 6

2.2 CHA2DS2-VASc Score Calculation 6

2.3 Statistical Analysis 7

2.3.1 Descriptive Analysis 7

2.3.2 Univariate Analysis 7

2.3.3 Model Assessment 7

2.3.4 Multivariable Analysis 8

2.3.5 Model Prediction Performance Metrics 8

3. Results 9

3.1 Patient Characteristics 9

3.2 Association between CHA2DS2-VASc and Stroke 10

3.3 All-cause Death 11

3.4 Model Prediction Performance 11

4. Discussion 12

4.1 Results 12

4.2 Previous Study 12

4.3 Limitations 13

4.4 Future Research 13

4.5 Conclusions 14

References 20

About this Master's Thesis

Rights statement
  • Permission granted by the author to include this thesis or dissertation in this repository. All rights reserved by the author. Please contact the author for information regarding the reproduction and use of this thesis or dissertation.
Subfield / Discipline
  • English
Research Field
Committee Chair / Thesis Advisor
Committee Members
Partnering Agencies
Last modified

Primary PDF

Supplemental Files