Short summary

The study aims to enhance stroke risk stratification in atrial fibrillation (AF) patients by assessing how the individual components of the CHA2DS2-VASc score and ABC-stroke score affect the hemostatic profile, and how the hemostatic profile differs among stroke risk groups, through analysis of hemostatic biomarkers in a cohort of OAC-naive AF patients. We will also explore the association of left atrial size/function and AF-burden with hemostatic biomarkers.

Rationale

Background: Atrial fibrillation (AF) is the most prevalent cardiac arrhythmia worldwide, and it is associated with a five-fold increased risk of ischemic stroke compared to a healthy population. Patients with stroke due to AF have a higher mortality and greater disability compared to stroke patients without AF. Oral anticoagulants (OACs), either vitamin K antagonists (VKAs) or direct oral anticoagulants (DOACs), are the cornerstones for the prevention of ischemic strokes in patients with AF. Different stroke risk stratification schemes have been developed to guide OAC therapy decision-making in AF patients. The most commonly used scheme is the CHA2DS2-VASc score, including the clinical risk factors congestive heart failure, hypertension, age, diabetes, prior stroke, vascular disease, and female gender. One point is assigned for every risk factor encompassed by the CHA2DS2-VASc score, except age ≥75 years and prior stroke, which account for two points each. The score ranges from 0-9 points, with higher scores representing a higher risk of stroke. The annual risk of ischemic stroke for CHA2DS2-VASc score of 0, 1, and 2 is approximately 0-0.68 %, 1.3-1.61 %, and 2.2-2.49 %, respectively. Novel stroke risk stratification schemes, also incorporating biomarkers, have been developed, e.g., the ABC-stroke score (Age, Biomarkers, and Clinical history). The components of the ABC-stroke score include age, plasma levels of N-terminal pro-B-type Natriuretic Peptide (NT-proBNP) and high-sensitivity troponin T (hs-TNT), and prior stroke/transient ischemic attack (TIA). ABC-stroke risk categories are defined as low-risk (<1 % predicted one-year risk of stroke), medium-risk (1-2 % predicted one-year risk of stroke), and high-risk (>2 % predicted one-year risk of stroke). Studies have shown that the ABC-stroke score outperforms the CHA2DS2-VASc score in terms of stroke risk prediction and stratification of AF patients. However, the CHA2DS2-VASc score is still recommended as the primary stroke risk stratification scheme for AF patients in current guidelines. Despite the broad use of stroke risk stratification schemes in clinical practice, stroke risk assessment remains particularly challenging in AF patients with intermediate stroke risk, i.e., men with a CHA2DS2-VASc score of 1 and women with a CHA2DS2-VASc score of 2, where current guidelines recommend considering OAC therapy. OAC therapy is generally recommended for AF patients at high risk of stroke, i.e., men with CHA2DS2-VASc score of ≥ 2 and women with CHA2DS2-VASc score of ≥ 3. OAC therapy is generally not recommended for AF patients at low risk of stroke, i.e., men with a CHA2DS2-VASc score of 0 and women with a CHA2DS2-VASc score of 1. The CHA2DS2-VASc score is broadly used in clinical practice due to its simplicity and low cost, however, the score has limitations. It exclusively includes clinical stroke-related risk factors, but not AF burden and left atrial size/function, which are proven independent markers of a hypercoagulable state. Treatment and stroke risk assessment in AF patients is moving towards a more individualized patient care, in which biomarkers play an essential role in improving risk stratification and personalizing treatment, as proposed in the ABC-stroke score. In recent years, measurements of hemostatic biomarkers in AF patients have been investigated in several studies, illustrating a hypercoagulable state. Of note, the CHA2DS2-VASc score also predicts stroke risk in patients without AF. Hence, there are concerns about the specificity for predicting AF-related stroke versus non-AF related stroke. Furthermore, the CHA2DS2-VASc score does not consider whether the modifiable components are well-controlled or not, which could lead to OAC overtreatment. For these reasons, there is still a need for refining stroke risk prediction using more AF-sensitive factors (i.e., AF-burden and left atrial size/function), in addition to the clinical factors, and for more exhaustive assessment of the modifiable components, to optimize stroke risk stratification in the future. It remains to be established whether the abovementioned AF-sensitive factors can be used to improve OAC decision-making, particularly in NVAF patients with intermediate stroke risk, where the balance between benefit and harm is less clear. When doubt persists, the presence of one or more non-CHA2DS2-VASc stroke risk factors could possibly strengthen the decision to initiate OAC therapy. The annual risk of ischemic stroke in NVAF patients depends on the overall CHA2DS2-VASc score. However, the annual risk of ischemic stroke in patients with NVAF may differ depending on the individual components of the CHA2DS2-VASc score that determine the score, despite the overall score being the same, as well as whether the modifiable components are well-controlled or not. Glowicki et al. conducted a study comparing the hemostatic profiles of AF patients with low stroke risk to those with intermediate stroke risk, as determined by the CHA2DS2-VASc score. As opposed to the study by Glowicki et al., our study will compare all three stroke risk groups, i.e., low, intermediate, and high risk, as determined by the CHA2DS2-VASc score and ABC-stroke score, and will incorporate a broader panel of hemostatic biomarkers, including the contact activation system. Furthermore, our study will investigate AF-burden and left atrial size/function, and their association with the hemostatic profile. Patients included in our study will be genuinely OAC-naïve, as opposed to the study by Glowicki et al. Comparison of hemostatic profiles between different stroke risk groups could be essential for the future improvement of OAC therapy decision-making, especially in NVAF patients with intermediate stroke risk. Objectives: 1) To evaluate how the individual components of the CHA2DS2-VASc score and ABC-stroke score are associated with hemostatic biomarkers, including an assessment of whether the modifiable components are well-controlled or not, and how the hemostatic profile differs among stroke risk groups in OAC-naïve NVAF patients. 2) To compare the hemostatic profiles of the different CHA2DS2-VASc scores and ABC-stroke scores in OAC-naïve NVAF patients. 3) To evaluate how AF-burden is associated with hemostatic biomarkers in OAC-naïve NVAF patients. 4) To evaluate how left atrial size/function is associated with hemostatic biomarkers in OAC-naïve NVAF patients.

Description of the cohort

Study design: The study is a cross-sectional, single-center observational study and will take place at Esbjerg Hospital - University Hospital of Southern Denmark, involving collaboration between the Unit for Thrombosis Research, Department of Clinical Diagnostics and the Department of Cardiology. Study population: The study population will consist of patients with newly diagnosed NVAF. Patients will have to be naïve to oral and parenteral anticoagulants prior to inclusion. We will check for this upon inclusion by systematic screening of an online medication database, called FMK, which is widely used among health care personnel in Denmark. According to the annual report (2023) from the AF database in Denmark (AFDK), there are about 800 patients with newly diagnosed AF in the catchment area of the University Hospital of Southern Denmark, Esbjerg, why we are confident that it will be possible to include the necessary number of patients. Collaboration with the general practitioners will be essential for patient inclusion. We will provide comprehensive information to general practitioners about our study and the importance of withholding OAC treatment before subacute referral to the Department of Cardiology. Patients with newly diagnosed NVAF who are willing to participate in our study and sign the patient consent form will be scheduled for fast track outpatient clinic visit within four days of their consultation with the general practitioner for blood sampling, transthoracic echocardiography (TTE), and heart rhythm monitoring. OAC treatment will be initiated immediately after blood sampling, based on current guidelines for the management of AF, and we will also do a tailored comprehensive work-up of the AF patients as necessary. Demographic data will be collected, as well. Likewise, symptoms attributable to AF will be quantified according to the modified EHRA-score (European Heart Rhythm Association). To exclude other causes of coagulopathy and characterize the patients at baseline, we will measure the following variables in each participant: Full blood count, Activated Partial Thromboplastin Time (APTT), International Normalized Ratio (INR), renal function, Glycated Hemoglobin (HbA1c), C-reactive Protein (CRP), liver function, and lipid profiles. Furthermore, plasma levels of NT-proBNP and hs-TNT will be determined with high-sensitivity immunoassays to estimate the ABC-stroke score. The risk of thromboembolic events will be estimated using the CHA2DS2-VASc score and ABC-stroke score for all patients in our study population, and the patients will be grouped according to their overall individual score. Moreover, each group will be subdivided according to the individual components of the CHA2DS2-VASc score and ABC-stroke score that constitute the overall score. We will determine whether hypertension and diabetes are effectively managed by conducting home blood pressure (BP) measurements and assessing HbA1c levels. BP will be measured at home three times in the morning and evening over three consecutive days, and mean BP will be calculated based on the measurements from days two and three. For simplicity, well-controlled hypertension will be defined as a BP ≤ 135/85 mmHg, and a systolic BP ≤ 145 mmHg for age groups < 80 years and ≥ 80 years, respectively. The treatment goals for type 1 and 2 diabetes patients are HbA1c ≤ 53 and ≤ 48 mmol/mol, respectively, and will be used as thresholds for whether diabetes is well controlled or not. Inclusion criteria: • Patients with newly diagnosed NVAF who are OAC-naïve prior to inclusion. • Age ≥ 18 years. • Signed informed consent. Exclusion criteria: • Ongoing OAC treatment prior to inclusion. • Valvular AF (mechanical heart valves or moderate-severe mitral stenosis). • Secondary AF due to an acute reversible precipitant (e.g., infection, surgery, thyrotoxicosis, etc.). • Pregnant or breast-feeding women. • Treatment with oral contraceptives. • End-stage renal disease (creatinine clearance <15 mL/min as calculated by the Cockcroft- Gault equation). • Connective tissue diseases. • Active cancer (cancer diagnosis not followed by curative procedures six months from the date of diagnosis). • Recent major surgery (< three months). • Acute coronary syndrome, stroke/TIA, and venous thromboembolism < three months prior to inclusion. • Thrombophilia. • Significant liver or hematological conditions.

Data and biological material

Biological material: A key component of our study is the measurement of hemostatic biomarkers in a population of OAC-naïve NVAF patients, with the aim of contributing to the improvement of stroke risk stratification in this population. Hence, blood will be drawn to measure hemostatic biomarkers. The only biological material used for the study is blood. The blood samples will be stored in the research biobank at University Hospital of Southern Denmark, Esbjerg. Blood sampling: Fasting blood samples will be drawn from an antecubital vein with minimal stasis, using a 21 gauge needle. The first 2 mL of blood will be discarded, and the following 6 x 2.7 mL blood will be collected into 0.109 M sodium citrate tubes for APTT, INR, and hemostasis variables. Then, 2 x 3 mL of blood will be collected in Li-Heparin tubes for lipids, liver enzymes, renal function, CRP, NT-proBNP, and hs-TNT. Finally, 3 mL of blood will be collected in EDTA-tubes for HbA1c and full blood count. To generate platelet poor plasma (PPP), the tubes will be centrifuged at 2000 g for 20 min. at room temperature. PPP will be stored at -80°C until analysis. Hemostatic biomarkers: We will measure biomarkers related to the primary and secondary hemostasis, as listed below: Primary hemostasis (platelet plug formation): To assess the influence of the primary hemostasis, we will measure the levels of von Willebrand factor (vWF) antigen using an in-house enzyme-linked immunosorbent assay (ELISA). If the concentrations of vWF are elevated compared with reference values, p-selectin and ADAMTS13 will be analyzed. Secondary hemostasis (fibrin formation and resolution): Assessment of the secondary hemostasis will be done by analyzing thrombin turnover, and investigate whether thrombin generation is initiated though the contact activation system (CAS), i.e., the intrinsic pathway of the coagulation cascade, or the tissue factor (TF) pathway, i.e., the extrinsic pathway. In relation to CAS, measures of kallikrein generation (lag time, peak kallikrein concentration, time to peak, and endogenous kallikrein potential (EKP)) and concentrations of cleaved high-molecular weight kininogen (cHK), factor XII, prekallikrein, and HK will be measured. Concentrations of C1-inhibitor will be measured with commercial C1-inhibitor antibodies. Thrombin generation will be assessed through measures of lag time, peak thrombin concentration, time to peak, and endogenous thrombin potential (ETP). Endogenous activation of prothrombin to thrombin will be estimated from concentrations of prothrombin fragment 1 + 2 (F1+2), by a commercial ELISA. Furthermore, factor VII, factor X, factor II, protein C, protein S, tissue factor pathway inhibitor (TFPI), and antithrombin (AT) will be measured using commercial ELISAs to assess concentrations of activators and inhibitors of blood coagulation. Fibrin turnover will be assessed through fibrin clot lysis and concentrations of fibrinogen, which will be measured on a BN II analyzer. To assess fibrinolysis, D-dimer, a fibrin degradation product, will be measured by an immunoturbidimetric method. Other markers of fibrinolysis will be measured as well, i.e., tissue-type plasminogen activator (t-PA), plasminogen activator inhibitor 1 (PAI-1), plasminogen, plasmin inhibitor (PI), factor XIII, and thrombin activatable fibrinolysis inhibitor (TAFI), using in house and commercial assays. AF burden and left atrial function: Patients will undergo TTE as well as seven days heart rhythm monitoring. Through TTE, a rhythm independent expression of left atrial function, known as left atrial function index (LAFI), will be calculated. LAFI is a ratio that incorporates analogues of cardiac output, atrial reservoir function, and left atrial (LA) size. Left ventricular ejection fraction (LVEF) will also be assessed as part of the TTE. AF-burden will be determined as a percentage based on the total number of AF-events during the seven days heart rhythm monitoring. Primary and secondary endpoints: Primary endpoints will be related to thrombin generation, kallikrein generation, and F1 + 2. Secondary endpoints will be the remaining hemostatic biomarkers, AF-burden, and LAFI. Data: When assessing the eligibility of the patients, we will need personal information, i.e., name, age, gender, address, and CPR-number. We will do a thorough assessment of their medical history, including medication intake through Sundhedsjournalen, Cosmic Arkiv, and FMK. This is to assess whether or not patients meet the eligibility criteria, and if they are OAC-naïve. Furthermore, we will assess the 12-lead electrocardiogram (ECG) and the blood work to document AF and to exclude end-stage renal disease, as well as secondary AF. Therefore, assessment of the patient journals is crucial to our study.

Collaborating researchers and departments

The Unit for Thrombosis Research, Department of Clinical Diagnostics, Esbjerg and Grindsted Hospital, University Hospital of Southern Denmark.

• CMO and associate professor, Anna-Marie Bloch Münster (co-supervisor).
• Professor, Else-Marie Bladbjerg (co-supervisor).

Department of Cardiology, Esbjerg and Grindsted Hospital, University Hospital of Southern Denmark.

• Professor and MD, Axel Brandes (main supervisor).

Faculty of Health Sciences, McMaster University.

• Professor and MD, Jeffrey Sean Healey (co-supervisor).