An Improved Risk Score for “Low-Risk” Atrial Fibrillation Patients with 0 and 1 Risk Factors

Yoav Arnson Cardiology Department, Meir Medical Center, Kfar Sava, Israel Clalit Research Institute, Clalit Health Services, Tel Aviv, Israel Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel Moshe Hoshen Clalit Research Institute, Clalit Health Services, Tel Aviv, Israel Adi Berliner Senderey Clalit Research Institute, Clalit Health Services, Tel Aviv, Israel Orna Reges Clalit Research Institute, Clalit Health Services, Tel Aviv, Israel Ran Balicer Clalit Research Institute, Clalit Health Services, Tel Aviv, Israel Faculty of Health Sciences, Ben Gurion University of the Negev, Beer Sheva, Israel Morton Leibowitz Clalit Research Institute, Clalit Health Services, Tel Aviv, Israel Meytal Avgil Tsadok Clalit Research Institute, Clalit Health Services, Tel Aviv, Israel Moti Haim Clalit Research Institute, Clalit Health Services, Tel Aviv, Israel Cardiology Department, Soroka Medical Center, Beer Sheva, Israel Faculty of Health Sciences, Ben Gurion University of the Negev, Beer Sheva, Israel

Background: The decision on anticoagulation therapy in patients with Non-Valvular Atrial Fibrillation (NVAF) is primarily dependent thromboembolic risk. Patients with a CHA₂DS₂-VASc of 0 are considered low-risk patients who do not benefit from anticoagulation, whereas thromboembolic risk and management in patients with a score of 1 is controversial.

Methods: Using a prospective historical cohort, we identified all incident non-valvular AF cases between 2004 and 2015, with either 0 or 1 thromboembolic risk factors besides sex (CHA₂DS₂-VASc score of 0 or 1 for men and 1 or 2 for women). The new Clalit Risk Score was computed using a logistic regression model on the incidence of stroke between 30-2,500 days after NVAF diagnosis, and further converted into a point score. We further tested the score over the composite outcomes of bleeding and death.

Results: We identified 14,536 patients. Average age was 51.5 ±12.5 years. The mean follow-up time was 3.9 years. In the final model, the most significant risk factors were age above 65 and diabetes, followed by CHF, hypertension, vascular disease and chronic kidney disease stage 2 and 3. Our study cohort scores ranged between 0-3.

There was a step-wise increase in stroke incidence for each point, ranging from a low of 0.8% incidence rate for score 0 up to 3.4% for scores of 2-3. Compared to patients with a score of 0, the odds ratio for stroke among the score 3 group was 4.3 (95% confidence interval 2.9-6.6), figure 1. Our risk score exhibited an area-under-the-curve (AUC) for prediction of stroke of 0.68 (95% CI; 0.65-0.71).

Conclusion: Patients with the new score of 2-3 had an overall stroke risk over 4-fold higher than patients with a score of 0. These patients can be re-classified as intermediate risk patients, and can clearly benefit from anticoagulation treatment

Presentation

Yoav Arnson