Introduction: The CHARGE-AF atrial fibrillation(AF) risk score shares similar characteristics to the ARIC heart failure score. Aim: To compare the CHARGE-AF score to the ARIC score in prediction of heart failure symptoms in patients with risk factors. Methods: Participants'≥65 yearswere recruited if theyhad presence of 1 or more risk factor for heart failure. Clinical HF risk was assessed using the CHARGE-AF and ARIC. Baseline electrocardiography, echocardiography and six minute walk test (SMWT) was performed. Heart failure symptoms were assessed at follow up using the Framingham criteria. ROC curves were generated and comparison made between both scores using the Hanley and McNeil method. Results: Total of 503 patients (mean±SD age 70.8±4.7 yrs, male 48% with median±IQR follow up 12 months±3). Baseline median CHARGE-AF and ARIC scores were 7.5%(3.8-11.3%) and 5.9%(2.6-9.3%) respectively. During the follow up period 55 patients developed heart failure symptoms. Patients with new onset heart failure were older with higher rates of diabetes mellitus(p=0.041), hypertension(p=0.036) and previous ischaemic heart disease(p=0.003). Echocardiography showed impaired global longitudinal(p<0.001) and circumferential strain(p=0.039) as well as higher left atrial volume( p=0.01) and left ventricular mass(p=0.002) in patients with new onset heart failure. SMWT was lower in patients with new onset heart failure 436.4 vs. 480.4m(p=0.002). ROC area under the curve was similar with both scores (ARIC 0.654(0.577-0.730,p<0.001) and CHARGE-AF 0.646 (0.565-0.727,p<0.001)), difference in AUC 0.008, SE of difference= 0.060, p=0.893. Conclusion: As well as assessing forAFrisk, theCHARGEAF score can be a useful screening tool to assess for heart failure risk in the community setting.
Ramkumar, S., Yang, H., Wang, Y., Nolan, M., Negishi, K., & Marwick, T. (2016). Comparing the Prediction of Heart Failure Symptoms Using a Clinical Score Assessing Risk of Atrial Fibrillation (CHARGE-AF) to Traditional Heart Failure Risk Scores. Heart, Lung and Circulation, 25, S105. https://doi.org/10.1016/j.hlc.2016.06.249