Introduction: Several cohort studies have reported racial differences in prevalence of atrial fibrillation (AF). This may be due to ascertainment bias. Thus, we aimed to determine incidence of AF with respect to race in patients with sinus node dysfunction and conduction disease.
Hypothesis: Ethnicity may have significant impact on development of AF in patients with sinus node dysfunction and conduction disease requiring pacemaker.
Methods: All patients followed at the University of Chicago with a dual-chamber pacemaker implanted between January 2005 to December 2011 and sinus node dysfunction (SND) or high-degree atrio-ventricular block (HGAVB) requiring dual chamber pacemaker were included. Patients were identified retrospectively. Electronic medical records, ECGs, device interrogation data, and procedure notes were examined. Logistic regression analysis was used to evaluate the association of race and device-detected AF incidence, controlling for clinical characteristics and medical comorbidities, including age, gender, HTN, diabetes, chronic obstructive pulmonary disease, obstructive sleep apnea, coronary artery disease (CAD), chronic kidney disease (CKD), and valvular heart disease.
Results: Three-hundred and twenty-four patients met criteria, including 125 (38%) Caucasians, 173 (53%) African Americans, 7 (2.2%) Hispanics, and 10 (3.1%) Asian-Americans. Average age was 68.9 +/- 14.8, 185 (57%) were female. Chi-squared analysis revealed no significant relationship between race and diabetes, CAD, or valvular disease in this population. Hypertension (p<0.01) and CKD (p=0.02) were associated with race; both were more common in African Americans and Hispanics. In a multivariate analysis, younger age (p=0.01; OR=1.02 (1.005-1.04)), African American race (p=0.03; OR=0.6 (0.3-0.9)), and Asian American race (p=0.04; OR=0.2 (0.03-0.8)) were associated with a decreased risk of device-detected AF.
Conclusions: African American race was associated with a decreased incidence of AF in patients with continuous monitoring, despite a relatively increased rate of CKD and hypertension in this population. This suggests that racial differences in AF cannot be fully explained by ascertainment bias.