AF is associated with self-reported syncope and falls in a general population cohort.
Item Type:Journal Article
Citation:Jansen S, Frewen J, Finucane C, de Rooij SE, van der Velde N, Kenny RA., AF is associated with self-reported syncope and falls in a general population cohort., Age Ageing, 44, 4, 2015, 598-603
598.full.pdf (PDF) 6.811Kb
Background: syncope is an important, but underestimated clinical problem in older persons. It is often overlooked in clinical practice or mistaken for falls. Atrial fibrillation (AF) is the most common cardiac arrhythmia, but little evidence exists regarding the association between AF, falls and syncope in the general population. Methods: cross-sectional analyses within a population sample of people aged 50+, taken from The Irish Longitudinal Study on Ageing. Ten-minute electrocardiogram recordings (n = 4,885) were analysed to detect AF. Syncope (self-reported faints or blackouts) and falls in the past year, co-morbidities, health measures and medications were gathered through computer-aided personal interviews. Multivariable logistic regression was performed to study associations between AF, falls and syncope. Results: mean age was 62 years (range: 50–91), 54% were female. Prevalence of AF was 3%, increasing to 8% in participants aged 75+. Of participants, 5% (n = 223) reported syncope and 20% (n = 972) reported falls. After adjustment for confounders, AF was significantly associated with faints and blackouts (odds ratio (OR) 2.0 [95% confidence interval (CI) 1.0–3.9]). After stratification by age category, we found that this association was strongest and only significant in participants aged 50–64 years (OR 4.4 [1.5–12.6]). Stratified for age group, AF was significantly associated with falls in participants aged 65–74 years (OR 2.0 [1.0–4.1]). Conclusions: adults aged 50+ with self-reported syncope and adults aged 65–74 years with falls are twice as likely to have AF at physical examination. These associations are independent of stroke, cardiovascular and psychotropic drugs and other confounders. Further longitudinal studies are needed to explore this association and potential causality further.
Type of material:Journal Article
Series/Report no:Age Ageing
Availability:Full text available