Identification, multi-modal characterisation and clinical outcomes of pre-atrial fibrillation sub-classes

Background A Pre-atrial fibrillation (Pre-AF) status has been postulated to exist when there are structural or electrical findings predisposing to AF, but it is not well described in relation to prognosis. Objectives This prospective cohort study aims to sub-classify Pre-AF by structural, electrical and biomarker parameters and quantify associations with major adverse events. Methods Prospective cohort study of individuals with high risk of developing AF (according to the FIND-AF risk score), who underwent echocardiographic, ECG and biomarker assessment. Pre-AF sub-classes were specified as follows: Pre-AF 1 (one of echocardiographic, ECG or biomarker parameters); Pre-AF 2 (two parameters); and Pre-AF 3 (three parameters). Individuals were followed up for a composite primary outcome of major adverse cardiovascular, cerebrovascular and cognitive events (incident AF, heart failure, ischaemic stroke/TIA, myocardial infarction, dementia or cardiovascular death). Results Of 254 participants (mean age 74.9 (+/- 5.8) years, 50% women), 82 (32.3%) were classified as Pre-AF 1, 90 (35.4%) as Pre-AF 2, and 82 (32.3%) as Pre-AF 3. Higher Pre-AF sub-class was associated with higher NT-pro BNP (84.0 ng/L vs 186.1 ng/L vs 558.2 ng/L, p<0.001) and larger left atrial volume (29.0 ml/m2 vs 39.8 ml/m2 vs 46.1 ml/m2, p < 0.001). Over a median follow-up of 24.4 months (IQR 19.6-26.3 months), the primary outcome was 23.2% (19/82) for Pre-AF 3, 11.1% (10/90) for Pre-AF 2 and 4.9% (4/82) for Pre-AF 1 (log-rank p = 0.002). Conclusions Sub-classification of Pre-AF using ECG, echocardiography and NT-proBNP has prognostic implications but is not specifically associated with diagnosis of incident AF in routine care.