Abnormal left atrial strain is associated with eventual diagnosis of atrial fibrillation in patients with embolic stroke of undetermined source

Type of funding sources: None.

Patients with embolic stroke of undetermined source (ESUS) have no immediate attributable cause despite initial evaluation. Occult paroxysmal atrial fibrillation (pAF) diagnosed on prolonged electrocardiographic monitoring may be subsequently found in 20% of patients initially classified as ESUS [1]. Anatomic and functional alterations of left atrium (LA) are known to predict pAF in the general population [2]. Recent studies have suggested that LA dysfunction is linked to the risk of development of pAF in stroke patients and may precede LA enlargement [1]. LA strain analysis may be helpful to predict the development of pAF as it can detect subclinical functional impairment [2]. The aim of this study was to analyse whether LA strain can be a marker for the development of new-onset AF in patients with ESUS and sinus rhythm.

Our single-centre observational cohort study examined 157 patients hospitalised for ESUS in our tertiary hospital between October 2014 and October 2017 who underwent AF monitoring with an implantable loop recorder (ILR). These patients were followed up for occurrence of new-onset pAF and recurrence of ischaemic stroke. All patients underwent transthoracic echocardiography (TTE) during index hospitalisation as part of the workup for ESUS. Echocardiographic images were obtained and analysed post-hoc for two-dimensional (2D) speckle tracing deformation parameters. These LA strain parameters were analysed against the primary outcome of AF detection and secondary outcome of recurrent ischaemic stroke.

A total of 157 ESUS patients, with a mean age of 61.0 (±11.6) years, were followed up for a median duration of 3.5 (interquartile range 3.29) years. ILR monitoring detected AF in 27 patients (17.2%). 27 patients developed recurrent ischaemic strokes. Of the 27 patients with newly diagnosed AF on ILR, 24 (88.9%) were commenced on oral anticoagulation; the remaining 3 patients were not on anticoagulation due to high bleeding risk and frailty.

Patients who had newly diagnosed AF had more impaired left atrial reservoir strain (LASr; 23.5% ± 10.3%, P = 0.042) and left atrial conduit strain (LAScd; -10.5% ± 5.3%, P = 0.003) compared with patients who remained in sinus rhythm. Multivariable logistic regression analysis adjusting for age, sex, hypertension, hyperlipidaemia, diabetes mellitus and left ventricular ejection fraction (LVEF) showed that LAScd was associated with occult AF (adjusted odds ratio [aOR] = 1.110, 95% CI, 1.019-1.209, P = 0.017) and the composite outcome of AF and recurrent ischaemic stroke (OR = 1.069, 95% CI, 1.014-1.138, P = 0.038). Abnormal LAScd was significantly associated with occult AF even when stratified by normal LA volume index (LAVI) (OR = 2.672, 95% CI, 1.035-8.548, P = 0.048) and high LAVI (OR = 1.713, 95% CI, 1.023-2.869, P = 0.041).

Impaired left atrial strain was associated with occult AF detection in patients with ESUS undergoing ILR monitoring.