EMPA-ESUS. a Randomised Control Trial to Investigate the Impact of Empagliflozin on Left Atrial Function in Patients with Embolic Stroke of Undetermined Source

Background:

Individuals who suffer an embolic stoke of undermined source represent 17% of all ischaemic stroke patients. When they undergo continuous cardiac rhythm monitoring with a loop recorder, a large portion of the population have been noted to develop atrial fibrillation (AF). In our own centre, 48.6% of patients have found to develop AF. This is considerably more than the proportion seen in patients undergoing ILR implantation for other reasons.

What remains unclear is whether this AF may have been responsible for the initial stroke, or is a reflection of an underlying atrial cardiopathy. Moreover, it is felt that this device detected AF is an important risk factor for stroke recurrence. Recent trials have suggested that whilst anticoagulation for device detected AF reduces future stroke, this is at the expense of a significant bleeding risk. Thus far, there have been no therapies that prevent the occurrence of AF, even in groups at high risk of AF development. Obviously such a strategy would negate the risk of bleeding associated with anticoagulation use. Such a strategy would reflect a paradigm shift in AF management.

Whilst studies have tried to explore the significance of abnormal atrial function in ESUS, generally crude assessments such as P-wave terminal force in V1 and brain natriuretic peptide had been utilised. Our group deep phenotyped over 300 ESUS patients. This work highlighted an important predictive role of left atrial strain for future device detected AF.

Left atrial strain has shown great promise as a marker of atrial dysfunction. There is growing evidence that sodium glucose 2 co-transporter inhibitors, such as empagliflozin, may prevent the development of AF. There have been over a dozen metaanalyses of the safety data from the large SGLT2i outcome trials, which have consistently shown that the SGLT2i arm was associated with lower incident AF compared to the control arm.

Similar results were seen in cohort studies including in Taiwan and the Food and Drug Association's adverse events reporting system. It is this finding that prompted the interest in utilising empagliflozin as the intervention in the trial.

Study aims and objectives:

The aims of this study are to assess if:

SGLT2 inhibitors can prevent adverse changes in left atrial function in patients who are felt to have an increased chance of AF development.

A nested MRI substudy will be used to try and provide mechanistic insight into how SGLT2 inhibitors may exert an antiarrhythmic effect.

Methods:

Given that the use of SGLT2 inhibitors to prevent atrial change has not been investigated before, we aim to investigate the hypotheses with an early stage randomised controlled trial.

For this, patients undergoing implantable loop recorder insertion for AF detection following an embolic stroke of undetermined source will be invited to participate. The participants will be randomised either to usual stroke care or to the intervention arm (use of SGLT2 inhibitor in addition to usual stroke care). They will be assessed at baseline and at 6-months following initiation of the medication. Both appointments will include echocardiography, electrocardiography and anthropometric assessments such as weight and grip strength.

From each arm, 10 patients will under also undergo advanced MRI imaging including MR spectroscopy, to try and investigate possible mechanisms for any effect of the SGLT2 inhibitor.

Inclusion criteria

• Age over 18 years
• Patient has had an embolic stroke of undetermined source, defined as:

• Ischaemic stroke or transient ischaemic attack diagnosed clinically + supporting radiological features on CT or MRI imaging
• No clinically significant extracranial or intracranial atherosclerosis found on neck imaging
• No evidence of prevalent cardioembolic source of embolism - atrial dysrhythmia, cardiac thrombus, prosthetic valve, cardiac tumours, mitral stenosis, recent MI, LVEF<35%, valvular vegetations, or infective endocarditis

AND

• No other specific stroke aetiology identified (pro-thrombotic states, arteritis, dissection, drug abuse)
• Patient has been referred for an implantable loop recorder implant to investigate possible atrial dysrhythmia
• Ability to give written, informed consent

Exclusion criteria

• Prevalent atrial fibrillation or atrial flutter
• Subsequent demonstration of a cardio-emoblic source
• Pre-existent indication for SGLT2i including heart failure with a reduced ejection fraction and Type 1 or Type 2 diabetes mellitus
• Contraindication to the commencement of SGLT2i including a history of, or risk factors for diabetic ketoacidosis, such as restricted food intake or a history of alcohol abuse
• Hypersensitivity to the active substance or to any of the excipients of empagliflozin
• Presence of alternate implantable cardiac devices such as a pacemaker or defibrillator, that would make ILR utilisation redundant
• Any condition that may confound the results or interfere with participation such as dementia
• An unwillingness to participate