Left Atrial Appendage Occlusion After Intracranial Haemorrhage

Atrial fibrillation affects roughly one in twenty Australians over sixty-five, and remains a leading driver of preventable cardioembolic stroke. For most patients, long-term oral anticoagulation with warfarin or a non-vitamin K antagonist offers reliable protection against thrombus formation in the left atrial appendage. The calculus shifts, however, when a patient has already survived an intracranial bleed.

Percutaneous left atrial appendage occlusion has matured from an experimental niche into a mainstream therapeutic option across tertiary Australian cardiac centres. Devices such as the Watchman FLX and the Amplatzer Amulet have earned TGA approval and are now routinely implanted at the Alfred in Melbourne, Royal Prince Alfred in Sydney, the Royal Brisbane and Women's, and the Royal Adelaide. The procedure appeals to patients and referrers because it offers mechanical stroke prophylaxis without the ongoing bleeding exposure of anticoagulant tablets.

The pool of candidates is substantial. Australians who have weathered a primary intracerebral haemorrhage or subdural haematoma frequently develop atrial fibrillation later in life, often as a consequence of hypertensive small vessel disease or amyloid angiopathy. Others present with atrial fibrillation first and then suffer a haemorrhagic transformation of an ischaemic stroke, or a spontaneous lobar bleed while on warfarin.

Each of these pathways funnels patients toward a clinician who must weigh the risks of resuming anticoagulation against the risks of leaving the appendage unprotected. The conversation about device-based alternatives becomes considerably more pressing in this setting, and Australian experience over the past decade has helped clarify when and how to proceed.

The clinical dilemma of stroke prevention after brain bleeding

The left atrial appendage is the source of more than ninety per cent of left atrial thrombi in non-valvular atrial fibrillation, which is why pharmacological stroke prevention revolves almost entirely around protecting this trabeculated pouch. For patients with a history of intracranial haemorrhage, the dilemma is that any agent powerful enough to prevent thrombus formation is also powerful enough to predispose the cerebral microvasculature to rebleeding, particularly when amyloid angiopathy or poorly controlled hypertension is in the background.

Risk stratification tools such as the CHA2DS2-VASc and HAS-BLED scores were never designed for this population. A patient with a prior lobar haemorrhage and atrial fibrillation will routinely score highly on both, leaving the treating team without a clear numerical answer. What is required is a clinical conversation that integrates the mechanism of the prior bleed, the time since the event, control of vascular risk factors, and the patient's own appetite for further bleeds versus further strokes.

This conversation is particularly relevant for the elderly, who form a disproportionate share of those affected by both conditions. In Australia, the population aged seventy-five and over is forecast to more than double by 2050, and stroke physicians and electrophysiologists are already seeing the downstream effects. A device-based alternative expands the menu considerably, and for many patients shifts the balance toward mechanical protection rather than indefinite antithrombotic medication.

Evidence base for occlusion in this high-risk cohort

The pivotal trials of left atrial appendage occlusion, including PROTECT AF and PREVAIL, excluded patients with recent intracranial haemorrhage, leaving the evidence base in this subgroup to be built largely from registries and observational cohorts. Australian and international data have consistently suggested that the periprocedural risk of occlusion in this population is comparable to that in unselected patients, provided the procedure is performed by an experienced operator and the patient is carefully prepared.

European series, including the EWOLUTION registry and the Belgian Amplatzer experience, have reported reassuring rates of ischaemic stroke reduction in patients with a bleeding history, with major bleeding events falling substantially after the early post-implant period. The Amplatzer Amulet IDE trial confirmed non-inferiority for safety and effectiveness compared with warfarin, with particularly favourable results in those whose primary reason for seeking an alternative was a previous major bleed.

Australian centres have contributed to this evidence through registries coordinated by the Australian and New Zealand Society of Cardiac and Thoracic Surgeons and through local audit programmes. Cohorts from the Alfred, Royal Melbourne and Royal Perth have shown low rates of device-related thrombus, pericardial effusion requiring intervention and procedure-related stroke. The consistency of these results has helped consolidate the procedure's position in local guidelines and supported the development of formal referral pathways from stroke units.

Device selection and procedural technique

Two devices dominate Australian practice. The Watchman FLX, approved by the TGA in 2019, is the most frequently implanted occlusion platform, favoured for its forgiving sealing criteria and broad international experience. The Amplatzer Amulet remains attractive for appendages with extreme morphology, where the dual-lobe design can offer more reliable anchoring and sealing in very wide ostia or shallow cavities.

Operator experience matters. Australian implanters typically work in teams that include a structural interventionalist or electrophysiologist, an echocardiographer for intraprocedural transoesophageal or intracardiac imaging, and an anaesthetist accustomed to managing sedation in older patients with multiple comorbidities. Most Australian centres perform between thirty and eighty implants each year, with outcomes that compare favourably with international benchmarks. The procedure takes between sixty and ninety minutes, with most patients discharged the following day.

Post-implant antithrombotic regimens remain a topic of active debate. Local protocols generally favour a short course of dual antiplatelet therapy followed by aspirin alone, particularly in patients whose primary indication for occlusion was a major bleeding event. Centres that also deliver AF ablation have noted that radiofrequency ablation of atrial tachycardia originating from the non-coronary aortic cusp shares some of the same periprocedural imaging and anticoagulation management challenges, and the cross-pollination of techniques has helped refine imaging standards across both procedural types.

Australian experience and the tyranny of distance

The Australian landscape introduces unique considerations. Many patients live substantial distances from the nearest implanting centre, with rural and remote patients often travelling from towns in the Pilbara, the Kimberley, western Queensland or far western New South Wales. Pre-procedural transoesophageal echocardiography, device planning CT and the implant itself usually require at least two and sometimes three visits to a metropolitan centre, a logistical burden that local coordinators help to navigate.

Telehealth has helped bridge part of this gap. Specialist nursing staff at the Alfred and Royal Prince Alfred coordinate video consultations for country patients, allowing initial work-up and consent discussions to occur closer to home. Several states have developed hub-and-spoke models in which regional cardiologists identify candidates, perform initial imaging, and refer patients to a metropolitan implanter, with subsequent follow-up returning to the regional cardiologist for ongoing device surveillance.

Cost remains an area of active discussion. The devices themselves are not currently subsidised through the Medicare Benefits Schedule for the specific indication of prior intracranial haemorrhage, although some private health insurers cover the procedure under selected policy tiers. Public hospital access varies by jurisdiction, with Victoria and New South Wales generally offering the broadest access. Efforts to secure a formal MSAC listing for the device in this population are ongoing, supported by local registry data and consumer advocacy groups.

Integrating occlusion with ablation practice

Operators with high-volume ablation practices have increasingly integrated left atrial appendage occlusion into a broader rhythm management strategy. Patients who undergo catheter ablation for atrial fibrillation and who have a contraindication to long-term anticoagulation because of a previous intracranial bleed are now often considered for a combined or staged procedure, with pulmonary vein isolation followed by appendage occlusion during the same admission or a few months later.

The technical lessons of high-volume ablation have informed occlusion practice in subtle but important ways. Operators familiar with contact force sensing technology have brought a more quantitative mindset to device deployment, using pressure feedback and real-time imaging to optimise appendage engagement. The discipline of routine transseptal access, careful sheath manipulation and structured periprocedural anticoagulation that defines modern AF ablation has translated directly into safer, more reproducible occlusion procedures.

For patients with extensive atrial substrate disease and recurrent atrial tachycardia after prior ablation, appendage occlusion can also serve a secondary role by reducing thromboembolic risk when further ablation attempts are planned. This is particularly relevant in Australian centres that manage complex redo cases, where the cumulative lifetime risk of stroke can be substantial.

Patient-centred decision making

The decision to proceed with left atrial appendage occlusion after an intracranial haemorrhage is rarely straightforward. Patients and families often arrive with strong views shaped by the trauma of the bleeding event, and clinicians must navigate between the very real fear of another bleed and the equally real fear of a disabling embolic stroke. Shared decision-making tools, including structured information booklets developed at the Royal Melbourne and the Baker Institute, have helped frame these conversations.

A practical pathway that many Australian centres now follow begins with referral from a stroke neurologist or geriatrician, followed by multidisciplinary review involving an electrophysiologist, a structural interventionalist and often a haematologist. The team assesses the mechanism of the prior bleed, the adequacy of blood pressure control, the presence of cerebral microbleeds on MRI, and the patient's overall trajectory. Where the balance favours occlusion, the patient is counselled about the procedure, the recovery and the realistic expectation of stroke risk reduction rather than elimination.

Follow-up is lifelong. Patients undergo routine imaging at forty-five to ninety days, then at twelve months, and at intervals thereafter, with the schedule adjusted according to device type and individual risk. Coordination between the implanting centre, the patient's local cardiologist and the general practitioner is essential, particularly for those in regional and remote communities who may otherwise struggle to access specialised follow-up.

For Australian patients, families and clinicians navigating the complex terrain of stroke prevention after a brain bleed, percutaneous left atrial appendage occlusion offers a pragmatic and increasingly well-evidenced alternative to long-term anticoagulation. Submit your latest work to the Journal of Arrhythmia and contribute to the growing body of regional evidence that informs this evolving field.