Brain and heart symbols representing the heart-brain stroke connection

Magnetofluid Left Atrial Appendage Occlusion — Fuwai Hospital and SIAT Invent Liquid-Based Alternative to Watchman

What if sealing off the most dangerous pocket in the human heart required no metal cage, no fabric mesh, no permanent implant at all — just a precisely injected magnetically controlled fluid that solidifies on command and can be dissolved if needed? A collaboration between Fuwai Hospital (China’s national cardiovascular center) and the Shenzhen Institute of Advanced Technology (SIAT) of the Chinese Academy of Sciences has done exactly that. Published in Nature in 2025, their magnetofluid left atrial appendage (LAA) occlusion system represents the first fundamentally new approach to LAA closure in over two decades — and it may render today’s umbrella-like devices obsolete.

Table of Contents

The Clinical Problem: Stroke From a Hidden Pocket

Brain and heart symbols representing the heart-brain stroke connection

Atrial fibrillation (AF) affects over 40 million people worldwide and is the most common sustained cardiac arrhythmia. Its deadliest consequence is not the irregular heartbeat itself, but the strokes it causes — and the origin of those strokes lies in a small, finger-like pouch called the left atrial appendage (LAA).

The Left Atrial Appendage: Anatomy of a Danger

The LAA is a muscular pouch extending from the left atrium. In normal sinus rhythm, the LAA contracts along with the rest of the atrium, expelling blood efficiently. But in atrial fibrillation, the atrial wall fibrillates rather than contracting, and the LAA becomes a stagnant pool where blood can pool and clot. Over 90% of thrombi in non-valvular atrial fibrillation originate in the LAA.

When a thrombus dislodges from the LAA, it can travel through the left atrium, into the left ventricle, and out into the arterial circulation — where the first major branch is the carotid artery leading to the brain. The result is an ischemic stroke, often devastating and sometimes fatal. AF-associated strokes are on average larger, more disabling, and more lethal than strokes from other causes.

The Anticoagulation Dilemma

The standard preventive strategy is long-term anticoagulation — warfarin or, increasingly, direct oral anticoagulants (DOACs) such as apixaban and rivaroxaban. While effective at reducing stroke risk, anticoagulants carry a significant bleeding risk, particularly in elderly patients and those with renal impairment, prior gastrointestinal bleeding, or a history of falls. For a substantial proportion of AF patients, the bleeding risk from anticoagulants is as dangerous as the stroke risk from AF — a clinical impasse known as the “anticoagulation contraindication” problem.

Current LAA Occlusion Devices and Their Limitations

Medical syringes and needles representing device-based interventions

LAA occlusion was developed precisely for these patients — those who cannot safely take anticoagulants. The concept is straightforward: mechanically seal off the LAA so that thrombi cannot escape, eliminating the source of stroke without the bleeding risk of systemic anticoagulation.

Watchman and Amulet: The Current Standard

The two most widely used LAA occlusion devices are the Watchman (Boston Scientific) and the Amulet (Abbott). Both are deployed percutaneously via catheter and work by positioning an expandable metal-and-fabric structure across the neck of the LAA. Over time, tissue grows over the device (endothelialization), permanently sealing the appendage.

Persistent Limitations

Despite their proven efficacy, these devices have well-documented shortcomings:

  • Size mismatch: The LAA varies enormously in shape and size between patients — from short and stubby to long and multilobed. Rigid devices come in fixed sizes, and incomplete apposition (a gap between the device and the LAA wall) is the most common complication, occurring in 10-15% of cases and potentially allowing residual flow and thrombus escape
  • Peridevice leak: Incomplete sealing around the device (peridevice leak) is documented in up to 30% of cases on follow-up imaging, though most leaks are small and clinically insignificant — some are not
  • Device-related thrombosis: The device surface itself can become a nidus for thrombus formation before endothelialization is complete, paradoxically creating the very problem the device is meant to prevent
  • Irreversibility: Once deployed, the device cannot be easily removed or repositioned. If the initial placement is suboptimal, the only option is to attempt percutaneous retrieval — a procedure with its own significant risks — or to manage the consequences
  • Long-term foreign body: The device remains in the heart permanently, with very long-term data (beyond 10-15 years) still limited

The Magnetofluid Solution: A Liquid That Seals on Command

MRI scan on medical monitor representing imaging-guided therapy

The Fuwai-SIAT team approached the problem from an entirely different direction. Instead of deploying a rigid or semi-rigid device that must conform to the LAA anatomy, they developed a magnetically controlled fluid that can be injected into the LAA, guided into position under real-time imaging, and solidified by applying an external magnetic field. The result is a custom-conforming seal that adapts to any LAA shape — and can be dissolved if needed.

The Magnetofluid Composition

The magnetofluid is a biocompatible ferrofluid — a liquid containing precisely engineered magnetic nanoparticles suspended in a carrier solution. Key properties include:

  • Biocompatible nanoparticles: Iron oxide-based nanoparticles (similar to those used in MRI contrast agents) with surface modifications to ensure biocompatibility and prevent aggregation
  • Thermosensitive carrier: A temperature-responsive polymer solution that is liquid at room/body temperature for injection and can be triggered to undergo a phase transition (solidification) by magnetic hyperthermia — when an alternating magnetic field is applied, the nanoparticles generate localized heat that causes the carrier polymer to gel
  • Biodegradability: Both the nanoparticles and the carrier polymer are designed to be biodegradable over a defined time period, or the solidified plug can be dissolved on demand by applying a different magnetic field pattern or a dissolving agent

Why a Liquid Solves the Shape Problem

The LAA is not a simple tube — it is a complex, often multilobed structure with an irregular orifice. No fixed-size device can perfectly conform to every variation. A liquid, however, flows into every crevice, fills every lobe, and creates a perfect cast of the LAA interior before solidifying. The result is a seal with zero peridevice leak by definition — the occlusion is not a device sitting in an opening, but a solid plug that is the exact negative of the appendage’s internal geometry.

How the System Works: Step by Step

Brain MRI scan representing stroke prevention imaging

The magnetofluid LAA occlusion procedure follows a carefully choreographed sequence:

Step 1: Catheter Navigation and LAA Access

Under transesophageal echocardiography (TEE) and fluoroscopic guidance, a catheter is advanced through the femoral vein, across the interatrial septum, and into the left atrium — the standard approach used for existing LAA occlusion procedures. The catheter tip is positioned at the orifice of the LAA.

Step 2: Magnetofluid Injection

The magnetofluid is slowly injected through the catheter into the LAA. Because the fluid is liquid at body temperature, it naturally flows into every recess and lobe of the appendage. Real-time imaging confirms complete filling of the LAA cavity.

Step 3: Magnetic Field-Guided Solidification

Once the LAA is completely filled, an external alternating magnetic field is applied. The magnetic nanoparticles in the fluid respond by generating localized heat (magnetic hyperthermia), which triggers the thermosensitive carrier polymer to undergo a sol-gel transition — the liquid solidifies into a firm, custom-conforming plug. The entire solidification process takes seconds to minutes and is monitored in real time by imaging.

Step 4: Verification and Catheter Withdrawal

Imaging confirms that the LAA is completely sealed with no residual flow. The catheter is withdrawn, and the solidified plug remains in place, anchored by its perfect conformity to the LAA geometry. Over time, the LAA wall remodels around the plug, and the polymer gradually biodegrades, leaving the LAA permanently closed by natural tissue.

Step 5: Reversibility (If Needed)

Unlike a permanently implanted metal device, the magnetofluid plug can be dissolved on demand. If the initial occlusion is incomplete, if the clinical situation changes, or if removal is desired for any reason, a specific dissolving protocol (a different magnetic field pattern combined with a biocompatible dissolving agent delivered by catheter) can liquefy the plug, allowing it to be aspirated and the procedure reattempted or abandoned.

The Nature Study: Key Results

ECG and heart model representing cardiac intervention research

The study, published in Nature (DOI: 10.1038/s41586-025-10091-1), presented results from extensive preclinical testing in animal models, demonstrating both the feasibility and safety of the magnetofluid LAA occlusion approach.

Efficacy Results

  • Complete LAA occlusion: In all treated animals, the magnetofluid achieved complete LAA occlusion as confirmed by angiography and echocardiography — no residual flow, no peridevice leak
  • Custom conformity: The solidified plug conformed perfectly to each animal’s unique LAA anatomy, regardless of shape or lobe configuration — a result impossible to guarantee with fixed-size devices
  • Stable occlusion at follow-up: The occlusion remained stable throughout the follow-up period, with no migration, displacement, or device embolization — complications that can occur with conventional devices

Safety Results

  • No device-related thrombosis: Unlike metal-and-fabric devices, the biocompatible magnetofluid surface did not promote thrombus formation during the pre-endothelialization period
  • Biocompatibility: Histological analysis showed no significant inflammatory response to the magnetofluid or its degradation products
  • Successful dissolution: The on-demand dissolution protocol was successfully demonstrated, confirming the reversibility of the occlusion
  • Magnetic field safety: The external magnetic field strengths used for solidification were within safety limits established for clinical magnetic resonance imaging

Advantages Over Existing Devices

Operating room representing cardiac procedural innovation

The magnetofluid approach addresses nearly every limitation of current LAA occlusion technology:

1. Universal Size: Eliminating Size Mismatch

Because the fluid conforms to any LAA shape, there is no need to select a device size, no risk of choosing too small or too large a device, and no possibility of incomplete apposition. A single magnetofluid formulation works for every patient anatomy.

2. Zero Peridevice Leak by Design

Peridevice leak occurs when a rigid device cannot perfectly appose the irregular LAA orifice. A liquid that fills the entire cavity and solidifies creates a seal with zero gaps by definition — the occlusion is the geometry.

3. Reversibility: A Safety Net That Current Devices Lack

The ability to dissolve and remove the occlusion on demand is perhaps the most transformative advantage. Current LAA devices are permanent implants — once deployed, they cannot be removed without major surgery. The magnetofluid provides a safety net: if the result is suboptimal, the procedure can be reversed and reattempted. This reversibility dramatically reduces the risk profile of LAA occlusion as a whole.

4. No Foreign Body Remains Long-Term

As the polymer biodegrades and is replaced by natural tissue, the long-term presence of a foreign body in the heart is eliminated. This addresses the unknown long-term risks of permanent metal implants and the rare but documented complications of device erosion and embolization that can occur years after implantation.

Fuwai Hospital and SIAT: The Innovation Ecosystem

This breakthrough was made possible by the collaboration between two of China’s most capable research institutions, each contributing distinct expertise.

Fuwai Hospital: China’s National Cardiovascular Center

Fuwai Hospital (阜外心血管病医院), affiliated with the Chinese Academy of Medical Sciences and Peking Union Medical College, is China’s — and the world’s — largest specialized cardiovascular hospital. With over 1,500 beds dedicated to cardiovascular disease, Fuwai performs more cardiac catheterizations, structural heart interventions, and cardiac surgeries than any other institution globally. Its experience in LAA occlusion is unmatched, having been among the first centers in China to adopt the Watchman device and having contributed extensively to the clinical evidence base for LAA closure.

SIAT: Shenzhen Institute of Advanced Technology

The Shenzhen Institute of Advanced Technology (SIAT), part of the Chinese Academy of Sciences, is a leading research institution in biomedical engineering, nanotechnology, and medical devices. SIAT’s expertise in magnetic nanoparticle engineering, thermosensitive biomaterials, and medical imaging was essential for developing the magnetofluid technology and the external magnetic control system.

What International Patients Should Know

The magnetofluid LAA occlusion system is currently in the preclinical and early translational development stage. While the Nature publication demonstrates compelling proof of concept, clinical trials in human patients have not yet begun.

Current Standard of Care at Fuwai

Fuwai Hospital currently offers the full range of LAA occlusion options, including Watchman, Amulet, and other approved devices. The hospital’s structural heart team has world-class expertise in LAA closure, with one of the largest case volumes globally and published outcomes that match or exceed international benchmarks. International patients with atrial fibrillation and anticoagulation contraindications can be evaluated for LAA occlusion at Fuwai using currently available devices.

Timeline for Magnetofluid Availability

  • Preclinical validation: Completed (published Nature 2025)
  • Device development and GMP manufacturing: In progress
  • Clinical trials: Expected to begin within 2-3 years, pending NMPA (National Medical Products Administration) approval
  • Commercial availability: If clinical trials are successful, the device could be available in China within 5-7 years

Accessing Fuwai Hospital

Fuwai Hospital is located in Beijing, which is covered under China’s 240-hour transit visa exemption and 30-day unilateral visa-free entry policies. The hospital has an international medical services department that coordinates care for patients from abroad, including consultation, evaluation, and procedure scheduling. For patients with complex atrial fibrillation or those who have experienced complications from existing LAA devices, Fuwai’s expertise in both conventional and innovative approaches to LAA closure makes it a leading destination for cardiovascular care in Asia.

Sources

  • Li S, Pan L, et al. “Magnetofluid-based left atrial appendage occlusion.” Nature, 2025. DOI: 10.1038/s41586-025-10091-1
  • Fuwai Hospital, Chinese Academy of Medical Sciences: Official Website
  • Shenzhen Institute of Advanced Technology (SIAT), Chinese Academy of Sciences: SIAT Official Website
  • Reddy VY, Sievert H, et al. “Percutaneous left atrial appendage closure for stroke prophylaxis in patients with atrial fibrillation: 2.3-Year Follow-up of the PROTECT AF (Watchman Left Atrial Appendage System for Embolic Protection in Patients with Atrial Fibrillation) Trial.” Circulation, 2013;127(6):720-729. DOI: 10.1161/CIRCULATIONAHA.112.114388
  • Landmesser U, Schmidt B, et al. “Left Atrial Appendage Occlusion: Current Status and Perspectives.” European Heart Journal, 2023;44(37):3597-3609. DOI: 10.1093/eurheartj/ehad439
  • Saw J, et al. “Peridevice Leak After Left Atrial Appendage Occlusion: Clinical Implications and Management Strategies.” JACC: Cardiovascular Interventions, 2021;14(10):1061-1073. DOI: 10.1016/j.jcin.2021.03.015
← Back to News