Magnetic fluid seals the heart's most dangerous pocket without a single metal implant

Nature, March 4, 2026

For patients with atrial fibrillation, the left atrial appendage is a liability. This small, finger-like pouch in the heart is where roughly 90% of stroke-causing blood clots form. Sealing it off has become a standard strategy, but the metal devices used to do so come with their own problems: they do not always fit the appendage's irregular shape, they can cause blood clots to form on the device itself, and gaps between the device and the tissue can allow leaks.

A team led by Prof. Xu Tiantian at the Shenzhen Institutes of Advanced Technology (Chinese Academy of Sciences) and Prof. Pan Xiangbin at Fuwai Hospital has proposed an entirely different approach. Instead of wedging a metal plug into the appendage, they fill it with a liquid that responds to magnets, conforms perfectly to the cavity's shape, and then solidifies. The results, published March 4 in Nature, demonstrate thrombus-free occlusion in animal models.

What the magnetofluid actually is

The material combines two components. Neodymium-iron-boron particles provide magnetic responsiveness, allowing external magnets to hold the fluid in place against the force of blood flow and heartbeat. The carrier fluid is a solution of ethylene-vinyl alcohol copolymer dissolved in dimethyl sulfoxide, which cures in situ once deployed. Polyvinyl alcohol powder was added to promote endocardialization, the process by which the heart's inner lining grows over the material.

The researchers determined magnet configurations and placement positions through simulations and benchtop experiments before moving to animals.

Preclinical results in pigs and rats

The technology was tested in Bama minipigs and Sprague-Dawley rats across both acute and chronic phases. Compared to clinical metallic occluders, the magnetofluid produced smoother, more uniform endocardial coverage. No thrombus formation was observed during the study period.

The adaptability advantage is straightforward: a liquid fills whatever shape it encounters. Metal devices, by contrast, come in a limited range of sizes and geometries. Left atrial appendages vary considerably between patients, and a poor fit can leave gaps that defeat the purpose of the procedure.

The gap this fills in clinical practice

Current left atrial appendage occlusion devices have been in use for years and are reasonably effective. But the complications they cause are not trivial. Device-related thrombosis, where clots form on the occluder itself, is reported in a meaningful percentage of patients. Peri-device leaks, where blood continues to flow around the edges, can persist and may require additional intervention.

Previous attempts to use liquid materials for occlusion failed because conventional liquids cannot withstand the mechanical forces inside the heart. Blood flows at high velocity through the left atrial appendage, and the vigorous contractions of the heartbeat create continuous disturbance. The magnetofluid addresses this by using magnetic retention to keep the material in place until the curing process locks it into a permanent seal.

What remains to be proven

This is preclinical work, and several questions remain unanswered. Long-term durability of the seal has not been established. The biocompatibility of the material over years or decades in a human heart is unknown. The chronic phase testing in animals, while encouraging, does not replicate the full lifespan exposure a human patient would experience.

There is also the question of retrievability. Metal devices, while imperfect, can in principle be removed or adjusted. A cured magnetofluid may not offer that option. If complications arise after deployment, the treatment path is unclear.

The transition from animal models to human clinical trials will require addressing these questions, along with demonstrating consistent results across the wide variety of appendage anatomies seen in patients with atrial fibrillation.

Still, the concept of replacing rigid metal implants with a conformable, self-curing material is a meaningful departure from current technology. If the approach proves safe and effective in humans, it could simplify a procedure that currently demands precise sizing and carries inherent risks of device mismatch.