Can a Surfactant Do What the Chamber Does? A Lab-on-a-Chip Test of Gas Embolism Therapy

Stylised editorial illustration of a branching capillary network glowing teal and cyan against deep navy, with small oxygen molecules diffusing inward toward a darkened central region

On the arterial side of the circulation, a bubble the width of a hair can end a life. Cerebral arterial gas embolism — gas that reaches the arteries supplying the brain — is an uncommon but potentially catastrophic complication of cardiothoracic surgery, extracorporeal circulation, mechanical circulatory support and catheter-based cardiac interventions. Its definitive treatment is one of the oldest uses of the hyperbaric chamber: breathing 100 percent oxygen at two to three times normal atmospheric pressure, which compresses the offending bubble, accelerates its dissolution, and floods starved tissue with oxygen.

Which is why a paper published on 1 October in the Journal of Colloid and Interface Science is worth reading carefully. A team at McGill University in Montreal and Memorial University in Newfoundland did not test a chamber. They built a lab-on-a-chip — a transparent network of microchannels designed to mimic human microvasculature — and asked a narrower, more radical question. If the problem is a bubble, can the bubble itself be made less dangerous before any patient is ever pressurised?

Why a bubble does not simply dissolve

Gas embolism is often described as though the bubble were a free sphere of gas in the bloodstream, waiting to be squeezed away by Boyle’s law. The physiology is messier. Immediately after a bubble forms, surface-adsorbed biomolecules from blood assemble into what the McGill authors call a “resilient membrane” around it. That layer slows the rate at which gas can dissolve back into the liquid phase, which means the treatment window is set not only by pressure and timing but by interfacial chemistry at a scale no chamber can reach.

The rationale the authors give is unusually direct for a physical-chemistry journal. Therapeutic and prevention strategies, they write, could focus on microscale interfacial processes “instead of, or in addition to the present, macroscale hyperbaric oxygen therapy.” Experimental testing of anti-bubble drugs in living animals is slow, ethically constrained and hard to reproduce, so the group engineered the alternative: a chip.

What the chip actually showed

The platform is a microfluidic network with branching channels, blood at controlled haematocrit, and adjustable driving pressure. Into it the researchers introduced Pluronic F-127, a surfactant already approved for in vivo medical use, and tracked what happened to gas bubbles — average volume, velocity, spatial distribution across bifurcations, accumulation over time, extent of coalescence, and haematocrit distribution.

Pluronic F-127 significantly reduced bubble size and the risk of vessel occlusion. The team pressed further with dimensionless Weber, Capillary and Euler numbers to argue the effect is mechanistic rather than incidental: the surfactant lowers the likelihood that a bubble is trapped at a branch point in the first place. Pluronic F-127 is not an exotic compound — it is a triblock copolymer routinely used as an excipient in pharmaceutical formulations, which is part of why the result is interesting, and also part of why it should not be over-read.

A chip is not a patient

The study is benchtop. There are no animals in it and no patients; no outcome data, no dose, no evidence that a surfactant introduced into a human circulation would reach the right capillary bed, persist long enough to matter, or avoid the haemodynamic and coagulation consequences of putting a surface-active agent into blood. The honest next step is an animal model — and the animal literature on gas embolism is a warning about how stubborn this problem is.

In 2013, a group led by Dr. R.P. Weenink, then at the Diving Medical Center of the Royal Netherlands Navy, reported a porcine study in Critical Care Medicine in which hyperbaric oxygen started two or four hours after cerebral arterial gas embolism did not improve cerebral function. Two hours is fast by real-world standards, in a laboratory where the embolism was induced deliberately. If the chamber — the accepted treatment — loses measurable ground within hours, a surfactant tested only at the microscale has a very long way to travel.

The clock is the real adversary

Where the clinical evidence is strongest is not in any new molecule but in time. A 2023 systematic review and individual patient data meta-analysis in Critical Care, led by Dr. R.A. Fakkert at Amsterdam UMC, pooled ten observational studies covering 263 patients with iatrogenic cerebral arterial gas embolism. Patients with a favourable outcome were treated with hyperbaric oxygen 2.4 hours earlier than those who were not (95 percent confidence interval 0.6–9.7 hours). In a mixed-effects model across eight studies and 126 patients, the probability of a favourable outcome fell from roughly 65 percent when hyperbaric oxygen was started immediately to about 30 percent when treatment was delayed by 15 hours.

That gradient has been visible for decades. A 2002 retrospective in Intensive Care Medicine by Dr. P. Blanc and colleagues followed 86 patients treated between 1980 and 1999 and found that those treated within six hours fared better than those treated later — and that arterial gas embolism, unlike venous gas embolism, arrived at the chamber later and recovered less often. A 2013 series from the University of Minnesota reviewed 36 consecutive patients across 22 years and found that treatment within six hours independently predicted a favourable outcome, with an odds ratio of 9.08, while infarct or oedema already visible on CT or MRI before treatment sharply reduced the odds.

The Minnesota paper also flags a trap that matters well outside the hyperbaric unit. Gas embolism looks like a stroke. Two of its 36 patients were given thrombolytics on that assumption, and one died of an intracerebral haemorrhage. The 2025 UHMS recommendations by Drs. Richard Moon and Simon Mitchell make the diagnostic point explicitly: vascular gas is frequently invisible on radiographic imaging, and its absence should not be used to exclude the diagnosis of arterial gas embolism. How the disease is recognised is still a bigger lever on outcome than how the bubble is chemically modified.

An indication that is not in dispute

It bears saying plainly, because so much of the public conversation about hyperbaric therapy concerns uses the FDA has not cleared: gas embolism is not one of those. Air or gas embolism is the first entry on the Undersea and Hyperbaric Medical Society’s list of approved indications, and Medicare’s national coverage determination 20.29 names “gas embolism” among the conditions for which hyperbaric oxygen therapy is covered when administered in a chamber.

The distinction from the wellness market matters. Mild hyperbaric devices marketed at 1.3 to 1.5 ATA cannot deliver the pressures used to treat an arterial gas embolism, and they are not cleared for it. The FDA’s August 2025 letter to health care providers — still the agency’s current device-safety guidance, dated 25 August 2025 — records that it is “aware of recent reports of fires that occurred with HBOT devices that resulted in serious injuries and deaths,” and asks facilities to control prohibited electrical and static-generating items, train and monitor staff, and follow manufacturer instructions for every device.

What a chip can and cannot change

None of this makes the McGill work a distraction. The bottleneck in gas embolism is not the physics of pressure; it is geography and time. Patients with iatrogenic embolism are usually already inside a hospital — sometimes already on cardiopulmonary bypass — and their outcome is decided by how quickly someone recognises the event and gets them to a chamber that may be across the city. A surfactant that could be given at the bedside during a cardiac procedure, if it ever worked in animals and then in people, would attack exactly that gap.

Today the evidence for it is a plastic channel, a copolymer and a dimensionless number. The chamber remains the definitive treatment, and the clock remains the adversary. The chip is an argument, not a therapy — and the next thing to watch is whether the bubble’s coat behaves the same way inside a living artery.

Sources

← Back to blog