Breathing Liquid: Lung Volume Holds the Key to Safer Total Liquid Ventilation in Newborns

For the most fragile newborns on Earth—babies born so prematurely that their lungs cannot yet do the job they were designed for—medicine has long sought a radical alternative to conventional mechanical ventilation. One of the most audacious ideas in modern neonatal intensive care is total liquid ventilation, or TLV, a technique in which the lungs are filled entirely with an oxygen-carrying liquid and ventilated by a machine rather than by air. Now, a team of researchers in Canada and France has provided some of the most detailed evidence yet about what goes wrong inside the airways during this procedure, and their findings could reshape how the technique is made safe enough for human infants. The study, published in Pediatric Research, reveals that the amount of liquid left in the lungs at the end of each breath—the end-expiratory lung volume—is the decisive factor determining whether a newborn’s airways stay open or silently collapse.

Total liquid ventilation is not science fiction. The concept dates back to the early 1970s, when researchers demonstrated that the alveoli of fluorocarbon-filled lungs could exchange gases across a liquid interface. Perfluorocarbons, the synthetic liquids used in TLV, possess two remarkable properties: they dissolve large quantities of oxygen and carbon dioxide, and they have extremely low surface tension. Because a premature infant’s lungs are deficient in surfactant—the body’s own surface-tension-lowering substance—filling them with a perfluorocarbon can, in principle, recruit collapsed alveoli uniformly, reduce the sheer forces that injure delicate tissue, and permit gentle ventilation at pressures that would be dangerous with gas. In TLV, a specialized ventilator with independent piston pumps fills the lungs with oxygenated liquid and then drains it, cycle after cycle, in a rhythm that mimics breathing.

Yet the technique has an Achilles’ heel that has shadowed its development for decades: airway collapse during expiration. As the liquid ventilator drains fluid from the lungs, the pressure inside the trachea and bronchi falls, and in a newborn—whose airway cartilage is soft and immature—the walls of these tubes can buckle inward under the pressure difference between the inside and outside of the airway. Previous work, including computational models and experiments in lambs and rabbits, had identified expiratory flow limitation and airway deformation as hazards of liquid-filled lungs. What has been missing is a direct, quantitative, in vivo picture of how much the airways actually deform during real TLV cycles, and what determines whether that deformation becomes dangerous. The new study set out to fill that gap.

The research team, led by Kevin Le Duc and Étienne Fortin-Pellerin of the Université de Sherbrooke, together with colleagues in mechanical engineering and molecular imaging, worked with nine piglets aged five to seven days and weighing around two kilograms. Piglets at this age are a well-established model for the neonatal airway, sharing key anatomical and mechanical features with human infants. The animals were placed on total liquid ventilation using perfluorooctyl bromide, a dense, radiopaque perfluorocarbon whose visibility on X-ray-based imaging makes it an ideal tracer for lung volume studies. When signs of tracheobronchial collapse appeared, additional perfluorocarbon was added to restore a normal expiratory pressure, and this baseline state was designated V0, corresponding to a median end-expiratory lung volume of 37 milliliters per kilogram.

From that baseline, the researchers then deliberately manipulated lung volume in a randomized sequence. Additional perfluorocarbon was administered in increments of one, three, and five milliliters per kilogram, creating ventilation levels labeled V1, V3, and V5. At each level, the team performed two kinds of computed tomography imaging. Static CT scans measured the end-expiratory lung volume itself, while dynamic CT scans captured something that had never before been quantified in a living neonatal model during TLV: the cross-sectional area of the trachea continuously throughout the expiratory phase of each breath. This dynamic imaging approach, adapted from techniques used to evaluate airway function in adult patients, allowed the investigators to watch the airway walls move in real time as liquid was withdrawn from the lungs.

The results were striking. At the baseline volume V0, the tracheal cross-sectional area varied by a median of 35 percent between the beginning and the end of expiration—a degree of deformation that, in a clinical setting, would occur entirely invisibly to the bedside team. The ventilator’s pressure readings gave no warning: the study found that severe expiratory tracheobronchial deformation can take place without any detectable drop in airway opening pressure, challenging the very safety markers that clinicians and engineers currently rely on to detect trouble during liquid ventilation. In other words, an infant’s airway could be collapsing substantially with every breath while the monitoring equipment reported that everything was normal.

When the researchers raised the end-expiratory lung volume by adding perfluorocarbon, the picture changed dramatically. At ventilation level V3, corresponding to a median end-expiratory lung volume of 42 milliliters per kilogram, tracheal deformation fell to a median of 18 percent, a statistically significant reduction. The mechanism is physically intuitive: a larger residual liquid volume at end-expiration maintains higher pressure within the airways throughout the expiratory phase, providing internal structural support that resists the compressive forces trying to collapse the soft-walled tubes. The liquid acts, in effect, as a hydraulic stent, splinting the trachea and bronchi open from the inside. This finding establishes, for the first time in vivo, a direct causal link between lung volume and airway stability during total liquid ventilation.

But the therapeutic window proved to be narrow, and the upper boundary of that window carries its own risks. When the team pushed lung volume to the highest level, V5, they observed a significant decrease in mean arterial pressure between the baseline and the highest volume condition. This hemodynamic consequence is consistent with a long-recognized principle in mechanical ventilation: overdistending the lungs raises intrathoracic pressure, which impedes venous return to the heart and reduces cardiac output. In a premature infant with limited cardiovascular reserve, such a drop in blood pressure could compromise perfusion of the brain, the heart, and other vital organs. Notably, blood gas values remained unchanged across all ventilation levels, meaning that the conventional measures of ventilation adequacy—oxygenation and carbon dioxide removal—gave no indication of either the airway deformation at low volumes or the circulatory compromise at high volumes.

The implications of this study extend well beyond the laboratory. For engineers designing the next generation of liquid ventilators, the message is that volume control, not merely pressure control, must be at the heart of the machine’s regulation strategy. The expiratory volume profile—the precise way liquid is withdrawn and how much is deliberately left behind at end-expiration—emerges as a critical, tunable parameter that can be optimized to keep airways open. For clinicians and translational researchers hoping to bring TLV to human neonates, the study defines a target range for end-expiratory lung volume, roughly between the levels that caused dangerous deformation and those that depressed blood pressure, and it warns that the currently used safety indicators are insufficient on their own. Complementary monitoring strategies, potentially including imaging or airway mechanics measurements, may be needed to detect occult collapse.

The road to clinical application remains long, and the authors are careful to frame their work as mechanistic evidence from an animal model rather than a recipe for the bedside. Piglets are not human infants, and the specific volume thresholds identified here cannot be transplanted directly into the neonatal intensive care unit. Nevertheless, the study delivers essential data for the safer translational development of a technique that could one day rescue babies whose lungs are too immature for any conventional approach. It also reinforces a broader lesson in respiratory medicine: the most dangerous events are sometimes the ones that leave no trace on the monitors. By making the invisible visible through dynamic CT imaging, this research has turned a hidden hazard into a measurable, and therefore manageable, quantity—bringing the dream of letting newborns breathe liquid one careful step closer to reality.

Subject of Research: The effect of end-expiratory lung volume on tracheal deformation and hemodynamic stability during total liquid ventilation in neonatal piglets

Article Title: Lung volume determines airway stability during total liquid ventilation: evidence from neonatal piglets

Article References: Lung volume determines airway stability during total liquid ventilation: evidence from neonatal piglets. (n.d.). https://doi.org/10.1038/s41390-026-05423-4

Image Credits: AI Generated

DOI: 10.1038/s41390-026-05423-4

Keywords: total liquid ventilation, perfluorocarbon, neonatal piglets, airway collapse, end-expiratory lung volume, dynamic CT imaging, tracheal deformation, premature infants, hemodynamics, pediatric research, respiratory mechanics, liquid ventilator