Endotracheal UVA therapy slowed early Pseudomonas aeruginosa spread in a porcine VAP model
Ventilator-associated pneumonia remains one of the most consequential complications of invasive mechanical ventilation, especially when extensively drug-resistant Pseudomonas aeruginosa is the organism at the center of the infection. In this sham-controlled large animal study, an endotracheally inserted narrow-band ultraviolet A catheter was tested as a way to blunt airway bacterial expansion before the infection could take hold in the lung parenchyma. The premise is biologically straightforward: if local antimicrobial energy can reduce tracheal colonization early enough, the downstream inflammatory cascade that leads to pneumonia may be delayed or partially interrupted.
Nine of ten intubated pigs completed the experiment. After inoculation with XDR P. aeruginosa, the animals were randomized to receive either 20-minute UVA sessions twice daily or sham treatment over a 100-hour observation period. Tracheal aspirates, bronchoalveolar lavage fluid, lung tissue, and tracheal samples were evaluated microbiologically and histologically, while inflammatory markers and catheter safety were monitored throughout the protocol.
The most immediate finding was not a dramatic elimination of bacteria from the lungs. Lung tissue P. aeruginosa burden was numerically lower in the UVA arm but did not differ significantly from sham treatment, with values of 1.48 [0–2.00] log CFU/g versus 3.37 [2.23–4.32] log CFU/g, respectively. The more convincing signal appeared earlier in the airway itself. Treated animals had significantly lower tracheal P. aeruginosa concentrations at 12 hours, and microbiologically confirmed pneumonia involved fewer lobes overall, 4 of 25 versus 15 of 25 in the control group.
Early airway suppression, not sterile lungs, defined the effect
That pattern fits the expected biology of ventilator-associated infection. In mechanically ventilated patients, bacterial adherence to the endotracheal tube and proximal trachea can serve as the reservoir from which organisms descend into distal airspaces. An intervention that reduces this reservoir may not completely sterilize lung tissue within a short window, yet it can still alter the tempo of spread and the extent of lobar involvement. The data here support exactly that type of intervention effect: transient attenuation of airway colonization with a measurable downstream impact on the early anatomy of pneumonia.
The histopathology tells the same story from a tissue perspective. UVA-treated animals had less neutrophil-mediated interstitial infiltration, indicating a weaker inflammatory response in the pulmonary tissue. In a setting where neutrophils are recruited rapidly to a contaminated lung, a lower infiltrative burden suggests less intense host injury during the early phase of infection. The epithelial lining fluid cytokine profile also shifted in an anti-inflammatory direction, with increased IL-10 and reduced IFN-γ. Those changes are compatible with a local immune milieu that is less primed for sustained tissue-damaging inflammation.
A single sentence captures the main translational tension here: the lung burden was not clearly eliminated, but the infection spread less aggressively.
Safety signals were limited, and tissue injury was not obvious
Safety is the question that has to be answered before any light-based endotracheal therapy can move closer to human study. In this porcine experiment, no histological abnormalities were detected, p53 expression did not change, and no major adverse events were observed. For an intratracheal device emitting UVA light, those observations matter because they speak to both mucosal tolerance and the possibility of unexpected photobiological injury.
The absence of p53 perturbation is particularly relevant in this context. p53 is often used as a marker when investigators are looking for cellular stress or potential genotoxic effects after exposure to a physical or chemical intervention. Within the limits of this short animal study, there was no signal that the chosen UVA exposure created overt tissue damage or a detectable carcinogenic stress response in the trachea or lung.
Still, the safety assessment was necessarily narrow. The study spans 100 hours, not weeks or months, and porcine anatomy is not a substitute for the complexity of prolonged human intensive care. The findings support local tolerability under the tested conditions, but they do not establish how repeated use would behave in a critically ill patient population exposed to multiple antibiotics, suctioning procedures, evolving secretions, and prolonged ventilation.
Why the cytokine shift matters for ventilator-associated pneumonia prevention
A VAP-prevention strategy does not need to sterilize every airway compartment to have clinical value. If it can interrupt the bacterial and inflammatory sequence early enough, it may reduce the probability that tracheobronchial colonization matures into frank pneumonia. The rise in IL-10 alongside the fall in IFN-γ is consistent with a microenvironment less dominated by pro-inflammatory signaling, which may help explain the lower neutrophil infiltration and the reduced number of lobes with microbiologically confirmed pneumonia.
That does not mean the device has proven clinical efficacy in patients. It does mean the mechanism is plausible and the signal is coherent across culture, histology, and cytokine data. The intervention appears to act at the interface between microbial load and host response, which is precisely where a prevention strategy for ventilator-associated pneumonia would need to operate.
The sponsor provided financial support for the work, while the investigators reported that the sponsor had no role in data collection, data analysis, interpretation, or manuscript writing. The study was conducted under animal ethics approval in Barcelona. Those details do not change the biologic result, but they frame the work as an early translational step rather than a definitive clinical proof.
What this means before human testing begins
The most defensible reading of this study is restrained. Endotracheal UVA therapy did not eradicate lung infection, yet it lowered early tracheal bacterial concentration, reduced the number of lobes with confirmed pneumonia, and dampened the inflammatory footprint in a porcine model of XDR P. aeruginosa VAP. That combination suggests a transient delay in airway colonization and a partial brake on early pneumonia progression.
For bedside practice, the implications are still indirect. Mechanically ventilated patients with high-risk gram-negative colonization are exactly the population where a local, non-antibiotic adjunct would be attractive, especially when resistance narrows therapeutic options. Even so, no clinical recommendation can follow from a nine-animal study. Human studies will need to determine whether the effect survives longer observation, whether repeated use remains safe, and whether the same reduction in airway burden translates into fewer clinically meaningful pneumonia events rather than only microbiologic shifts.
The next step is not to assume benefit, but to test whether this local phototherapy can be integrated into the highly vulnerable period after intubation, when airway colonization can turn into full-blown ventilator-associated pneumonia with little margin for error.