The research, published in Frontiers in Transplantation, tackled a well-known weak point in the standard Toronto protocol for ex vivo lung perfusion, or EVLP. In clinical practice, EVLP is typically limited to four to six hours, during which a donor lung is warmed to body temperature, ventilated, and perfused with a nutrient-rich solution so transplant teams can judge whether a marginal organ is usable. During prolonged perfusion, however, the perfusate degrades: pH falls as lactate accumulates, glucose is consumed by metabolically active tissue, and electrolyte concentrations wander out of physiological range. Glucose depletion has previously been linked to edema and impaired cellular function, making the biochemical environment itself a candidate target for extending safe perfusion times.

The team, led by Justin Issard and colleagues at Marie Lannelongue Hospital and Université Paris-Saclay, harvested twelve heart-lung blocks from donor pigs and cold-preserved them before subjecting each to twelve hours of normothermic EVLP following the Toronto protocol. All lungs were perfused with Steen solution, the albumin- and dextran-containing perfusate used in most clinical EVLP programs, with mean priming volumes of roughly 1.3 to 1.5 liters. Every two hours, the researchers exchanged 100 milliliters of perfusate — about five to nine percent of the total volume. In the standard group, that exchange was fresh Steen solution. In the corrected group, it was a custom mixture of one-third Steen solution, one-third 2.5 percent glucose solution, and one-third 1.4 percent sodium bicarbonate, formulated to hold sodium, potassium, calcium, chloride, pH, and glucose within physiological limits.

The corrective strategy worked as designed. Throughout the twelve-hour perfusion, the corrected group maintained pH between 7.3 and 7.5, potassium between 3.8 and 4.2 millimoles per liter, chloride below 120 millimoles per liter, and glucose above 10 millimoles per liter. Statistical analysis of the time-course data, using two-way repeated-measures ANOVA, showed significant group effects and group-by-time interactions for pH measured in both the pulmonary artery and the left atrium, as well as for lactate, sodium, potassium, and chloride concentrations. Pulmonary vascular resistance, a key safety indicator, did not differ between groups or change over time, suggesting the correction strategy did not stress the pulmonary circulation.

Functionally, the corrected lungs performed better while on the circuit. Dynamic compliance — the ease with which the lungs inflate with each breath — was significantly higher in the corrected group throughout EVLP. Oxygenation, measured as the PaO2/FiO2 ratio, also increased over time in both groups but remained significantly higher overall in the corrected group. These are exactly the parameters transplant clinicians watch when deciding whether a marginal graft is safe to implant, so the finding suggests that biochemical homeostasis translates into measurable organ quality during prolonged perfusion.

Structural analysis told a more nuanced story. Histological scoring of the right lungs after EVLP showed that alveolar edema was substantially lower in the corrected group, a finding consistent with better fluid balance and preserved barrier function. Yet the same lungs showed more hyaline membranes — protein-rich deposits within alveoli that typically signal injury or repair — and electron microscopy revealed endothelial hypertrophy, subendothelial edema, and disrupted intercellular junctions in the corrected group that were absent in the standard group. The authors propose that hyaline membranes may reflect a reparative or inflammatory response to alveolar injury rather than ongoing edema, possibly driven by the increased glucose availability creating a pro-inflammatory alveolar microenvironment, or by altered perfusate protein composition favoring fibrin deposition.

One of the most intriguing structural findings involved the pores of Kohn, the microscopic channels that allow air to move laterally between adjacent alveoli. Scanning electron microscopy showed a significantly higher alveolar pore density in the corrected group — 711 plus or minus 307 pores per cubic millimeter compared with 529 plus or minus 210 in the standard group. Greater pore density may enhance collateral ventilation and contribute to the improved compliance and oxygenation observed, though the researchers caution that increased pore density has also been described in emphysematous remodeling, so the change could reflect a mixture of alveolar recruitment and tissue stress after twelve hours of normothermic perfusion.

Molecular and vascular analyses added further complexity. Western blotting showed decreased expression of endothelial nitric oxide synthase and its activated phosphorylated form in both perfused groups compared with control lungs, with further reductions in Twist and vascular endothelial cadherin, particularly in the standard group. Mitochondrial proteins MTERF and COX15 were upregulated in perfused lungs, significantly so in the corrected group, hinting at metabolic adaptation or stress. Vascular myography revealed that acetylcholine-induced, endothelium-dependent relaxation was impaired in all perfused lungs compared with controls, and more so in the corrected group, while smooth muscle-dependent relaxation with sildenafil was similar across groups. Cytokine levels in the perfusate, including IL-6, IL-18, TNF-alpha, and IL-1beta, were largely unaffected by the correction strategy.

The critical test — transplantation — produced a sobering result. Ten left lungs, five from each group, were deemed suitable and transplanted into donor-related pigs for four hours of observation. Post-transplant oxygenation, measured as the delta PO2 between pulmonary artery and left atrial blood and as the PaO2/FiO2 ratio, showed no significant difference between the corrected and standard groups. The authors attribute this partly to the presence of the recipients’ native right lung, which dilutes any graft-specific signal in the ventilatory and gas-exchange measurements, and partly to the short follow-up period. Selective ventilation or bilateral transplantation would be needed to isolate the transplanted lung’s contribution.

The study’s implications reach beyond the twelve-hour mark. Extending EVLP duration could unlock reconditioning interventions that are impractical within a few hours — antibiotic therapy for infected donor lungs, adenoviral gene transfer, immunomodulation, or mesenchymal stem cell administration — potentially expanding the pool of usable grafts at a time when demand far outstrips supply. The authors suggest that combining corrected perfusate composition with intermittent EVLP strategies, which alternate normothermic perfusion with cold storage phases to limit cumulative endothelial stress, could eventually enable lung preservation for several days. They also flag the key unresolved problems: the bicarbonate buffer caused hypernatremia that was difficult to control, endothelial protection remains an unmet need, and glucose delivery must be tuned to avoid hyperglycemic endothelial stress. Future work, they write, should move beyond simple biochemical correction toward targeted strategies that maintain endothelial integrity and capillary barrier function, evaluate 24-hour protocols, and explore adjuncts such as hemofiltration, adsorbent membrane filtration, and oncotic pressure management.

Subject of Research: Perfusate biochemical correction during prolonged ex vivo lung perfusion of donor lungs in a porcine transplantation model

Article Title: Prolonged ex vivo lung perfusion with corrected perfusate: functional and structural benefits in swine

Article References: Issard, J., Mercier, J., Ménager, J.-B., Fadel, G., Nobori, Y., Izem, M., Ghigna, M.-R., Fadel, E., Antigny, F., Mordant, P., & Mercier, O. (2026). Prolonged ex vivo lung perfusion with corrected perfusate: functional and structural benefits in swine. Frontiers in Transplantation, 5, Article 1899960. https://doi.org/10.3389/frtra.2026.1899960

Image Credits: AI Generated

DOI: 10.3389/frtra.2026.1899960

Keywords: ex vivo lung perfusion, lung transplantation, EVLP, perfusate correction, donor lungs, porcine model, Steen solution, lung edema, endothelial function, pores of Kohn, Toronto protocol, graft preservation