OBJECTIVE: Lung transplantation remains limited by the shortage of healthy organs. Cross-circulation with a healthy swine recipient provides a durable physiologic environment to recover injured donor lungs. In a clinical application, a recipient awaiting lung transplantation could be placed on cross-circulation to recover damaged donor lungs, enabling eventual transplantation. Our objective was to assess the ability of recipient swine with respiratory compromise to tolerate cross-circulation and support recovery of donor lungs subjected to extended cold ischemia.
METHODS: Swine donor lungs (n = 6) were stored at 4 °C for 24 hours while recipient swine (n = 6) underwent gastric aspiration injury before cross-circulation. Longitudinal multiscale analyses (blood gas, bronchoscopy, radiography, histopathology, cytokine quantification) were performed to evaluate recipient swine and extracorporeal lungs on cross-circulation.
RESULTS: Recipient swine lung injury resulted in sustained, impaired oxygenation (arterial oxygen tension/inspired oxygen fraction ratio 205 ± 39 mm Hg vs 454 ± 111 mm Hg at baseline). Radiographic, bronchoscopic, and histologic assessments demonstrated bilateral infiltrates, airway cytokine elevation, and significantly worsened lung injury scores. Recipient swine provided sufficient metabolic support for extracorporeal lungs to demonstrate robust functional improvement (0 hours, arterial oxygen tension/inspired oxygen fraction ratio 138 ± 28.2 mm Hg; 24 hours, 539 ± 156 mm Hg). Multiscale analyses demonstrated improved gross appearance, aeration, and cellular regeneration in extracorporeal lungs by 24 hours.
CONCLUSIONS: We demonstrate that acutely injured recipient swine tolerate cross-circulation and enable recovery of donor lungs subjected to extended cold storage. This proof-of-concept study supports feasibility of cross-circulation for recipients with isolated lung disease who are candidates for this clinical application.

UNASSIGNED: Cytomegalovirus infection after lung transplant is associated with increased morbidity and mortality. Inflammation, infection, and longer ischemic times are important risk factors for cytomegalovirus infection. Ex vivo lung perfusion has helped to successfully increase the use of high-risk donors over the last decade. However, the impact of ex vivo lung perfusion on post-transplant cytomegalovirus infection is unknown.
UNASSIGNED: We performed a retrospective analysis of all adult lung transplant recipients from 2010 to 2020. The primary end point was comparison of cytomegalovirus viremia between patients who received ex vivo lung perfusion donor lungs and patients who received non-ex vivo lung perfusion donor lungs. Cytomegalovirus viremia was defined as cytomegalovirus viral load greater than 1000 IU/mL within 2 years post-transplant. Secondary end points were the time from lung transplant to cytomegalovirus viremia, peak cytomegalovirus viral load, and survival. Outcomes were also compared between the different donor recipient cytomegalovirus serostatus matching groups.
UNASSIGNED: Included were 902 recipients of non-ex vivo lung perfusion lungs and 403 recipients of ex vivo lung perfusion lungs. There was no significant difference in the distribution of the cytomegalovirus serostatus matching groups. A total of 34.6% of patients in the non-ex vivo lung perfusion group developed cytomegalovirus viremia, as did 30.8% in the ex vivo lung perfusion group (P = .17). There was no difference in time to viremia, peak viral loads, or survival when comparing both groups. Likewise, all outcomes were comparable in the non-ex vivo lung perfusion and ex vivo lung perfusion groups within each serostatus matching group.
UNASSIGNED: The practice of using more injured donor organs via ex vivo lung perfusion has not affected cytomegalovirus viremia rates and severity in lung transplant recipients in our center.

Xenotransplantation promises to alleviate the issue of donor organ shortages and to decrease waiting times for transplantation. Recent advances in genetic engineering have allowed for the creation of pigs with up to 16 genetic modifications. Several combinations of genetic modifications have been associated with extended graft survival and life-supporting function in experimental heart and kidney xenotransplants. Lung xenotransplantation carries specific challenges related to the large surface area of the lung vascular bed, its innate immune system\'s intrinsic hyperreactivity to perceived \'danger\', and its anatomic vulnerability to airway flooding after even localized loss of alveolocapillary barrier function. This article discusses the current status of lung xenotransplantation, and challenges related to immunology, physiology, anatomy, and infection. Tissue engineering as a feasible alternative to develop a viable lung replacement solution is discussed.

VIDEO ABSTRACT.

OBJECTIVE: Ex vivo lung perfusion has emerged as a novel technique to safely preserve lungs before transplantation. Recent studies have demonstrated an accumulation of inflammatory molecules in the perfusate during ex vivo lung perfusion. These proinflammatory molecules, including damage-associated molecular patterns and inflammatory cytokines, may contribute to acute and chronic allograft dysfunction. At present, ex vivo lung perfusion is performed clinically at normothermic temperature (37°C). The effect of lowering temperature to the subnormothermic range during ex vivo lung perfusion has not been reported. In this study, we hypothesized that lower ex vivo lung perfusion temperature will lead to a reduction in allograft inflammation and result in improved post-transplant graft function.
METHODS: Lewis rat heart-lung blocs underwent 4 hours of ex vivo lung perfusion in 3 temperature groups: 37°C (MP37), 30°C (MP30), and 25°C (MP25). In the control group, lung grafts were preserved by static cold storage before transplantation. After ex vivo lung perfusion or static cold storage, the left lung was transplanted for 2 hours before the animal was killed. Sera and tissue were collected and analyzed.
RESULTS: There were no differences in partial pressure of arterial oxygenation to fraction of inspired oxygen ratios during 4 hours of ex vivo lung perfusion between temperature groups. Tumor necrosis factor α significantly increased in the MP37 group during ex vivo lung perfusion, whereas this was not seen at lower temperatures. Extracellular DNA and high-mobility group box 1 perfusate concentrations increased significantly during ex vivo lung perfusion in all groups, but the rate of increase was diminished at lower temperature. Two hours post-transplant, there were no significant differences in partial pressure of arterial oxygenation to fraction of inspired oxygen ratios of the lung graft or serum damage-associated molecular pattern levels among groups. On histologic grading after transplantation, greater injury was observed in the MP30 and MP37 groups, but not MP25, when compared with static cold storage.
CONCLUSIONS: Subnormothermic ex vivo lung perfusion at 25°C reduces the production of inflammatory mediators during ex vivo lung perfusion and is associated with reduced histologic graft injury after transplantation.

Despite progress in modern medical therapy, pulmonary hypertension remains an unremitting disease. Once severe or refractory to medical therapy, advanced percutaneous and surgical interventions can palliate right ventricular overload, bridge to transplantation, and overall extend a patient\'s course. These approaches include atrial septostomy, Potts shunt, and extracorporeal life support. Bilateral lung transplantation is the ultimate treatment for eligible patients, although the need for suitable lungs continues to outpace availability. Measures such as ex vivo lung perfusion are ongoing to expand donor lung availability, increase rates of transplant, and decrease waitlist mortality.

Cell-free DNA (cfDNA), such as mitochondrial DNA (mtDNA) and nuclear DNA (nuDNA), are known to be released from injured cells and as such have been explored as biomarkers for tissue injury in different clinical settings. Ex vivo lung perfusion (EVLP) has been developed as an effective technique for marginal donor lung functional assessment. We hypothesized that the level of cfDNA in EVLP perfusate may reflect tissue injury and thus can be developed as a biomarker to quantify the degree of donor lung injury or predict the development of primary graft dysfunction (PGD) after lung transplantation (LTx).
The perfusate from 62 donor lungs transplanted at our institution between May 2010 and December 2015 was sampled for cfDNA at 1 and 4 hours of perfusion. Sequences of genes encoding nicotinamide adenine dinucleotide dehydrogenase 1 (NADH-1) and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) were used to represent mtDNA and nuDNA, respectively. Levels were quantified by real-time polymerase chain reaction and correlated with clinical outcome after LTx.
In our entire cohort, 14 patients developed PGD grade 3 (PGD3) within 72 hours after LTx. The non-PGD group included 48 patients (PGD0-1). Concentrations of mtDNA in the perfusate of the PGD3 group were significantly higher than those in non-PGD group at 1 hour of EVLP (1874 ± 844 vs 1259 ± 885 copies/μL; P = .011). The perfusate of the PGD3 group had significantly higher levels of nuDNA compared with the non-PGD group at both 1 hour (1498 ± 1895 vs 675 ± 391 copies/μL; P = .008) and 4 hours (4521 ± 5810 vs 1764 ± 1494 copies/μL; P = .001). In donation after cardiac death (DCD) cases, mtDNA levels were significantly higher in the PGD3 group compared with the non-PGD group at 1 hour of EVLP (2060 ± 997 vs 1184 ± 782 copies/μL; P = .040), and the levels of nuDNA were significantly higher in the PGD3 group compared with the non-PGD group at both 1 hour (1021 ± 495 vs 606 ± 305 copies/μL; P = .041) and 4 hours (2358 ± 1028 vs 1185 ± 967 copies/μL; P = .006). In donation after brain death (DBD) cases, cfDNA scores did not show a significant difference.
We found that the amount of cfDNA, especially nuDNA, in EVLP perfusate was higher in the severe PGD group (PGD3) compared with the non-PGD group. This proof-of-concept study supports the concept that the analysis of cfDNA levels in EVLP perfusate can help estimate the damage to donor lungs before implantation. Larger studies are needed to validate this concept.

OBJECTIVE: Ex vivo lung perfusion (EVLP) is reportedly a useful strategy that permits marginal donor lungs to be evaluated and reconditioned for successful lung transplantation (LTx). This systematic review and meta-analysis was performed to evaluate the outcomes of EVLP conducted for marginal donor lungs.
METHODS: We searched PubMed, the Cochrane Library, and Embase to select studies describing the results of LTx following EVLP for marginal donor lungs compared with standard LTx without EVLP. We performed a meta-analysis to examine donor baseline characteristics, recipient baseline characteristics, and postoperative outcomes.
RESULTS: Of 1380 studies, 8 studies involving 1191 patients met the inclusion criteria. Compared with the non-EVLP group (ie, standard LTx without EVLP), the EVLP group (ie, EVLP of marginal donors following LTx) had similar donor age and sex and recipient baseline age, sex, body mass index, bridge by ventilator/extracorporeal life support/extracorporeal membrane oxygenation, and rate of double LTx but more abnormal donor lung radiographs (P = .0002), a higher smoking history rate (P = .03), and worse donor arterial oxygen tension/inspired oxygen fraction (P < .00001). However, there were no significant differences in outcomes between the EVLP and non-EVLP groups with respect to the length of postoperative intubation, postoperative extracorporeal life support/extracorporeal membrane oxygenation use, length of intensive care unit stay, length of hospital stay, 72-hour primary graft dysfunction of grade 3, 30-day survival, or 1-year survival (all P values > .05).
CONCLUSIONS: Posttransplant outcomes were similar between EVLP-treated LTx and standard LTx without EVLP, although the quality of donor lungs was worse with EVLP-treated LTx.

Ex vivo lung perfusion (EVLP) is developed to increase the quantity and quality of suitable grafts for lung transplantation. Standardly, lungs are mounted supine with the risk of fluid accumulation in the dorsal regions. Therefore, we investigated the impact of experimental prone position on graft function during EVLP.
Porcine lungs were mounted on a normothermic EVLP for 6 h in supine [S], (n = 7) or prone position [P], (n = 7). Physiology during EVLP was recorded. After EVLP, biopsies were assessed for wet-to-dry weight (W/D) ratios and pathology, broncho-alveolar lavage was measured, and the left lung was computed tomography (CT) scanned.
Physiological parameters were similar between both groups, despite a higher pulmonary vascular resistance in [P] (P = 0.0002). In [S], W/D ratios and CT density of dorsal areas were higher compared to ventral (P = 0.0017 and P = 0.053, respectively). In [P], W/D and CT density between ventral and dorsal regions were similar, meaning that pulmonary edema was distributed more homogeneously throughout the lung. Histology and cytokine levels in perfusate and broncho-alveolar lavage did not differ between both groups.
Prone positioning during EVLP is feasible and leads to more homogenous distribution of interstitial fluid. Supine position resulted in more concentrated edema accumulation in lower dependent regions.

For more accurate lung evaluation in ex vivo lung perfusion (EVLP), we have devised a new parameter, PaO2 /FiO2 ratio difference (PFD); PFD1-0.4  = P/F ratio at FiO2 1.0 - P/F ratio at FiO2 0.4. The aim of this study is to compare PFD and transplant suitability, and physiological parameters utilized in cellular EVLP. Thirty-nine human donor lungs were perfused. At 2 h of EVLP, PFD1-0.4 was compared with transplant suitability and physiological parameters. In a second study, 10 pig lungs were perfused in same fashion. PFD1-0.4 was calculated by blood from upper and lower lobe pulmonary veins and compared with lobe wet/dry ratio and pathological findings. In human model, receiver operating characteristic curve analysis showed PFD1-0.4 had the highest area under curve, 0.90, sensitivity, 0.96, to detect nonsuitable lungs, and significant negative correlation with lung weight ratio (R2  = 0.26, P < 0.001). In pig model, PFD1-0.4 on lower and upper lobe pulmonary veins were significantly associated with corresponding lobe wet/dry ratios (R2  = 0.51, P = 0.019; R2  = 0.37, P = 0.060), respectively. PFD1-0.4 in EVLP demonstrated a significant correlation with lung weight ratio and allowed more precise assessment of individual lobes in detecting lung edema. Moreover, it might support decision-making in evaluation with current EVLP criteria.