PMC:5400024 / 9870-10005 JSON TXT

Profiling inflammation and tissue injury markers in perfusate and bronchoalveolar lavage fluid during human ex vivo lung perfusion Abstract Abstract OBJECTIVES: Availability of donor lungs suitable for transplant falls short of current demand and contributes to waiting list mortality. Ex vivo lung perfusion (EVLP) offers the opportunity to objectively assess and recondition organs unsuitable for immediate transplant. Identifying robust biomarkers that can stratify donor lungs during EVLP to use or non-use or for specific interventions could further improve its clinical impact. METHODS: In this pilot study, 16 consecutive donor lungs unsuitable for immediate transplant were assessed by EVLP. Key inflammatory mediators and tissue injury markers were measured in serial perfusate samples collected hourly and in bronchoalveolar lavage fluid (BALF) collected before and after EVLP. Levels were compared between donor lungs that met criteria for transplant and those that did not. RESULTS: Seven of the 16 donor lungs (44%) improved during EVLP and were transplanted with uniformly good outcomes. Tissue and vascular injury markers lactate dehydrogenase, HMGB-1 and Syndecan-1 were significantly lower in perfusate from transplanted lungs. A model combining IL-1β and IL-8 concentrations in perfusate could predict final EVLP outcome after 2 h assessment. In addition, perfusate IL-1β concentrations showed an inverse correlation to recipient oxygenation 24 h post-transplant. CONCLUSIONS: This study confirms the feasibility of using inflammation and tissue injury markers in perfusate and BALF to identify donor lungs most likely to improve for successful transplant during clinical EVLP. These results support examining this issue in a larger study. INTRODUCTION Normothermic ex vivo lung perfusion (EVLP) has emerged as a promising technique to expand the donor pool by assessing and reconditioning donor lungs previously considered unsuitable for transplant [1]. The lung is highly susceptible to injury in the critical care environment, and in the hours or days leading up to the donor’s demise it may be exposed to the sequelae of brain-stem death together with infection, aspiration, barotrauma, fluid overload or multiple transfusions [2–4]. Because the extent of lung injury is difficult to assess at the time of organ procurement, donor acceptance criteria are therefore conservative and poor discriminators of injury and infection in the donor lung [5]. The increased use of extended-criteria donors may further elevate the risk of primary graft dysfunction (PGD) and other more severe postoperative complications [6–8]. The use of ex vivo reconditioned donor lungs is steadily growing and now accounts for about 20% of the activity in some established centres [9–11]. The decision to accept organs for transplant after EVLP has, however, been based on the same questionable physiological measures and organ appearance used during standard procurement. Reported discard rates of 10–60% of perfused lungs suggest that some donor lungs may be inappropriately used or declined for transplant after EVLP [1]. In donor lungs transplanted without exvivo evaluation, there is an established relationship between their inflammatory burden and early outcome [12–15]. However, this phenomenon has not been as extensively investigated during EVLP. In this proof-of-concept study, we evaluated a panel of inflammatory mediators and tissue injury-associated proteins in both perfusate and bronchoalveolar lavage fluid (BALF) from human donor lungs undergoing clinical EVLP with intent for transplant. The panel was based on our own previous work and on available studies of biological markers in standard lung transplant and preclinical and clinical observations during EVLP. The aim was to assess the potential for specific protein markers in perfusate and BALF to distinguish which donor lungs, initially deemed unsuitable for immediate transplantation are most likely to successfully recondition during EVLP and thereby provide a basis for further investigations in a larger validation cohort. MATERIALS AND METHODS Study design We conducted a prospective observational study investigating protein expression in donor lungs exposed to a standardized EVLP protocol. Approval was granted by our local research ethics committee, and informed consent for research was obtained from donor families and lung transplant recipients (REC 09/H0905/10). Ex vivo lung perfusion protocol Adult donor lungs deemed unsuitable for lung transplant by all five UK lung transplant centres but meeting strict EVLP criteria were included in the study (Supplementary Table S1), the results of which were previously published by our group [16]. Lungs were procured in a routine fashion and transported to our institution. The EVLP assessment followed a standardized acellular protocol using the Toronto technique with a closed left atrium and reduced perfusate flow, previously described in detail [3, 16]. Transplant suitability was assessed hourly during perfusion. Lungs meeting transplant criteria (Supplementary Table S2) at two consecutive time points were cooled and transplanted. Lungs deemed to have futile prospects for improvement were taken off the circuit and discarded. Two transplanted and five non-transplanted lungs were perfused ≥5h before a transplant decision was made. Sample collection and processing A research BAL was performed for all lungs by wedging an adult bronchoscope in a subsegmental bronchus of the right or left lower lobe [17]. Saline (40 ml) was instilled through the suction channel followed by gentle aspiration and sample collection prior to commencing ventilation at the beginning of EVLP. This process was repeated in the same lobe but in a different subsegmental bronchus before disconnecting the ventilation at the end of perfusion. In addition, hourly perfusate samples of 2.5 ml were collected until the assessment was stopped. Protein expression analysis All protein expressions measured in perfusate were adjusted to the predicted total lung capacity (pTLC) of the donor as an estimate of perfused donor lung volume and were reported as corrected perfusate concentrations (pg/ml). The pTLC was calculated in a routine fashion based on donor gender and height [18]. If one lung was deemed unusable due to severe consolidation or extensive contusion on inspection, or if the intended recipient required a single lung transplant on a specific side, only one lung was procured. For single-lung perfusions, the pTLC was adjusted to a factor of 0.55 for right lung and 0.45 for left lung perfusion [19]. Lactate dehydrogenase assay Lactate dehydrogenase (LDH) levels were measured in perfusate and BALF with a colorimetric LDH cytotoxicity assay kit according to manufacturer’s instructions and reported as arbitrary units (U) (Thermo Fisher Scientific Inc., Rockford, IL, USA). Multiplex inflammatory cytokine array Interleukin (IL)-1β, IL-6, IL-8, TNF-α, and IL-10 were analysed with an MSD Multi-Array® (Meso Scale Diagnostics, LLC, Rockville, MD, USA). The assay was performed according to the manufacturer’s instructions. Technical issues prevented reading of perfusate samples from donor lung EVLP03; therefore, these samples were excluded from the analysis. Enzyme-linked immunosorbent assays Syndecan-1, IL-33, S100A9 (all R&D Systems, Inc., Minneapolis, MN, USA) and high-mobility group box-1 (HMGB-1) (Shino-Test Corporation, Kanagawa, Japan) were measured with commercially available ELISA kits according to the manufacturers’ instructions. Statistical analysis Donor characteristics and physiological parameters are expressed as medians with interquartile ranges and were compared between transplanted and non-transplanted lungs using Mann–Whitney U tests. Paired samples, start-end of perfusion, were compared with Wilcoxon signed-ranks tests. Log protein expressions were compared between transplanted and non-transplanted lungs with multiple t-tests. Correlations between IL-8 and IL-1β expressions and seven post-transplant outcomes (PaO2:FiO2 24-h post-transplant; PGD3 at 72 h post-transplant; ventilation time; intensive care unit stay hospital stay‘; percent of predicted FEV1 at 6 months post-transplant; and percent of predicted FVC at 6 months post-transplant) were analysed by Pearson’s correlation tests. The data were transformed back into non-logged values for reporting as mean (T) for transplanted and mean (NT) for non-transplanted lungs with standard deviations (SD). Multiple testing of donor parameters, protein analyses, and correlations was corrected using the Benjamini-Hochberg false discovery rate (FDR) controlling procedure [20]. Because this was a feasibility study aiming to identify potential markers for further investigation in a validation cohort, an FDR corrected P-value <0.1 was deemed significant. Corrected P-values are reported. A multiple logistic regression model was fitted using the 11 log transformed protein covariates and their squared counterparts as independent variables and the EVLP outcome as the dependent variable. The optimal model was established by leave-one-out cross-validation using the software package R from R Core Team (2014) (R Foundation for Statistical Computing, Vienna, Austria) [21]. RESULTS Study group and donor characteristics Sixteen consecutive clinical EVLP assessments of lungs deemed unacceptable for standard transplant were included in the study (Table 1). Seven (44%) were accepted for transplant after EVLP assessment and were implanted as four bilateral and three single-lung procedures. Nine (56%) failed to achieve transplant criteria and were excluded. The characteristics of the donor groups of transplanted and non-transplanted lungs were not significantly different and are shown in Tables 1 and 2. Table 1: Donor characteristics Donor no. Age/ gender Smoker Donor cause of death Donor type TLC (L) Micro on offer Radiographic infiltrates Secretions EVLP indication EVLP 01 18 m No Diabetic coma DL DBD 7.30 Nil Bilateral Nil PaO2 <300 mmHg EVLP 02 51 f Yes ICH DL DBD 4.31 Nil Clear Nil PaO2 <300 mmHg EVLP 03 33 m Yes ICH DL DBD 7.14 Nil Left diffuse Purulent PaO2 <300 mmHg EVLP 04 15 f No Hanging DL DCD 5.10 Nil Bilateral Nil PaO2 <300 mmHg EVLP 05 52 m No ICH DL DBD 5.78 Nil Clear Mucopurulent PaO2 <300 mmHg EVLP 06 57 m Yes ICH DL DBD 7.54 Nil Right basal Mucopurulent oedema EVLP 07 52 f Yes ICH SL DBD 2.70 Nil Bibasal Purulent PaO2 <300 mmHg EVLP 08 20 m Yes Hanging SL DBD 3.18 Nil Right basal Nil PV PaO2 <225 mmHg + oedema EVLP 09 36 m No Brain tumour SL DBD 3.11 Yesa Left diffuse Nil PaO2 <300 mmHg EVLP 10 47 m No HBI DL DBD 6.74 Nil Clear Nil PaO2 <300 mmHg EVLP 11 45 f No TBI DL DBD 6.18 Nil Clear Nil PV PaO2 <225 mmHg + mild contusion RLL EVLP 12 56 m No ICH DL DBD 6.50 Nil Clear Purulent PaO2 <300 mmHg EVLP 13 18 f Yes ICH DL DCD 5.03 Nil Bilateral Mucopurulent oedema EVLP 14 53 m Yes ICH DL DBD 7.06 Nil Clear Purulent PaO2 <300 mmHg EVLP 15 48 f No ICH DL DBD 4.57 Nil Bibasal Purulent PaO2 <300 mmHg EVLP 16 38 m Yes Hanging DL DCD 6.98 Nil Left basal Nil oedema Shaded donor lungs were not transplanted after EVLP . BALF: bronchoalveolar lavage fluid; EVLP : ex vivo lung perfusion; ICH: intracranial haemorrhage; HBI: hypoxic brain injury secondary to variceal bleed; TBI: traumatic brain injury following a road traffic accident; DL: double lung perfusion; SL: single lung perfusion; DBD: organ donation after brain death; DCD: organ donation after circulatory death; TLC: predicted total lung capacity of perfused lung; PaO2: partial pressure of arterial oxygen; PV: selective pulmonary vein blood gas analyses; RLL: right lower lobe. a Sputum culture with moderate growth of Staphylococcus aureus, Haemophilus influenzae, and Streptococcus pneumoniae. Table 2: EVLP and transplant outcomes Donor no. Optimized donor PaO2:FiO2 (mmHg) PaO2:FiO2 after EVLP (mmHg) Ischaemic time (min)a EVLP time (min)b Total time ex vivo (min)c Transplant PGD 3 at 72 h 90-day survival EVLP 01 149 521 345 290 1020 Yes–BL Yes Yes EVLP 02 248 342 360 360 N/A No–emphysema EVLP 03 209 167 330 270 N/A No–gross oedema EVLP 04 222 426 450 360 870 Yes–SL No Yes EVLP 05 167 440 320 330 N/A No–heavy gram-stain in BALF EVLP 06 357 525 395 240 582 Yes–SL No Yes EVLP 07 178 557 315 255 1048 Yes–SL Yes Yes EVLP 08 372 392 340 240 N/A No–emphysema EVLP 09 293 330 430 300 N/A No–gross oedema EVLP 10 171 291 305 300 N/A No–gross oedema EVLP 11 526 452 340 330 N/A No–gross oedema EVLP 12 191 512 395 220 785 Yes–BL No Yesd EVLP 13 360 352 345 180 N/A No–gross oedema EVLP 14 165 516 270 190 496 Yes–BL No Yes EVLP 15 293 443 360 165 713 Yes–BL No Yes EVLP 16 527 322 335 180 N/A No–gross oedema Shaded donor lungs were not transplanted after EVLP. PaO2:FiO2: the ratio of partial pressure of arterial oxygen to fraction of inspired oxygen; EVLP: ex vivo lung perfusion; PGD: primary graft dysfunction at 72 h; SL: single lung transplant; BL: bilateral lung transplant; N/A: not applicable; BALF: bronchoalveolar lavage fluid. a Start of donor lung flush at procurement to start of pulmonary artery (PA) perfusion on EVLP circuit. b Start of PA perfusion to disconnection from EVLP circuit. c Start of donor lung flush at procurement to reperfusion in recipient. d Death due to H1N1 infection 11 months post-transplant. Ex vivo lung perfusion outcomes The median partial pressure of oxygen on 100% inspired oxygen (PaO2:FiO2) at the end of EVLP was 516 (478–523) mmHg for transplanted and 342 (322–392) mmHg for non-transplanted donor lungs (Mann–Whitney U = 4, P = 0.02). The median PaO2:FiO2 improvement after EVLP was 321 (186–362) mmHg in transplanted donor lungs (Wilcoxon Z = −2.37, P = 0.03). No improvement was seen in lungs that failed assessment; the median PaO2:FiO2 change was 20 (−42–94) mmHg, (Wilcoxon Z = −0.18, P = 0.91) (Fig. 1). Figure 1: Physiological characteristics of transplanted and non-transplanted EVLP donor lungs. Box and whisker plots with boxes showing medians with interquartile range and whiskers at max and min. **P<0.05. EVLP; start time: point when the lung had been successfully rewarmed to 37 °C in the beginning of the EVLP; end time: point before start of cooling at the end of the EVLP. Of the nine lungs that failed EVLP reconditioning, six displayed worsening pulmonary oedema and two showed signs of persistent hyperinflation and raised PVR, suggesting diffuse injury and a degree of intrinsic emphysema. This suggestion was later verified by a pathologist from pulmonary biopsies. Lastly, one set of lungs was turned down due to heavy microbial contamination with multiple gram-negative organisms identified in the tracheal aspirate. Transplant outcomes In the seven donor lungs that were transplanted, perfusion and ventilation parameters remained stable or improved during assessment. The data on changes in lung compliance, airway resistance, PVR, perfusate flow, and ventilation and perfusion pressures are shown in Fig. 1. None of the recipients required extracorporeal life support. Two patients (29%) suffered from PGD (Grade 3) at 72 h post-transplant. However, both patients responded well to standard management. All seven recipients survived to hospital discharge. One recipient died of severe pneumonia at 11 months post-transplant due to an influenza infection unrelated to the EVLP procedure. The other six (86%) remain alive and well >2 years after transplant (Table 2). Analysis of perfusate and bronchoalveolar lavage fluid Assessed lungs were divided into two groups based on decision after EVLP: transplanted (n = 7) or non-transplanted (n = 9). Decision on transplant suitability was based on strict criteria summarized in Supplementary Table S2. Samples were analysed retrospectively for protein expressions. For an initial general assessment of tissue injury, we measured LDH levels. LDH was found in substantially lower levels in perfusate from lungs meeting criteria for transplant (T) after EVLP compared to those that did not meet the criteria for transplant (NT) lungs. Suitability for transplant was evident after 2 h of perfusion: mean (T) = 0.150 U (SD 0.030, n = 14) and mean (NT) = 0.223 U (SD 0.082, n = 18) [t(30) = 2.77, P = 0.03] (Fig. 2). The LDH perfusate levels remained significantly different when assessed over all sample time points: mean (T) = 0.149 U (SD 0.033, n = 24) and mean (NT) = 0.259 U (SD 0.088, n = 40) [t(62) = 5.45, P < 0.001]. No difference in LDH levels was seen in BALF samples collected pre- and post-perfusion. Figure 2: Protein expressions during EVLP in perfusate from transplanted and non-transplanted donor lungs. Protein levels expressed as medians with interquartile ranges. **P<0.05. EVLP: ex vivo lung perfusion; LDH: lactate dehydrogenase; IL: interleukin; TNF-α: tumour necrosis factor alpha; HMGB-1: high-mobility group box-1; Time: hours of ex vivo lung perfusion after start of pulmonary artery perfusion. To assess the integrity of the pulmonary vascular compartment, we measured Syndecan-1 levels as a marker of endothelial glycocaylx disruption. Syndecan-1 levels were significantly lower in perfusate samples from transplanted donor lungs: mean (T) = 111 (pg/ml) (SD 98, n = 24) compared to mean (NT) = 281 (pg/ml) (SD 219, n = 40) [t(62) = 4.14, P < 0.001] (Fig. 3). Figure 3: Protein expressions in perfusate and BALF from transplanted and non-transplanted EVLP donor lungs. Interleaved scatter plots with molecular marker levels expressed in (pg/ml) and lines representing means. *P<0.1. **P<0.05. ***P<0.001. BALF: bronchoalveolar lavage fluid; EVLP: ex vivo lung perfusion; IL: interleukin; TNF-α: tumour necrosis factor alpha; LDH: lactate dehydrogenase; HMGB-1: high-mobility group box-1. We noted a similarly consistent pattern towards lower release of proinflammatory cytokines into the circulating perfusate from transplanted compared to non-transplanted lungs. The average level of IL-8 was nearly eight times higher in perfusate samples from non-transplanted lungs: mean (T) = 165 (pg/ml) (SD 155, n = 24) compared to mean (NT) = 1310 (pg/ml) (SD 1510, n = 35) [t(57) = 2.27, P = 0.06]. Levels of the anti-inflammatory cytokine IL-10 increased noticeably over the first 2 h of perfusion in the transplanted group, with high initial fold changes in both perfusate and BALF (Fig. 4). Perfusate levels then diminished and were less than half those in non-transplanted lungs when measured over the full assessment: mean (T) = 9 (pg/ml) (SD 11, n = 24) compared to mean (NT) = 23 (pg/ml) (SD 32, n = 40) [t(57) = 2.02, P = 0.07]. No difference in levels of IL-10 in BALF was seen between the groups. Figure 4: Protein median fold changes in perfusate and BALF from EVLP donor lungs. Bar graphs showing median fold changes with interquartile range. BALF: bronchoalveolar lavage fluid; EVLP: ex vivo lung perfusion; IL: interleukin; TNF-α: tumour necrosis factor alpha; LDH: lactate dehydrogenase; HMGB-1: high-mobility group box-1. A panel of three damage-associated molecular patterns, IL-33, HMGB-1 and S100A9, was used to assess the extent of cellular injury in the donor lung. Lower levels of IL-33 and HMGB-1 were found in perfusate from transplanted donor lungs. The average IL-33 levels in perfusate were mean (T) = 3 (pg/ml) (SD 5, n = 24) and mean (NT) = 9 (pg/ml) (SD 10, n = 40) [t(62) = 2.93, P = 0.02]. HMGB-1 was highly expressed in both perfusate and BALF. The HMGB-1 levels in perfusate were mean (T) = 8318 (pg/ml) (SD 4474, n = 24) and mean (NT) = 10 545 (pg/ml) (SD 4,313, n = 40) [t(62) = 2.10, P = 0.07]. No difference in lavage fluid levels of IL-33 or HMGB-1 was seen between the groups. We noted a consistent pattern towards increasing perfusate protein levels and a more pronounced separation between transplanted and non-transplanted lungs over time of perfusion (Fig. 2). This pattern was most noticeable for the investigated inflammatory cytokines. The only protein marker not showing this pattern was S100A9, which appeared lower in the non-transplanted group. A multivariate analysis model was fitted to assess the predictive value of combining two or more protein markers. The optimal model was established by leave-one-out cross-validation and is shown in Fig. 5 as a scatterplot of IL-1β and IL-8 perfusate levels from the 16 assessments. The shaded prediction region was derived from the regression analysis with covariates IL-1β, IL-8 and (IL-8)2 after 2 h of perfusion. The model suggests that the closer to the centre of the area, the higher was the probability of a donor lung being found suitable for transplant after EVLP assessment. This early IL-1β–IL-8 signature remained equally as robust when applied to the full set of samples (Fig. 5B). An ROC curve showing the potential benefit of the model in predicting EVLP outcome compared to random is shown in Fig. 5C, with a sensitivity of 0.81 and specificity of 0.92 at 2 h of perfusion. Figure 5: (A) Early IL-1β-IL-8 signature in EVLP perfusate after 2 h of perfusion. (B) IL-1β-IL-8 signature in EVLP perfusate after full assessment. (C) ROC curve. (A) and (B) Scatter plots of IL-1β and IL-8 levels in perfusate samples from transplanted and non-transplanted EVLP donor lungs. The shaded area is the prediction region from the optimal logistic regression model with covariates IL-1β, IL-8 and (IL-8)2. (C) Receiver operating characteristic (ROC) curve of the model as a predictor of EVLP outcome. Calculated sensitivity (y axis) is plotted against the 1-specificity formula (x axis) of the logistic regression function 2 h into perfusion and after full assessment. EVLP: ex vivo lung perfusion; IL: interleukin. The modelled proteins were lastly assessed for correlations to post-transplant outcome measures. Donor IL-1β levels in the perfusate at time of transplant decision showed a significant negative correlation to recipient PaO2:FiO2 24 h post-transplant (r = −093, P = 0.03) (Fig. 6). No other correlations remained significant after corrections for multiple testing in this cohort. Figure 6: Association between donor IL-1β level in EVLP perfusate at the time of transplant decision and recipient P/F ratio 24 h post lung transplantation. Pearson correlation plot. **P<0.05. EVLP: ex vivo lung perfusion; IL: interleukin; PaO2:FiO2: partial pressure of oxygen on 100% inspired oxygen. DISCUSSION In this proof-of-concept study, the feasibility of using inflammation and tissue injury markers in both perfusate and BALF from human donor lungs undergoing EVLP to validate the decision to use or decline them for clinical transplant was evaluated. All 16 donor lungs in this study were assessed consecutively with intent for transplant following a strict protocol with predefined criteria for EVLP and transplant suitability to reduce the risk of bias in decision making. All protein analyses were done retrospectively and had no impact on the transplant decision. Our results demonstrate a difference in protein expression in perfusate from lungs that were transplanted compared with those that were declined. This difference became more apparent for each hour of perfusion, and BALF samples showed a comparable but weaker signal from the airway compartment. The difference in donor lungs was effectively demonstrated by a logistic regression model combining inflammatory cytokines IL-1β and IL-8 after 2 h of perfusion. The model revealed a clear separation of the transplanted lungs from those declined. Importantly, the model remained equally as robust when applied to perfusate samples from all time points. Furthermore, perfusate IL-1β concentrations in transplanted lungs demonstrated a clear negative correlation to early recipient oxygenation post-transplant. Even though our results are noteworthy, overinterpretation of this regression model should be avoided at this stage. Further investigations are required to assess its biological plausibility. A few declined lungs displayed a very low IL-1β/IL-8 signal—below that of lungs deemed reconditioned and transplanted with good outcomes after the EVLP assessment. If this finding indicates that those lungs were wrongfully declined and could safely have been transplanted, that EVLP-assessed lungs deemed suitable for transplant truly express an intermediate level of inflammation, or that this is a signal specific to this relatively small cohort needs further assessment in follow-up studies. The measurement of LDH levels in perfusate had the potential to significantly predict EVLP outcome as early as 2 h into perfusion. Although real-time ligand binding assays for inflammatory cytokines and other tissue injury markers may soon become accessible [22], LDH has the benefit of being a widely recognized marker of tissue injury in serum and BALF and rapidly available as a point-of-care test [23, 24]. The Brisbane lung transplant group have described measures of endothelial dysfunction as potentially useful in EVLP [25, 26]. Our results strengthen that belief. Syndecan-1, released by disruption of the endothelial glycocalyx, may be a discriminatory marker in EVLP perfusate and should be further evaluated. Increasing levels of inflammatory cytokines in perfusate and tissue have been shown during experimental lung perfusion in porcine models and a small preclinical human lung study, which might reflect procedure-mediated inflammation [27–29]. Our findings during full clinical EVLP support recent observations made by Machuca and colleagues [30] from Toronto and demonstrate the feasibility of identifying biomarkers that might improve EVLP assessment. IL-8, proposed as one of the best markers of EVLP transplant outcome in the Machuca study, has consistently shown potential in previous studies of donor lung injury from our group [12, 13]. The potential of IL-8 to discriminate successful EVLP in our study was enhanced by combining it with IL-1β. Our observations add significantly to previous observations because we have included an assessment of BALF inflammatory and tissue injury markers compared to EVLP perfusate measures. Even though BALF samples in this study did not add any apparent clinical value to perfusate sampling, we feel it is too early to draw any firm conclusions, and we will continue to assess a broad spectrum of markers and lung compartment samples in future studies. In addition, because protein levels are measured in a fixed volume of 2 litres of perfusate solution, we corrected protein levels for the size of lung tissue perfused. This potential confounder has been disregarded in previous studies. The pTLC of the donor lung ranged from 4.31 to 7.54 litres among the double lungs perfused, and the smallest single lung had a pTLC of only 2.70 litres. Unadjusted marker levels are therefore likely to differ widely between donor lungs regardless of the degree of lung injury present. A limitation of our study is its relatively small sample size, with 16 human EVLP assessments included, yet this number is sufficient to demonstrate the feasibility of using perfusate markers to discriminate successful EVLP. Outcomes were almost universally good, and this study does not identify predictors of early lung injury. Larger studies, with a cohort of patients doing less well, are needed to address that particular question. Caution is, however, required in overinterpretation of moderate to severe PGD (Grade 2–3) after transplant as a reason for donor lungs to be declined after EVLP, because many recipients with PGD 2–3 will recover and have satisfactory early outcomes. Truly marginal donor lungs subjected to EVLP for reconditioning purposes form by nature a subpopulation that is likely to have a higher inflammatory burden than standard donor lungs. We believe that biomarker profiles of these lungs could help support the clinical decision to use or decline the donor lungs after EVLP and allow continued safe transplant activity. Furthermore, this approach offers the potential to attenuate the inflammatory response or encourage recovery of vascular integrity in the donor lung before implantation in a stratified medicine approach, which may improve early outcomes after lung transplant and help to safely maximize lung use from the existing donor pool. The results of this study need further evaluation in a larger validation cohort of lungs exposed to ex vivo perfusion but demonstrate the feasibility of identifying an early perfusate signature potentially predictive of successful EVLP reconditioning. SUPPLEMENTARY MATERIAL Supplementary material (Tables 1 and 2) is available at EJCTS online. Fundi ng This work was also partly funded by the National Institute for Health Research Blood and Transplant Research Unit (NIHR BTRU) in Organ Donation and Transplantation at the University of Cambridge in collaboration with Newcastle University and in partnership with NHS Blood and Transplant (NHSBT). The views expressed are those of the author(s) and not necessarily those of the NHS, the NIHR, the Department of Health or NHSBT. The funding organizations had no role in the collection of data, its analysis or interpretation, and had no right to approve or disapprove publication of the finished manuscript. This study was supported by a research grant awarded by the United Kingdom Cystic Fibrosis Trust as a result of a donation made by the Robert Luff Foundation for research on lung transplants. Conflict of interest: none declared. Supplementary Material Supplementary Data Click here for additional data file.

Document structure show

article-title	Profiling inflammation and tissue injury markers in perfusate and bronchoalveolar lavage fluid during human ex vivo lung perfusion
abstract	Abstract OBJECTIVES: Availability of donor lungs suitable for transplant falls short of current demand and contributes to waiting list mortality. Ex vivo lung perfusion (EVLP) offers the opportunity to objectively assess and recondition organs unsuitable for immediate transplant. Identifying robust biomarkers that can stratify donor lungs during EVLP to use or non-use or for specific interventions could further improve its clinical impact. METHODS: In this pilot study, 16 consecutive donor lungs unsuitable for immediate transplant were assessed by EVLP. Key inflammatory mediators and tissue injury markers were measured in serial perfusate samples collected hourly and in bronchoalveolar lavage fluid (BALF) collected before and after EVLP. Levels were compared between donor lungs that met criteria for transplant and those that did not. RESULTS: Seven of the 16 donor lungs (44%) improved during EVLP and were transplanted with uniformly good outcomes. Tissue and vascular injury markers lactate dehydrogenase, HMGB-1 and Syndecan-1 were significantly lower in perfusate from transplanted lungs. A model combining IL-1β and IL-8 concentrations in perfusate could predict final EVLP outcome after 2 h assessment. In addition, perfusate IL-1β concentrations showed an inverse correlation to recipient oxygenation 24 h post-transplant. CONCLUSIONS: This study confirms the feasibility of using inflammation and tissue injury markers in perfusate and BALF to identify donor lungs most likely to improve for successful transplant during clinical EVLP. These results support examining this issue in a larger study.
title	Abstract
p	OBJECTIVES: Availability of donor lungs suitable for transplant falls short of current demand and contributes to waiting list mortality. Ex vivo lung perfusion (EVLP) offers the opportunity to objectively assess and recondition organs unsuitable for immediate transplant. Identifying robust biomarkers that can stratify donor lungs during EVLP to use or non-use or for specific interventions could further improve its clinical impact.
p	METHODS: In this pilot study, 16 consecutive donor lungs unsuitable for immediate transplant were assessed by EVLP. Key inflammatory mediators and tissue injury markers were measured in serial perfusate samples collected hourly and in bronchoalveolar lavage fluid (BALF) collected before and after EVLP. Levels were compared between donor lungs that met criteria for transplant and those that did not.
p	RESULTS: Seven of the 16 donor lungs (44%) improved during EVLP and were transplanted with uniformly good outcomes. Tissue and vascular injury markers lactate dehydrogenase, HMGB-1 and Syndecan-1 were significantly lower in perfusate from transplanted lungs. A model combining IL-1β and IL-8 concentrations in perfusate could predict final EVLP outcome after 2 h assessment. In addition, perfusate IL-1β concentrations showed an inverse correlation to recipient oxygenation 24 h post-transplant.
p	CONCLUSIONS: This study confirms the feasibility of using inflammation and tissue injury markers in perfusate and BALF to identify donor lungs most likely to improve for successful transplant during clinical EVLP. These results support examining this issue in a larger study.
body	INTRODUCTION Normothermic ex vivo lung perfusion (EVLP) has emerged as a promising technique to expand the donor pool by assessing and reconditioning donor lungs previously considered unsuitable for transplant [1]. The lung is highly susceptible to injury in the critical care environment, and in the hours or days leading up to the donor’s demise it may be exposed to the sequelae of brain-stem death together with infection, aspiration, barotrauma, fluid overload or multiple transfusions [2–4]. Because the extent of lung injury is difficult to assess at the time of organ procurement, donor acceptance criteria are therefore conservative and poor discriminators of injury and infection in the donor lung [5]. The increased use of extended-criteria donors may further elevate the risk of primary graft dysfunction (PGD) and other more severe postoperative complications [6–8]. The use of ex vivo reconditioned donor lungs is steadily growing and now accounts for about 20% of the activity in some established centres [9–11]. The decision to accept organs for transplant after EVLP has, however, been based on the same questionable physiological measures and organ appearance used during standard procurement. Reported discard rates of 10–60% of perfused lungs suggest that some donor lungs may be inappropriately used or declined for transplant after EVLP [1]. In donor lungs transplanted without exvivo evaluation, there is an established relationship between their inflammatory burden and early outcome [12–15]. However, this phenomenon has not been as extensively investigated during EVLP. In this proof-of-concept study, we evaluated a panel of inflammatory mediators and tissue injury-associated proteins in both perfusate and bronchoalveolar lavage fluid (BALF) from human donor lungs undergoing clinical EVLP with intent for transplant. The panel was based on our own previous work and on available studies of biological markers in standard lung transplant and preclinical and clinical observations during EVLP. The aim was to assess the potential for specific protein markers in perfusate and BALF to distinguish which donor lungs, initially deemed unsuitable for immediate transplantation are most likely to successfully recondition during EVLP and thereby provide a basis for further investigations in a larger validation cohort. MATERIALS AND METHODS Study design We conducted a prospective observational study investigating protein expression in donor lungs exposed to a standardized EVLP protocol. Approval was granted by our local research ethics committee, and informed consent for research was obtained from donor families and lung transplant recipients (REC 09/H0905/10). Ex vivo lung perfusion protocol Adult donor lungs deemed unsuitable for lung transplant by all five UK lung transplant centres but meeting strict EVLP criteria were included in the study (Supplementary Table S1), the results of which were previously published by our group [16]. Lungs were procured in a routine fashion and transported to our institution. The EVLP assessment followed a standardized acellular protocol using the Toronto technique with a closed left atrium and reduced perfusate flow, previously described in detail [3, 16]. Transplant suitability was assessed hourly during perfusion. Lungs meeting transplant criteria (Supplementary Table S2) at two consecutive time points were cooled and transplanted. Lungs deemed to have futile prospects for improvement were taken off the circuit and discarded. Two transplanted and five non-transplanted lungs were perfused ≥5h before a transplant decision was made. Sample collection and processing A research BAL was performed for all lungs by wedging an adult bronchoscope in a subsegmental bronchus of the right or left lower lobe [17]. Saline (40 ml) was instilled through the suction channel followed by gentle aspiration and sample collection prior to commencing ventilation at the beginning of EVLP. This process was repeated in the same lobe but in a different subsegmental bronchus before disconnecting the ventilation at the end of perfusion. In addition, hourly perfusate samples of 2.5 ml were collected until the assessment was stopped. Protein expression analysis All protein expressions measured in perfusate were adjusted to the predicted total lung capacity (pTLC) of the donor as an estimate of perfused donor lung volume and were reported as corrected perfusate concentrations (pg/ml). The pTLC was calculated in a routine fashion based on donor gender and height [18]. If one lung was deemed unusable due to severe consolidation or extensive contusion on inspection, or if the intended recipient required a single lung transplant on a specific side, only one lung was procured. For single-lung perfusions, the pTLC was adjusted to a factor of 0.55 for right lung and 0.45 for left lung perfusion [19]. Lactate dehydrogenase assay Lactate dehydrogenase (LDH) levels were measured in perfusate and BALF with a colorimetric LDH cytotoxicity assay kit according to manufacturer’s instructions and reported as arbitrary units (U) (Thermo Fisher Scientific Inc., Rockford, IL, USA). Multiplex inflammatory cytokine array Interleukin (IL)-1β, IL-6, IL-8, TNF-α, and IL-10 were analysed with an MSD Multi-Array® (Meso Scale Diagnostics, LLC, Rockville, MD, USA). The assay was performed according to the manufacturer’s instructions. Technical issues prevented reading of perfusate samples from donor lung EVLP03; therefore, these samples were excluded from the analysis. Enzyme-linked immunosorbent assays Syndecan-1, IL-33, S100A9 (all R&D Systems, Inc., Minneapolis, MN, USA) and high-mobility group box-1 (HMGB-1) (Shino-Test Corporation, Kanagawa, Japan) were measured with commercially available ELISA kits according to the manufacturers’ instructions. Statistical analysis Donor characteristics and physiological parameters are expressed as medians with interquartile ranges and were compared between transplanted and non-transplanted lungs using Mann–Whitney U tests. Paired samples, start-end of perfusion, were compared with Wilcoxon signed-ranks tests. Log protein expressions were compared between transplanted and non-transplanted lungs with multiple t-tests. Correlations between IL-8 and IL-1β expressions and seven post-transplant outcomes (PaO2:FiO2 24-h post-transplant; PGD3 at 72 h post-transplant; ventilation time; intensive care unit stay hospital stay‘; percent of predicted FEV1 at 6 months post-transplant; and percent of predicted FVC at 6 months post-transplant) were analysed by Pearson’s correlation tests. The data were transformed back into non-logged values for reporting as mean (T) for transplanted and mean (NT) for non-transplanted lungs with standard deviations (SD). Multiple testing of donor parameters, protein analyses, and correlations was corrected using the Benjamini-Hochberg false discovery rate (FDR) controlling procedure [20]. Because this was a feasibility study aiming to identify potential markers for further investigation in a validation cohort, an FDR corrected P-value <0.1 was deemed significant. Corrected P-values are reported. A multiple logistic regression model was fitted using the 11 log transformed protein covariates and their squared counterparts as independent variables and the EVLP outcome as the dependent variable. The optimal model was established by leave-one-out cross-validation using the software package R from R Core Team (2014) (R Foundation for Statistical Computing, Vienna, Austria) [21]. RESULTS Study group and donor characteristics Sixteen consecutive clinical EVLP assessments of lungs deemed unacceptable for standard transplant were included in the study (Table 1). Seven (44%) were accepted for transplant after EVLP assessment and were implanted as four bilateral and three single-lung procedures. Nine (56%) failed to achieve transplant criteria and were excluded. The characteristics of the donor groups of transplanted and non-transplanted lungs were not significantly different and are shown in Tables 1 and 2. Table 1: Donor characteristics Donor no. Age/ gender Smoker Donor cause of death Donor type TLC (L) Micro on offer Radiographic infiltrates Secretions EVLP indication EVLP 01 18 m No Diabetic coma DL DBD 7.30 Nil Bilateral Nil PaO2 <300 mmHg EVLP 02 51 f Yes ICH DL DBD 4.31 Nil Clear Nil PaO2 <300 mmHg EVLP 03 33 m Yes ICH DL DBD 7.14 Nil Left diffuse Purulent PaO2 <300 mmHg EVLP 04 15 f No Hanging DL DCD 5.10 Nil Bilateral Nil PaO2 <300 mmHg EVLP 05 52 m No ICH DL DBD 5.78 Nil Clear Mucopurulent PaO2 <300 mmHg EVLP 06 57 m Yes ICH DL DBD 7.54 Nil Right basal Mucopurulent oedema EVLP 07 52 f Yes ICH SL DBD 2.70 Nil Bibasal Purulent PaO2 <300 mmHg EVLP 08 20 m Yes Hanging SL DBD 3.18 Nil Right basal Nil PV PaO2 <225 mmHg + oedema EVLP 09 36 m No Brain tumour SL DBD 3.11 Yesa Left diffuse Nil PaO2 <300 mmHg EVLP 10 47 m No HBI DL DBD 6.74 Nil Clear Nil PaO2 <300 mmHg EVLP 11 45 f No TBI DL DBD 6.18 Nil Clear Nil PV PaO2 <225 mmHg + mild contusion RLL EVLP 12 56 m No ICH DL DBD 6.50 Nil Clear Purulent PaO2 <300 mmHg EVLP 13 18 f Yes ICH DL DCD 5.03 Nil Bilateral Mucopurulent oedema EVLP 14 53 m Yes ICH DL DBD 7.06 Nil Clear Purulent PaO2 <300 mmHg EVLP 15 48 f No ICH DL DBD 4.57 Nil Bibasal Purulent PaO2 <300 mmHg EVLP 16 38 m Yes Hanging DL DCD 6.98 Nil Left basal Nil oedema Shaded donor lungs were not transplanted after EVLP . BALF: bronchoalveolar lavage fluid; EVLP : ex vivo lung perfusion; ICH: intracranial haemorrhage; HBI: hypoxic brain injury secondary to variceal bleed; TBI: traumatic brain injury following a road traffic accident; DL: double lung perfusion; SL: single lung perfusion; DBD: organ donation after brain death; DCD: organ donation after circulatory death; TLC: predicted total lung capacity of perfused lung; PaO2: partial pressure of arterial oxygen; PV: selective pulmonary vein blood gas analyses; RLL: right lower lobe. a Sputum culture with moderate growth of Staphylococcus aureus, Haemophilus influenzae, and Streptococcus pneumoniae. Table 2: EVLP and transplant outcomes Donor no. Optimized donor PaO2:FiO2 (mmHg) PaO2:FiO2 after EVLP (mmHg) Ischaemic time (min)a EVLP time (min)b Total time ex vivo (min)c Transplant PGD 3 at 72 h 90-day survival EVLP 01 149 521 345 290 1020 Yes–BL Yes Yes EVLP 02 248 342 360 360 N/A No–emphysema EVLP 03 209 167 330 270 N/A No–gross oedema EVLP 04 222 426 450 360 870 Yes–SL No Yes EVLP 05 167 440 320 330 N/A No–heavy gram-stain in BALF EVLP 06 357 525 395 240 582 Yes–SL No Yes EVLP 07 178 557 315 255 1048 Yes–SL Yes Yes EVLP 08 372 392 340 240 N/A No–emphysema EVLP 09 293 330 430 300 N/A No–gross oedema EVLP 10 171 291 305 300 N/A No–gross oedema EVLP 11 526 452 340 330 N/A No–gross oedema EVLP 12 191 512 395 220 785 Yes–BL No Yesd EVLP 13 360 352 345 180 N/A No–gross oedema EVLP 14 165 516 270 190 496 Yes–BL No Yes EVLP 15 293 443 360 165 713 Yes–BL No Yes EVLP 16 527 322 335 180 N/A No–gross oedema Shaded donor lungs were not transplanted after EVLP. PaO2:FiO2: the ratio of partial pressure of arterial oxygen to fraction of inspired oxygen; EVLP: ex vivo lung perfusion; PGD: primary graft dysfunction at 72 h; SL: single lung transplant; BL: bilateral lung transplant; N/A: not applicable; BALF: bronchoalveolar lavage fluid. a Start of donor lung flush at procurement to start of pulmonary artery (PA) perfusion on EVLP circuit. b Start of PA perfusion to disconnection from EVLP circuit. c Start of donor lung flush at procurement to reperfusion in recipient. d Death due to H1N1 infection 11 months post-transplant. Ex vivo lung perfusion outcomes The median partial pressure of oxygen on 100% inspired oxygen (PaO2:FiO2) at the end of EVLP was 516 (478–523) mmHg for transplanted and 342 (322–392) mmHg for non-transplanted donor lungs (Mann–Whitney U = 4, P = 0.02). The median PaO2:FiO2 improvement after EVLP was 321 (186–362) mmHg in transplanted donor lungs (Wilcoxon Z = −2.37, P = 0.03). No improvement was seen in lungs that failed assessment; the median PaO2:FiO2 change was 20 (−42–94) mmHg, (Wilcoxon Z = −0.18, P = 0.91) (Fig. 1). Figure 1: Physiological characteristics of transplanted and non-transplanted EVLP donor lungs. Box and whisker plots with boxes showing medians with interquartile range and whiskers at max and min. P<0.05. EVLP; start time: point when the lung had been successfully rewarmed to 37 °C in the beginning of the EVLP; end time: point before start of cooling at the end of the EVLP. Of the nine lungs that failed EVLP reconditioning, six displayed worsening pulmonary oedema and two showed signs of persistent hyperinflation and raised PVR, suggesting diffuse injury and a degree of intrinsic emphysema. This suggestion was later verified by a pathologist from pulmonary biopsies. Lastly, one set of lungs was turned down due to heavy microbial contamination with multiple gram-negative organisms identified in the tracheal aspirate. Transplant outcomes In the seven donor lungs that were transplanted, perfusion and ventilation parameters remained stable or improved during assessment. The data on changes in lung compliance, airway resistance, PVR, perfusate flow, and ventilation and perfusion pressures are shown in Fig. 1. None of the recipients required extracorporeal life support. Two patients (29%) suffered from PGD (Grade 3) at 72 h post-transplant. However, both patients responded well to standard management. All seven recipients survived to hospital discharge. One recipient died of severe pneumonia at 11 months post-transplant due to an influenza infection unrelated to the EVLP procedure. The other six (86%) remain alive and well >2 years after transplant (Table 2). Analysis of perfusate and bronchoalveolar lavage fluid Assessed lungs were divided into two groups based on decision after EVLP: transplanted (n = 7) or non-transplanted (n = 9). Decision on transplant suitability was based on strict criteria summarized in Supplementary Table S2. Samples were analysed retrospectively for protein expressions. For an initial general assessment of tissue injury, we measured LDH levels. LDH was found in substantially lower levels in perfusate from lungs meeting criteria for transplant (T) after EVLP compared to those that did not meet the criteria for transplant (NT) lungs. Suitability for transplant was evident after 2 h of perfusion: mean (T) = 0.150 U (SD 0.030, n = 14) and mean (NT) = 0.223 U (SD 0.082, n = 18) [t(30) = 2.77, P = 0.03] (Fig. 2). The LDH perfusate levels remained significantly different when assessed over all sample time points: mean (T) = 0.149 U (SD 0.033, n = 24) and mean (NT) = 0.259 U (SD 0.088, n = 40) [t(62) = 5.45, P < 0.001]. No difference in LDH levels was seen in BALF samples collected pre- and post-perfusion. Figure 2: Protein expressions during EVLP in perfusate from transplanted and non-transplanted donor lungs. Protein levels expressed as medians with interquartile ranges. P<0.05. EVLP: ex vivo lung perfusion; LDH: lactate dehydrogenase; IL: interleukin; TNF-α: tumour necrosis factor alpha; HMGB-1: high-mobility group box-1; Time: hours of ex vivo lung perfusion after start of pulmonary artery perfusion. To assess the integrity of the pulmonary vascular compartment, we measured Syndecan-1 levels as a marker of endothelial glycocaylx disruption. Syndecan-1 levels were significantly lower in perfusate samples from transplanted donor lungs: mean (T) = 111 (pg/ml) (SD 98, n = 24) compared to mean (NT) = 281 (pg/ml) (SD 219, n = 40) [t(62) = 4.14, P < 0.001] (Fig. 3). Figure 3: Protein expressions in perfusate and BALF from transplanted and non-transplanted EVLP donor lungs. Interleaved scatter plots with molecular marker levels expressed in (pg/ml) and lines representing means. P<0.1. P<0.05. P<0.001. BALF: bronchoalveolar lavage fluid; EVLP: ex vivo lung perfusion; IL: interleukin; TNF-α: tumour necrosis factor alpha; LDH: lactate dehydrogenase; HMGB-1: high-mobility group box-1. We noted a similarly consistent pattern towards lower release of proinflammatory cytokines into the circulating perfusate from transplanted compared to non-transplanted lungs. The average level of IL-8 was nearly eight times higher in perfusate samples from non-transplanted lungs: mean (T) = 165 (pg/ml) (SD 155, n = 24) compared to mean (NT) = 1310 (pg/ml) (SD 1510, n = 35) [t(57) = 2.27, P = 0.06]. Levels of the anti-inflammatory cytokine IL-10 increased noticeably over the first 2 h of perfusion in the transplanted group, with high initial fold changes in both perfusate and BALF (Fig. 4). Perfusate levels then diminished and were less than half those in non-transplanted lungs when measured over the full assessment: mean (T) = 9 (pg/ml) (SD 11, n = 24) compared to mean (NT) = 23 (pg/ml) (SD 32, n = 40) [t(57) = 2.02, P = 0.07]. No difference in levels of IL-10 in BALF was seen between the groups. Figure 4: Protein median fold changes in perfusate and BALF from EVLP donor lungs. Bar graphs showing median fold changes with interquartile range. BALF: bronchoalveolar lavage fluid; EVLP: ex vivo lung perfusion; IL: interleukin; TNF-α: tumour necrosis factor alpha; LDH: lactate dehydrogenase; HMGB-1: high-mobility group box-1. A panel of three damage-associated molecular patterns, IL-33, HMGB-1 and S100A9, was used to assess the extent of cellular injury in the donor lung. Lower levels of IL-33 and HMGB-1 were found in perfusate from transplanted donor lungs. The average IL-33 levels in perfusate were mean (T) = 3 (pg/ml) (SD 5, n = 24) and mean (NT) = 9 (pg/ml) (SD 10, n = 40) [t(62) = 2.93, P = 0.02]. HMGB-1 was highly expressed in both perfusate and BALF. The HMGB-1 levels in perfusate were mean (T) = 8318 (pg/ml) (SD 4474, n = 24) and mean (NT) = 10 545 (pg/ml) (SD 4,313, n = 40) [t(62) = 2.10, P = 0.07]. No difference in lavage fluid levels of IL-33 or HMGB-1 was seen between the groups. We noted a consistent pattern towards increasing perfusate protein levels and a more pronounced separation between transplanted and non-transplanted lungs over time of perfusion (Fig. 2). This pattern was most noticeable for the investigated inflammatory cytokines. The only protein marker not showing this pattern was S100A9, which appeared lower in the non-transplanted group. A multivariate analysis model was fitted to assess the predictive value of combining two or more protein markers. The optimal model was established by leave-one-out cross-validation and is shown in Fig. 5 as a scatterplot of IL-1β and IL-8 perfusate levels from the 16 assessments. The shaded prediction region was derived from the regression analysis with covariates IL-1β, IL-8 and (IL-8)2 after 2 h of perfusion. The model suggests that the closer to the centre of the area, the higher was the probability of a donor lung being found suitable for transplant after EVLP assessment. This early IL-1β–IL-8 signature remained equally as robust when applied to the full set of samples (Fig. 5B). An ROC curve showing the potential benefit of the model in predicting EVLP outcome compared to random is shown in Fig. 5C, with a sensitivity of 0.81 and specificity of 0.92 at 2 h of perfusion. Figure 5: (A) Early IL-1β-IL-8 signature in EVLP perfusate after 2 h of perfusion. (B) IL-1β-IL-8 signature in EVLP perfusate after full assessment. (C) ROC curve. (A) and (B) Scatter plots of IL-1β and IL-8 levels in perfusate samples from transplanted and non-transplanted EVLP donor lungs. The shaded area is the prediction region from the optimal logistic regression model with covariates IL-1β, IL-8 and (IL-8)2. (C) Receiver operating characteristic (ROC) curve of the model as a predictor of EVLP outcome. Calculated sensitivity (y axis) is plotted against the 1-specificity formula (x axis) of the logistic regression function 2 h into perfusion and after full assessment. EVLP: ex vivo lung perfusion; IL: interleukin. The modelled proteins were lastly assessed for correlations to post-transplant outcome measures. Donor IL-1β levels in the perfusate at time of transplant decision showed a significant negative correlation to recipient PaO2:FiO2 24 h post-transplant (r = −093, P = 0.03) (Fig. 6). No other correlations remained significant after corrections for multiple testing in this cohort. Figure 6: Association between donor IL-1β level in EVLP perfusate at the time of transplant decision and recipient P/F ratio 24 h post lung transplantation. Pearson correlation plot. P<0.05. EVLP: ex vivo lung perfusion; IL: interleukin; PaO2:FiO2: partial pressure of oxygen on 100% inspired oxygen. DISCUSSION In this proof-of-concept study, the feasibility of using inflammation and tissue injury markers in both perfusate and BALF from human donor lungs undergoing EVLP to validate the decision to use or decline them for clinical transplant was evaluated. All 16 donor lungs in this study were assessed consecutively with intent for transplant following a strict protocol with predefined criteria for EVLP and transplant suitability to reduce the risk of bias in decision making. All protein analyses were done retrospectively and had no impact on the transplant decision. Our results demonstrate a difference in protein expression in perfusate from lungs that were transplanted compared with those that were declined. This difference became more apparent for each hour of perfusion, and BALF samples showed a comparable but weaker signal from the airway compartment. The difference in donor lungs was effectively demonstrated by a logistic regression model combining inflammatory cytokines IL-1β and IL-8 after 2 h of perfusion. The model revealed a clear separation of the transplanted lungs from those declined. Importantly, the model remained equally as robust when applied to perfusate samples from all time points. Furthermore, perfusate IL-1β concentrations in transplanted lungs demonstrated a clear negative correlation to early recipient oxygenation post-transplant. Even though our results are noteworthy, overinterpretation of this regression model should be avoided at this stage. Further investigations are required to assess its biological plausibility. A few declined lungs displayed a very low IL-1β/IL-8 signal—below that of lungs deemed reconditioned and transplanted with good outcomes after the EVLP assessment. If this finding indicates that those lungs were wrongfully declined and could safely have been transplanted, that EVLP-assessed lungs deemed suitable for transplant truly express an intermediate level of inflammation, or that this is a signal specific to this relatively small cohort needs further assessment in follow-up studies. The measurement of LDH levels in perfusate had the potential to significantly predict EVLP outcome as early as 2 h into perfusion. Although real-time ligand binding assays for inflammatory cytokines and other tissue injury markers may soon become accessible [22], LDH has the benefit of being a widely recognized marker of tissue injury in serum and BALF and rapidly available as a point-of-care test [23, 24]. The Brisbane lung transplant group have described measures of endothelial dysfunction as potentially useful in EVLP [25, 26]. Our results strengthen that belief. Syndecan-1, released by disruption of the endothelial glycocalyx, may be a discriminatory marker in EVLP perfusate and should be further evaluated. Increasing levels of inflammatory cytokines in perfusate and tissue have been shown during experimental lung perfusion in porcine models and a small preclinical human lung study, which might reflect procedure-mediated inflammation [27–29]. Our findings during full clinical EVLP support recent observations made by Machuca and colleagues [30] from Toronto and demonstrate the feasibility of identifying biomarkers that might improve EVLP assessment. IL-8, proposed as one of the best markers of EVLP transplant outcome in the Machuca study, has consistently shown potential in previous studies of donor lung injury from our group [12, 13]. The potential of IL-8 to discriminate successful EVLP in our study was enhanced by combining it with IL-1β. Our observations add significantly to previous observations because we have included an assessment of BALF inflammatory and tissue injury markers compared to EVLP perfusate measures. Even though BALF samples in this study did not add any apparent clinical value to perfusate sampling, we feel it is too early to draw any firm conclusions, and we will continue to assess a broad spectrum of markers and lung compartment samples in future studies. In addition, because protein levels are measured in a fixed volume of 2 litres of perfusate solution, we corrected protein levels for the size of lung tissue perfused. This potential confounder has been disregarded in previous studies. The pTLC of the donor lung ranged from 4.31 to 7.54 litres among the double lungs perfused, and the smallest single lung had a pTLC of only 2.70 litres. Unadjusted marker levels are therefore likely to differ widely between donor lungs regardless of the degree of lung injury present. A limitation of our study is its relatively small sample size, with 16 human EVLP assessments included, yet this number is sufficient to demonstrate the feasibility of using perfusate markers to discriminate successful EVLP. Outcomes were almost universally good, and this study does not identify predictors of early lung injury. Larger studies, with a cohort of patients doing less well, are needed to address that particular question. Caution is, however, required in overinterpretation of moderate to severe PGD (Grade 2–3) after transplant as a reason for donor lungs to be declined after EVLP, because many recipients with PGD 2–3 will recover and have satisfactory early outcomes. Truly marginal donor lungs subjected to EVLP for reconditioning purposes form by nature a subpopulation that is likely to have a higher inflammatory burden than standard donor lungs. We believe that biomarker profiles of these lungs could help support the clinical decision to use or decline the donor lungs after EVLP and allow continued safe transplant activity. Furthermore, this approach offers the potential to attenuate the inflammatory response or encourage recovery of vascular integrity in the donor lung before implantation in a stratified medicine approach, which may improve early outcomes after lung transplant and help to safely maximize lung use from the existing donor pool. The results of this study need further evaluation in a larger validation cohort of lungs exposed to ex vivo perfusion but demonstrate the feasibility of identifying an early perfusate signature potentially predictive of successful EVLP reconditioning. SUPPLEMENTARY MATERIAL Supplementary material (Tables 1 and 2) is available at EJCTS online. Fundi ng This work was also partly funded by the National Institute for Health Research Blood and Transplant Research Unit (NIHR BTRU) in Organ Donation and Transplantation at the University of Cambridge in collaboration with Newcastle University and in partnership with NHS Blood and Transplant (NHSBT). The views expressed are those of the author(s) and not necessarily those of the NHS, the NIHR, the Department of Health or NHSBT. The funding organizations had no role in the collection of data, its analysis or interpretation, and had no right to approve or disapprove publication of the finished manuscript. This study was supported by a research grant awarded by the United Kingdom Cystic Fibrosis Trust as a result of a donation made by the Robert Luff Foundation for research on lung transplants. Conflict of interest: none declared. Supplementary Material Supplementary Data Click here for additional data file.
sec	INTRODUCTION Normothermic ex vivo lung perfusion (EVLP) has emerged as a promising technique to expand the donor pool by assessing and reconditioning donor lungs previously considered unsuitable for transplant [1]. The lung is highly susceptible to injury in the critical care environment, and in the hours or days leading up to the donor’s demise it may be exposed to the sequelae of brain-stem death together with infection, aspiration, barotrauma, fluid overload or multiple transfusions [2–4]. Because the extent of lung injury is difficult to assess at the time of organ procurement, donor acceptance criteria are therefore conservative and poor discriminators of injury and infection in the donor lung [5]. The increased use of extended-criteria donors may further elevate the risk of primary graft dysfunction (PGD) and other more severe postoperative complications [6–8]. The use of ex vivo reconditioned donor lungs is steadily growing and now accounts for about 20% of the activity in some established centres [9–11]. The decision to accept organs for transplant after EVLP has, however, been based on the same questionable physiological measures and organ appearance used during standard procurement. Reported discard rates of 10–60% of perfused lungs suggest that some donor lungs may be inappropriately used or declined for transplant after EVLP [1]. In donor lungs transplanted without exvivo evaluation, there is an established relationship between their inflammatory burden and early outcome [12–15]. However, this phenomenon has not been as extensively investigated during EVLP. In this proof-of-concept study, we evaluated a panel of inflammatory mediators and tissue injury-associated proteins in both perfusate and bronchoalveolar lavage fluid (BALF) from human donor lungs undergoing clinical EVLP with intent for transplant. The panel was based on our own previous work and on available studies of biological markers in standard lung transplant and preclinical and clinical observations during EVLP. The aim was to assess the potential for specific protein markers in perfusate and BALF to distinguish which donor lungs, initially deemed unsuitable for immediate transplantation are most likely to successfully recondition during EVLP and thereby provide a basis for further investigations in a larger validation cohort.
title	INTRODUCTION
p	Normothermic ex vivo lung perfusion (EVLP) has emerged as a promising technique to expand the donor pool by assessing and reconditioning donor lungs previously considered unsuitable for transplant [1]. The lung is highly susceptible to injury in the critical care environment, and in the hours or days leading up to the donor’s demise it may be exposed to the sequelae of brain-stem death together with infection, aspiration, barotrauma, fluid overload or multiple transfusions [2–4]. Because the extent of lung injury is difficult to assess at the time of organ procurement, donor acceptance criteria are therefore conservative and poor discriminators of injury and infection in the donor lung [5]. The increased use of extended-criteria donors may further elevate the risk of primary graft dysfunction (PGD) and other more severe postoperative complications [6–8].
p	The use of ex vivo reconditioned donor lungs is steadily growing and now accounts for about 20% of the activity in some established centres [9–11]. The decision to accept organs for transplant after EVLP has, however, been based on the same questionable physiological measures and organ appearance used during standard procurement. Reported discard rates of 10–60% of perfused lungs suggest that some donor lungs may be inappropriately used or declined for transplant after EVLP [1].
p	In donor lungs transplanted without exvivo evaluation, there is an established relationship between their inflammatory burden and early outcome [12–15]. However, this phenomenon has not been as extensively investigated during EVLP.
p	In this proof-of-concept study, we evaluated a panel of inflammatory mediators and tissue injury-associated proteins in both perfusate and bronchoalveolar lavage fluid (BALF) from human donor lungs undergoing clinical EVLP with intent for transplant. The panel was based on our own previous work and on available studies of biological markers in standard lung transplant and preclinical and clinical observations during EVLP. The aim was to assess the potential for specific protein markers in perfusate and BALF to distinguish which donor lungs, initially deemed unsuitable for immediate transplantation are most likely to successfully recondition during EVLP and thereby provide a basis for further investigations in a larger validation cohort.
sec	MATERIALS AND METHODS Study design We conducted a prospective observational study investigating protein expression in donor lungs exposed to a standardized EVLP protocol. Approval was granted by our local research ethics committee, and informed consent for research was obtained from donor families and lung transplant recipients (REC 09/H0905/10). Ex vivo lung perfusion protocol Adult donor lungs deemed unsuitable for lung transplant by all five UK lung transplant centres but meeting strict EVLP criteria were included in the study (Supplementary Table S1), the results of which were previously published by our group [16]. Lungs were procured in a routine fashion and transported to our institution. The EVLP assessment followed a standardized acellular protocol using the Toronto technique with a closed left atrium and reduced perfusate flow, previously described in detail [3, 16]. Transplant suitability was assessed hourly during perfusion. Lungs meeting transplant criteria (Supplementary Table S2) at two consecutive time points were cooled and transplanted. Lungs deemed to have futile prospects for improvement were taken off the circuit and discarded. Two transplanted and five non-transplanted lungs were perfused ≥5h before a transplant decision was made. Sample collection and processing A research BAL was performed for all lungs by wedging an adult bronchoscope in a subsegmental bronchus of the right or left lower lobe [17]. Saline (40 ml) was instilled through the suction channel followed by gentle aspiration and sample collection prior to commencing ventilation at the beginning of EVLP. This process was repeated in the same lobe but in a different subsegmental bronchus before disconnecting the ventilation at the end of perfusion. In addition, hourly perfusate samples of 2.5 ml were collected until the assessment was stopped. Protein expression analysis All protein expressions measured in perfusate were adjusted to the predicted total lung capacity (pTLC) of the donor as an estimate of perfused donor lung volume and were reported as corrected perfusate concentrations (pg/ml). The pTLC was calculated in a routine fashion based on donor gender and height [18]. If one lung was deemed unusable due to severe consolidation or extensive contusion on inspection, or if the intended recipient required a single lung transplant on a specific side, only one lung was procured. For single-lung perfusions, the pTLC was adjusted to a factor of 0.55 for right lung and 0.45 for left lung perfusion [19]. Lactate dehydrogenase assay Lactate dehydrogenase (LDH) levels were measured in perfusate and BALF with a colorimetric LDH cytotoxicity assay kit according to manufacturer’s instructions and reported as arbitrary units (U) (Thermo Fisher Scientific Inc., Rockford, IL, USA). Multiplex inflammatory cytokine array Interleukin (IL)-1β, IL-6, IL-8, TNF-α, and IL-10 were analysed with an MSD Multi-Array® (Meso Scale Diagnostics, LLC, Rockville, MD, USA). The assay was performed according to the manufacturer’s instructions. Technical issues prevented reading of perfusate samples from donor lung EVLP03; therefore, these samples were excluded from the analysis. Enzyme-linked immunosorbent assays Syndecan-1, IL-33, S100A9 (all R&D Systems, Inc., Minneapolis, MN, USA) and high-mobility group box-1 (HMGB-1) (Shino-Test Corporation, Kanagawa, Japan) were measured with commercially available ELISA kits according to the manufacturers’ instructions. Statistical analysis Donor characteristics and physiological parameters are expressed as medians with interquartile ranges and were compared between transplanted and non-transplanted lungs using Mann–Whitney U tests. Paired samples, start-end of perfusion, were compared with Wilcoxon signed-ranks tests. Log protein expressions were compared between transplanted and non-transplanted lungs with multiple t-tests. Correlations between IL-8 and IL-1β expressions and seven post-transplant outcomes (PaO2:FiO2 24-h post-transplant; PGD3 at 72 h post-transplant; ventilation time; intensive care unit stay hospital stay‘; percent of predicted FEV1 at 6 months post-transplant; and percent of predicted FVC at 6 months post-transplant) were analysed by Pearson’s correlation tests. The data were transformed back into non-logged values for reporting as mean (T) for transplanted and mean (NT) for non-transplanted lungs with standard deviations (SD). Multiple testing of donor parameters, protein analyses, and correlations was corrected using the Benjamini-Hochberg false discovery rate (FDR) controlling procedure [20]. Because this was a feasibility study aiming to identify potential markers for further investigation in a validation cohort, an FDR corrected P-value <0.1 was deemed significant. Corrected P-values are reported. A multiple logistic regression model was fitted using the 11 log transformed protein covariates and their squared counterparts as independent variables and the EVLP outcome as the dependent variable. The optimal model was established by leave-one-out cross-validation using the software package R from R Core Team (2014) (R Foundation for Statistical Computing, Vienna, Austria) [21].
title	MATERIALS AND METHODS
sec	Study design We conducted a prospective observational study investigating protein expression in donor lungs exposed to a standardized EVLP protocol. Approval was granted by our local research ethics committee, and informed consent for research was obtained from donor families and lung transplant recipients (REC 09/H0905/10).
title	Study design
p	We conducted a prospective observational study investigating protein expression in donor lungs exposed to a standardized EVLP protocol. Approval was granted by our local research ethics committee, and informed consent for research was obtained from donor families and lung transplant recipients (REC 09/H0905/10).
sec	Ex vivo lung perfusion protocol Adult donor lungs deemed unsuitable for lung transplant by all five UK lung transplant centres but meeting strict EVLP criteria were included in the study (Supplementary Table S1), the results of which were previously published by our group [16]. Lungs were procured in a routine fashion and transported to our institution. The EVLP assessment followed a standardized acellular protocol using the Toronto technique with a closed left atrium and reduced perfusate flow, previously described in detail [3, 16]. Transplant suitability was assessed hourly during perfusion. Lungs meeting transplant criteria (Supplementary Table S2) at two consecutive time points were cooled and transplanted. Lungs deemed to have futile prospects for improvement were taken off the circuit and discarded. Two transplanted and five non-transplanted lungs were perfused ≥5h before a transplant decision was made. Sample collection and processing A research BAL was performed for all lungs by wedging an adult bronchoscope in a subsegmental bronchus of the right or left lower lobe [17]. Saline (40 ml) was instilled through the suction channel followed by gentle aspiration and sample collection prior to commencing ventilation at the beginning of EVLP. This process was repeated in the same lobe but in a different subsegmental bronchus before disconnecting the ventilation at the end of perfusion. In addition, hourly perfusate samples of 2.5 ml were collected until the assessment was stopped.
title	Ex vivo lung perfusion protocol
p	Adult donor lungs deemed unsuitable for lung transplant by all five UK lung transplant centres but meeting strict EVLP criteria were included in the study (Supplementary Table S1), the results of which were previously published by our group [16]. Lungs were procured in a routine fashion and transported to our institution. The EVLP assessment followed a standardized acellular protocol using the Toronto technique with a closed left atrium and reduced perfusate flow, previously described in detail [3, 16]. Transplant suitability was assessed hourly during perfusion. Lungs meeting transplant criteria (Supplementary Table S2) at two consecutive time points were cooled and transplanted. Lungs deemed to have futile prospects for improvement were taken off the circuit and discarded. Two transplanted and five non-transplanted lungs were perfused ≥5h before a transplant decision was made.
sec	Sample collection and processing A research BAL was performed for all lungs by wedging an adult bronchoscope in a subsegmental bronchus of the right or left lower lobe [17]. Saline (40 ml) was instilled through the suction channel followed by gentle aspiration and sample collection prior to commencing ventilation at the beginning of EVLP. This process was repeated in the same lobe but in a different subsegmental bronchus before disconnecting the ventilation at the end of perfusion. In addition, hourly perfusate samples of 2.5 ml were collected until the assessment was stopped.
title	Sample collection and processing
p	A research BAL was performed for all lungs by wedging an adult bronchoscope in a subsegmental bronchus of the right or left lower lobe [17]. Saline (40 ml) was instilled through the suction channel followed by gentle aspiration and sample collection prior to commencing ventilation at the beginning of EVLP. This process was repeated in the same lobe but in a different subsegmental bronchus before disconnecting the ventilation at the end of perfusion. In addition, hourly perfusate samples of 2.5 ml were collected until the assessment was stopped.
sec	Protein expression analysis All protein expressions measured in perfusate were adjusted to the predicted total lung capacity (pTLC) of the donor as an estimate of perfused donor lung volume and were reported as corrected perfusate concentrations (pg/ml). The pTLC was calculated in a routine fashion based on donor gender and height [18]. If one lung was deemed unusable due to severe consolidation or extensive contusion on inspection, or if the intended recipient required a single lung transplant on a specific side, only one lung was procured. For single-lung perfusions, the pTLC was adjusted to a factor of 0.55 for right lung and 0.45 for left lung perfusion [19].
title	Protein expression analysis
p	All protein expressions measured in perfusate were adjusted to the predicted total lung capacity (pTLC) of the donor as an estimate of perfused donor lung volume and were reported as corrected perfusate concentrations (pg/ml). The pTLC was calculated in a routine fashion based on donor gender and height [18]. If one lung was deemed unusable due to severe consolidation or extensive contusion on inspection, or if the intended recipient required a single lung transplant on a specific side, only one lung was procured. For single-lung perfusions, the pTLC was adjusted to a factor of 0.55 for right lung and 0.45 for left lung perfusion [19].
sec	Lactate dehydrogenase assay Lactate dehydrogenase (LDH) levels were measured in perfusate and BALF with a colorimetric LDH cytotoxicity assay kit according to manufacturer’s instructions and reported as arbitrary units (U) (Thermo Fisher Scientific Inc., Rockford, IL, USA).
title	Lactate dehydrogenase assay
p	Lactate dehydrogenase (LDH) levels were measured in perfusate and BALF with a colorimetric LDH cytotoxicity assay kit according to manufacturer’s instructions and reported as arbitrary units (U) (Thermo Fisher Scientific Inc., Rockford, IL, USA).
sec	Multiplex inflammatory cytokine array Interleukin (IL)-1β, IL-6, IL-8, TNF-α, and IL-10 were analysed with an MSD Multi-Array® (Meso Scale Diagnostics, LLC, Rockville, MD, USA). The assay was performed according to the manufacturer’s instructions. Technical issues prevented reading of perfusate samples from donor lung EVLP03; therefore, these samples were excluded from the analysis.
title	Multiplex inflammatory cytokine array
p	Interleukin (IL)-1β, IL-6, IL-8, TNF-α, and IL-10 were analysed with an MSD Multi-Array® (Meso Scale Diagnostics, LLC, Rockville, MD, USA). The assay was performed according to the manufacturer’s instructions. Technical issues prevented reading of perfusate samples from donor lung EVLP03; therefore, these samples were excluded from the analysis.
sec	Enzyme-linked immunosorbent assays Syndecan-1, IL-33, S100A9 (all R&D Systems, Inc., Minneapolis, MN, USA) and high-mobility group box-1 (HMGB-1) (Shino-Test Corporation, Kanagawa, Japan) were measured with commercially available ELISA kits according to the manufacturers’ instructions.
title	Enzyme-linked immunosorbent assays
p	Syndecan-1, IL-33, S100A9 (all R&D Systems, Inc., Minneapolis, MN, USA) and high-mobility group box-1 (HMGB-1) (Shino-Test Corporation, Kanagawa, Japan) were measured with commercially available ELISA kits according to the manufacturers’ instructions.
sec	Statistical analysis Donor characteristics and physiological parameters are expressed as medians with interquartile ranges and were compared between transplanted and non-transplanted lungs using Mann–Whitney U tests. Paired samples, start-end of perfusion, were compared with Wilcoxon signed-ranks tests. Log protein expressions were compared between transplanted and non-transplanted lungs with multiple t-tests. Correlations between IL-8 and IL-1β expressions and seven post-transplant outcomes (PaO2:FiO2 24-h post-transplant; PGD3 at 72 h post-transplant; ventilation time; intensive care unit stay hospital stay‘; percent of predicted FEV1 at 6 months post-transplant; and percent of predicted FVC at 6 months post-transplant) were analysed by Pearson’s correlation tests. The data were transformed back into non-logged values for reporting as mean (T) for transplanted and mean (NT) for non-transplanted lungs with standard deviations (SD). Multiple testing of donor parameters, protein analyses, and correlations was corrected using the Benjamini-Hochberg false discovery rate (FDR) controlling procedure [20]. Because this was a feasibility study aiming to identify potential markers for further investigation in a validation cohort, an FDR corrected P-value <0.1 was deemed significant. Corrected P-values are reported. A multiple logistic regression model was fitted using the 11 log transformed protein covariates and their squared counterparts as independent variables and the EVLP outcome as the dependent variable. The optimal model was established by leave-one-out cross-validation using the software package R from R Core Team (2014) (R Foundation for Statistical Computing, Vienna, Austria) [21].
title	Statistical analysis
p	Donor characteristics and physiological parameters are expressed as medians with interquartile ranges and were compared between transplanted and non-transplanted lungs using Mann–Whitney U tests. Paired samples, start-end of perfusion, were compared with Wilcoxon signed-ranks tests. Log protein expressions were compared between transplanted and non-transplanted lungs with multiple t-tests. Correlations between IL-8 and IL-1β expressions and seven post-transplant outcomes (PaO2:FiO2 24-h post-transplant; PGD3 at 72 h post-transplant; ventilation time; intensive care unit stay hospital stay‘; percent of predicted FEV1 at 6 months post-transplant; and percent of predicted FVC at 6 months post-transplant) were analysed by Pearson’s correlation tests. The data were transformed back into non-logged values for reporting as mean (T) for transplanted and mean (NT) for non-transplanted lungs with standard deviations (SD). Multiple testing of donor parameters, protein analyses, and correlations was corrected using the Benjamini-Hochberg false discovery rate (FDR) controlling procedure [20]. Because this was a feasibility study aiming to identify potential markers for further investigation in a validation cohort, an FDR corrected P-value <0.1 was deemed significant. Corrected P-values are reported.
p	A multiple logistic regression model was fitted using the 11 log transformed protein covariates and their squared counterparts as independent variables and the EVLP outcome as the dependent variable. The optimal model was established by leave-one-out cross-validation using the software package R from R Core Team (2014) (R Foundation for Statistical Computing, Vienna, Austria) [21].
sec	RESULTS Study group and donor characteristics Sixteen consecutive clinical EVLP assessments of lungs deemed unacceptable for standard transplant were included in the study (Table 1). Seven (44%) were accepted for transplant after EVLP assessment and were implanted as four bilateral and three single-lung procedures. Nine (56%) failed to achieve transplant criteria and were excluded. The characteristics of the donor groups of transplanted and non-transplanted lungs were not significantly different and are shown in Tables 1 and 2. Table 1: Donor characteristics Donor no. Age/ gender Smoker Donor cause of death Donor type TLC (L) Micro on offer Radiographic infiltrates Secretions EVLP indication EVLP 01 18 m No Diabetic coma DL DBD 7.30 Nil Bilateral Nil PaO2 <300 mmHg EVLP 02 51 f Yes ICH DL DBD 4.31 Nil Clear Nil PaO2 <300 mmHg EVLP 03 33 m Yes ICH DL DBD 7.14 Nil Left diffuse Purulent PaO2 <300 mmHg EVLP 04 15 f No Hanging DL DCD 5.10 Nil Bilateral Nil PaO2 <300 mmHg EVLP 05 52 m No ICH DL DBD 5.78 Nil Clear Mucopurulent PaO2 <300 mmHg EVLP 06 57 m Yes ICH DL DBD 7.54 Nil Right basal Mucopurulent oedema EVLP 07 52 f Yes ICH SL DBD 2.70 Nil Bibasal Purulent PaO2 <300 mmHg EVLP 08 20 m Yes Hanging SL DBD 3.18 Nil Right basal Nil PV PaO2 <225 mmHg + oedema EVLP 09 36 m No Brain tumour SL DBD 3.11 Yesa Left diffuse Nil PaO2 <300 mmHg EVLP 10 47 m No HBI DL DBD 6.74 Nil Clear Nil PaO2 <300 mmHg EVLP 11 45 f No TBI DL DBD 6.18 Nil Clear Nil PV PaO2 <225 mmHg + mild contusion RLL EVLP 12 56 m No ICH DL DBD 6.50 Nil Clear Purulent PaO2 <300 mmHg EVLP 13 18 f Yes ICH DL DCD 5.03 Nil Bilateral Mucopurulent oedema EVLP 14 53 m Yes ICH DL DBD 7.06 Nil Clear Purulent PaO2 <300 mmHg EVLP 15 48 f No ICH DL DBD 4.57 Nil Bibasal Purulent PaO2 <300 mmHg EVLP 16 38 m Yes Hanging DL DCD 6.98 Nil Left basal Nil oedema Shaded donor lungs were not transplanted after EVLP . BALF: bronchoalveolar lavage fluid; EVLP : ex vivo lung perfusion; ICH: intracranial haemorrhage; HBI: hypoxic brain injury secondary to variceal bleed; TBI: traumatic brain injury following a road traffic accident; DL: double lung perfusion; SL: single lung perfusion; DBD: organ donation after brain death; DCD: organ donation after circulatory death; TLC: predicted total lung capacity of perfused lung; PaO2: partial pressure of arterial oxygen; PV: selective pulmonary vein blood gas analyses; RLL: right lower lobe. a Sputum culture with moderate growth of Staphylococcus aureus, Haemophilus influenzae, and Streptococcus pneumoniae. Table 2: EVLP and transplant outcomes Donor no. Optimized donor PaO2:FiO2 (mmHg) PaO2:FiO2 after EVLP (mmHg) Ischaemic time (min)a EVLP time (min)b Total time ex vivo (min)c Transplant PGD 3 at 72 h 90-day survival EVLP 01 149 521 345 290 1020 Yes–BL Yes Yes EVLP 02 248 342 360 360 N/A No–emphysema EVLP 03 209 167 330 270 N/A No–gross oedema EVLP 04 222 426 450 360 870 Yes–SL No Yes EVLP 05 167 440 320 330 N/A No–heavy gram-stain in BALF EVLP 06 357 525 395 240 582 Yes–SL No Yes EVLP 07 178 557 315 255 1048 Yes–SL Yes Yes EVLP 08 372 392 340 240 N/A No–emphysema EVLP 09 293 330 430 300 N/A No–gross oedema EVLP 10 171 291 305 300 N/A No–gross oedema EVLP 11 526 452 340 330 N/A No–gross oedema EVLP 12 191 512 395 220 785 Yes–BL No Yesd EVLP 13 360 352 345 180 N/A No–gross oedema EVLP 14 165 516 270 190 496 Yes–BL No Yes EVLP 15 293 443 360 165 713 Yes–BL No Yes EVLP 16 527 322 335 180 N/A No–gross oedema Shaded donor lungs were not transplanted after EVLP. PaO2:FiO2: the ratio of partial pressure of arterial oxygen to fraction of inspired oxygen; EVLP: ex vivo lung perfusion; PGD: primary graft dysfunction at 72 h; SL: single lung transplant; BL: bilateral lung transplant; N/A: not applicable; BALF: bronchoalveolar lavage fluid. a Start of donor lung flush at procurement to start of pulmonary artery (PA) perfusion on EVLP circuit. b Start of PA perfusion to disconnection from EVLP circuit. c Start of donor lung flush at procurement to reperfusion in recipient. d Death due to H1N1 infection 11 months post-transplant. Ex vivo lung perfusion outcomes The median partial pressure of oxygen on 100% inspired oxygen (PaO2:FiO2) at the end of EVLP was 516 (478–523) mmHg for transplanted and 342 (322–392) mmHg for non-transplanted donor lungs (Mann–Whitney U = 4, P = 0.02). The median PaO2:FiO2 improvement after EVLP was 321 (186–362) mmHg in transplanted donor lungs (Wilcoxon Z = −2.37, P = 0.03). No improvement was seen in lungs that failed assessment; the median PaO2:FiO2 change was 20 (−42–94) mmHg, (Wilcoxon Z = −0.18, P = 0.91) (Fig. 1). Figure 1: Physiological characteristics of transplanted and non-transplanted EVLP donor lungs. Box and whisker plots with boxes showing medians with interquartile range and whiskers at max and min. P<0.05. EVLP; start time: point when the lung had been successfully rewarmed to 37 °C in the beginning of the EVLP; end time: point before start of cooling at the end of the EVLP. Of the nine lungs that failed EVLP reconditioning, six displayed worsening pulmonary oedema and two showed signs of persistent hyperinflation and raised PVR, suggesting diffuse injury and a degree of intrinsic emphysema. This suggestion was later verified by a pathologist from pulmonary biopsies. Lastly, one set of lungs was turned down due to heavy microbial contamination with multiple gram-negative organisms identified in the tracheal aspirate. Transplant outcomes In the seven donor lungs that were transplanted, perfusion and ventilation parameters remained stable or improved during assessment. The data on changes in lung compliance, airway resistance, PVR, perfusate flow, and ventilation and perfusion pressures are shown in Fig. 1. None of the recipients required extracorporeal life support. Two patients (29%) suffered from PGD (Grade 3) at 72 h post-transplant. However, both patients responded well to standard management. All seven recipients survived to hospital discharge. One recipient died of severe pneumonia at 11 months post-transplant due to an influenza infection unrelated to the EVLP procedure. The other six (86%) remain alive and well >2 years after transplant (Table 2). Analysis of perfusate and bronchoalveolar lavage fluid Assessed lungs were divided into two groups based on decision after EVLP: transplanted (n = 7) or non-transplanted (n = 9). Decision on transplant suitability was based on strict criteria summarized in Supplementary Table S2. Samples were analysed retrospectively for protein expressions. For an initial general assessment of tissue injury, we measured LDH levels. LDH was found in substantially lower levels in perfusate from lungs meeting criteria for transplant (T) after EVLP compared to those that did not meet the criteria for transplant (NT) lungs. Suitability for transplant was evident after 2 h of perfusion: mean (T) = 0.150 U (SD 0.030, n = 14) and mean (NT) = 0.223 U (SD 0.082, n = 18) [t(30) = 2.77, P = 0.03] (Fig. 2). The LDH perfusate levels remained significantly different when assessed over all sample time points: mean (T) = 0.149 U (SD 0.033, n = 24) and mean (NT) = 0.259 U (SD 0.088, n = 40) [t(62) = 5.45, P < 0.001]. No difference in LDH levels was seen in BALF samples collected pre- and post-perfusion. Figure 2: Protein expressions during EVLP in perfusate from transplanted and non-transplanted donor lungs. Protein levels expressed as medians with interquartile ranges. P<0.05. EVLP: ex vivo lung perfusion; LDH: lactate dehydrogenase; IL: interleukin; TNF-α: tumour necrosis factor alpha; HMGB-1: high-mobility group box-1; Time: hours of ex vivo lung perfusion after start of pulmonary artery perfusion. To assess the integrity of the pulmonary vascular compartment, we measured Syndecan-1 levels as a marker of endothelial glycocaylx disruption. Syndecan-1 levels were significantly lower in perfusate samples from transplanted donor lungs: mean (T) = 111 (pg/ml) (SD 98, n = 24) compared to mean (NT) = 281 (pg/ml) (SD 219, n = 40) [t(62) = 4.14, P < 0.001] (Fig. 3). Figure 3: Protein expressions in perfusate and BALF from transplanted and non-transplanted EVLP donor lungs. Interleaved scatter plots with molecular marker levels expressed in (pg/ml) and lines representing means. P<0.1. P<0.05. P<0.001. BALF: bronchoalveolar lavage fluid; EVLP: ex vivo lung perfusion; IL: interleukin; TNF-α: tumour necrosis factor alpha; LDH: lactate dehydrogenase; HMGB-1: high-mobility group box-1. We noted a similarly consistent pattern towards lower release of proinflammatory cytokines into the circulating perfusate from transplanted compared to non-transplanted lungs. The average level of IL-8 was nearly eight times higher in perfusate samples from non-transplanted lungs: mean (T) = 165 (pg/ml) (SD 155, n = 24) compared to mean (NT) = 1310 (pg/ml) (SD 1510, n = 35) [t(57) = 2.27, P = 0.06]. Levels of the anti-inflammatory cytokine IL-10 increased noticeably over the first 2 h of perfusion in the transplanted group, with high initial fold changes in both perfusate and BALF (Fig. 4). Perfusate levels then diminished and were less than half those in non-transplanted lungs when measured over the full assessment: mean (T) = 9 (pg/ml) (SD 11, n = 24) compared to mean (NT) = 23 (pg/ml) (SD 32, n = 40) [t(57) = 2.02, P = 0.07]. No difference in levels of IL-10 in BALF was seen between the groups. Figure 4: Protein median fold changes in perfusate and BALF from EVLP donor lungs. Bar graphs showing median fold changes with interquartile range. BALF: bronchoalveolar lavage fluid; EVLP: ex vivo lung perfusion; IL: interleukin; TNF-α: tumour necrosis factor alpha; LDH: lactate dehydrogenase; HMGB-1: high-mobility group box-1. A panel of three damage-associated molecular patterns, IL-33, HMGB-1 and S100A9, was used to assess the extent of cellular injury in the donor lung. Lower levels of IL-33 and HMGB-1 were found in perfusate from transplanted donor lungs. The average IL-33 levels in perfusate were mean (T) = 3 (pg/ml) (SD 5, n = 24) and mean (NT) = 9 (pg/ml) (SD 10, n = 40) [t(62) = 2.93, P = 0.02]. HMGB-1 was highly expressed in both perfusate and BALF. The HMGB-1 levels in perfusate were mean (T) = 8318 (pg/ml) (SD 4474, n = 24) and mean (NT) = 10 545 (pg/ml) (SD 4,313, n = 40) [t(62) = 2.10, P = 0.07]. No difference in lavage fluid levels of IL-33 or HMGB-1 was seen between the groups. We noted a consistent pattern towards increasing perfusate protein levels and a more pronounced separation between transplanted and non-transplanted lungs over time of perfusion (Fig. 2). This pattern was most noticeable for the investigated inflammatory cytokines. The only protein marker not showing this pattern was S100A9, which appeared lower in the non-transplanted group. A multivariate analysis model was fitted to assess the predictive value of combining two or more protein markers. The optimal model was established by leave-one-out cross-validation and is shown in Fig. 5 as a scatterplot of IL-1β and IL-8 perfusate levels from the 16 assessments. The shaded prediction region was derived from the regression analysis with covariates IL-1β, IL-8 and (IL-8)2 after 2 h of perfusion. The model suggests that the closer to the centre of the area, the higher was the probability of a donor lung being found suitable for transplant after EVLP assessment. This early IL-1β–IL-8 signature remained equally as robust when applied to the full set of samples (Fig. 5B). An ROC curve showing the potential benefit of the model in predicting EVLP outcome compared to random is shown in Fig. 5C, with a sensitivity of 0.81 and specificity of 0.92 at 2 h of perfusion. Figure 5: (A) Early IL-1β-IL-8 signature in EVLP perfusate after 2 h of perfusion. (B) IL-1β-IL-8 signature in EVLP perfusate after full assessment. (C) ROC curve. (A) and (B) Scatter plots of IL-1β and IL-8 levels in perfusate samples from transplanted and non-transplanted EVLP donor lungs. The shaded area is the prediction region from the optimal logistic regression model with covariates IL-1β, IL-8 and (IL-8)2. (C) Receiver operating characteristic (ROC) curve of the model as a predictor of EVLP outcome. Calculated sensitivity (y axis) is plotted against the 1-specificity formula (x axis) of the logistic regression function 2 h into perfusion and after full assessment. EVLP: ex vivo lung perfusion; IL: interleukin. The modelled proteins were lastly assessed for correlations to post-transplant outcome measures. Donor IL-1β levels in the perfusate at time of transplant decision showed a significant negative correlation to recipient PaO2:FiO2 24 h post-transplant (r = −093, P = 0.03) (Fig. 6). No other correlations remained significant after corrections for multiple testing in this cohort. Figure 6: Association between donor IL-1β level in EVLP perfusate at the time of transplant decision and recipient P/F ratio 24 h post lung transplantation. Pearson correlation plot. P<0.05. EVLP: ex vivo lung perfusion; IL: interleukin; PaO2:FiO2: partial pressure of oxygen on 100% inspired oxygen.
title	RESULTS
sec	Study group and donor characteristics Sixteen consecutive clinical EVLP assessments of lungs deemed unacceptable for standard transplant were included in the study (Table 1). Seven (44%) were accepted for transplant after EVLP assessment and were implanted as four bilateral and three single-lung procedures. Nine (56%) failed to achieve transplant criteria and were excluded. The characteristics of the donor groups of transplanted and non-transplanted lungs were not significantly different and are shown in Tables 1 and 2. Table 1: Donor characteristics Donor no. Age/ gender Smoker Donor cause of death Donor type TLC (L) Micro on offer Radiographic infiltrates Secretions EVLP indication EVLP 01 18 m No Diabetic coma DL DBD 7.30 Nil Bilateral Nil PaO2 <300 mmHg EVLP 02 51 f Yes ICH DL DBD 4.31 Nil Clear Nil PaO2 <300 mmHg EVLP 03 33 m Yes ICH DL DBD 7.14 Nil Left diffuse Purulent PaO2 <300 mmHg EVLP 04 15 f No Hanging DL DCD 5.10 Nil Bilateral Nil PaO2 <300 mmHg EVLP 05 52 m No ICH DL DBD 5.78 Nil Clear Mucopurulent PaO2 <300 mmHg EVLP 06 57 m Yes ICH DL DBD 7.54 Nil Right basal Mucopurulent oedema EVLP 07 52 f Yes ICH SL DBD 2.70 Nil Bibasal Purulent PaO2 <300 mmHg EVLP 08 20 m Yes Hanging SL DBD 3.18 Nil Right basal Nil PV PaO2 <225 mmHg + oedema EVLP 09 36 m No Brain tumour SL DBD 3.11 Yesa Left diffuse Nil PaO2 <300 mmHg EVLP 10 47 m No HBI DL DBD 6.74 Nil Clear Nil PaO2 <300 mmHg EVLP 11 45 f No TBI DL DBD 6.18 Nil Clear Nil PV PaO2 <225 mmHg + mild contusion RLL EVLP 12 56 m No ICH DL DBD 6.50 Nil Clear Purulent PaO2 <300 mmHg EVLP 13 18 f Yes ICH DL DCD 5.03 Nil Bilateral Mucopurulent oedema EVLP 14 53 m Yes ICH DL DBD 7.06 Nil Clear Purulent PaO2 <300 mmHg EVLP 15 48 f No ICH DL DBD 4.57 Nil Bibasal Purulent PaO2 <300 mmHg EVLP 16 38 m Yes Hanging DL DCD 6.98 Nil Left basal Nil oedema Shaded donor lungs were not transplanted after EVLP . BALF: bronchoalveolar lavage fluid; EVLP : ex vivo lung perfusion; ICH: intracranial haemorrhage; HBI: hypoxic brain injury secondary to variceal bleed; TBI: traumatic brain injury following a road traffic accident; DL: double lung perfusion; SL: single lung perfusion; DBD: organ donation after brain death; DCD: organ donation after circulatory death; TLC: predicted total lung capacity of perfused lung; PaO2: partial pressure of arterial oxygen; PV: selective pulmonary vein blood gas analyses; RLL: right lower lobe. a Sputum culture with moderate growth of Staphylococcus aureus, Haemophilus influenzae, and Streptococcus pneumoniae. Table 2: EVLP and transplant outcomes Donor no. Optimized donor PaO2:FiO2 (mmHg) PaO2:FiO2 after EVLP (mmHg) Ischaemic time (min)a EVLP time (min)b Total time ex vivo (min)c Transplant PGD 3 at 72 h 90-day survival EVLP 01 149 521 345 290 1020 Yes–BL Yes Yes EVLP 02 248 342 360 360 N/A No–emphysema EVLP 03 209 167 330 270 N/A No–gross oedema EVLP 04 222 426 450 360 870 Yes–SL No Yes EVLP 05 167 440 320 330 N/A No–heavy gram-stain in BALF EVLP 06 357 525 395 240 582 Yes–SL No Yes EVLP 07 178 557 315 255 1048 Yes–SL Yes Yes EVLP 08 372 392 340 240 N/A No–emphysema EVLP 09 293 330 430 300 N/A No–gross oedema EVLP 10 171 291 305 300 N/A No–gross oedema EVLP 11 526 452 340 330 N/A No–gross oedema EVLP 12 191 512 395 220 785 Yes–BL No Yesd EVLP 13 360 352 345 180 N/A No–gross oedema EVLP 14 165 516 270 190 496 Yes–BL No Yes EVLP 15 293 443 360 165 713 Yes–BL No Yes EVLP 16 527 322 335 180 N/A No–gross oedema Shaded donor lungs were not transplanted after EVLP. PaO2:FiO2: the ratio of partial pressure of arterial oxygen to fraction of inspired oxygen; EVLP: ex vivo lung perfusion; PGD: primary graft dysfunction at 72 h; SL: single lung transplant; BL: bilateral lung transplant; N/A: not applicable; BALF: bronchoalveolar lavage fluid. a Start of donor lung flush at procurement to start of pulmonary artery (PA) perfusion on EVLP circuit. b Start of PA perfusion to disconnection from EVLP circuit. c Start of donor lung flush at procurement to reperfusion in recipient. d Death due to H1N1 infection 11 months post-transplant.
title	Study group and donor characteristics
p	Sixteen consecutive clinical EVLP assessments of lungs deemed unacceptable for standard transplant were included in the study (Table 1). Seven (44%) were accepted for transplant after EVLP assessment and were implanted as four bilateral and three single-lung procedures. Nine (56%) failed to achieve transplant criteria and were excluded. The characteristics of the donor groups of transplanted and non-transplanted lungs were not significantly different and are shown in Tables 1 and 2. Table 1: Donor characteristics Donor no. Age/ gender Smoker Donor cause of death Donor type TLC (L) Micro on offer Radiographic infiltrates Secretions EVLP indication EVLP 01 18 m No Diabetic coma DL DBD 7.30 Nil Bilateral Nil PaO2 <300 mmHg EVLP 02 51 f Yes ICH DL DBD 4.31 Nil Clear Nil PaO2 <300 mmHg EVLP 03 33 m Yes ICH DL DBD 7.14 Nil Left diffuse Purulent PaO2 <300 mmHg EVLP 04 15 f No Hanging DL DCD 5.10 Nil Bilateral Nil PaO2 <300 mmHg EVLP 05 52 m No ICH DL DBD 5.78 Nil Clear Mucopurulent PaO2 <300 mmHg EVLP 06 57 m Yes ICH DL DBD 7.54 Nil Right basal Mucopurulent oedema EVLP 07 52 f Yes ICH SL DBD 2.70 Nil Bibasal Purulent PaO2 <300 mmHg EVLP 08 20 m Yes Hanging SL DBD 3.18 Nil Right basal Nil PV PaO2 <225 mmHg + oedema EVLP 09 36 m No Brain tumour SL DBD 3.11 Yesa Left diffuse Nil PaO2 <300 mmHg EVLP 10 47 m No HBI DL DBD 6.74 Nil Clear Nil PaO2 <300 mmHg EVLP 11 45 f No TBI DL DBD 6.18 Nil Clear Nil PV PaO2 <225 mmHg + mild contusion RLL EVLP 12 56 m No ICH DL DBD 6.50 Nil Clear Purulent PaO2 <300 mmHg EVLP 13 18 f Yes ICH DL DCD 5.03 Nil Bilateral Mucopurulent oedema EVLP 14 53 m Yes ICH DL DBD 7.06 Nil Clear Purulent PaO2 <300 mmHg EVLP 15 48 f No ICH DL DBD 4.57 Nil Bibasal Purulent PaO2 <300 mmHg EVLP 16 38 m Yes Hanging DL DCD 6.98 Nil Left basal Nil oedema Shaded donor lungs were not transplanted after EVLP . BALF: bronchoalveolar lavage fluid; EVLP : ex vivo lung perfusion; ICH: intracranial haemorrhage; HBI: hypoxic brain injury secondary to variceal bleed; TBI: traumatic brain injury following a road traffic accident; DL: double lung perfusion; SL: single lung perfusion; DBD: organ donation after brain death; DCD: organ donation after circulatory death; TLC: predicted total lung capacity of perfused lung; PaO2: partial pressure of arterial oxygen; PV: selective pulmonary vein blood gas analyses; RLL: right lower lobe. a Sputum culture with moderate growth of Staphylococcus aureus, Haemophilus influenzae, and Streptococcus pneumoniae. Table 2: EVLP and transplant outcomes Donor no. Optimized donor PaO2:FiO2 (mmHg) PaO2:FiO2 after EVLP (mmHg) Ischaemic time (min)a EVLP time (min)b Total time ex vivo (min)c Transplant PGD 3 at 72 h 90-day survival EVLP 01 149 521 345 290 1020 Yes–BL Yes Yes EVLP 02 248 342 360 360 N/A No–emphysema EVLP 03 209 167 330 270 N/A No–gross oedema EVLP 04 222 426 450 360 870 Yes–SL No Yes EVLP 05 167 440 320 330 N/A No–heavy gram-stain in BALF EVLP 06 357 525 395 240 582 Yes–SL No Yes EVLP 07 178 557 315 255 1048 Yes–SL Yes Yes EVLP 08 372 392 340 240 N/A No–emphysema EVLP 09 293 330 430 300 N/A No–gross oedema EVLP 10 171 291 305 300 N/A No–gross oedema EVLP 11 526 452 340 330 N/A No–gross oedema EVLP 12 191 512 395 220 785 Yes–BL No Yesd EVLP 13 360 352 345 180 N/A No–gross oedema EVLP 14 165 516 270 190 496 Yes–BL No Yes EVLP 15 293 443 360 165 713 Yes–BL No Yes EVLP 16 527 322 335 180 N/A No–gross oedema Shaded donor lungs were not transplanted after EVLP. PaO2:FiO2: the ratio of partial pressure of arterial oxygen to fraction of inspired oxygen; EVLP: ex vivo lung perfusion; PGD: primary graft dysfunction at 72 h; SL: single lung transplant; BL: bilateral lung transplant; N/A: not applicable; BALF: bronchoalveolar lavage fluid. a Start of donor lung flush at procurement to start of pulmonary artery (PA) perfusion on EVLP circuit. b Start of PA perfusion to disconnection from EVLP circuit. c Start of donor lung flush at procurement to reperfusion in recipient. d Death due to H1N1 infection 11 months post-transplant.
table-wrap	Table 1: Donor characteristics Donor no. Age/ gender Smoker Donor cause of death Donor type TLC (L) Micro on offer Radiographic infiltrates Secretions EVLP indication EVLP 01 18 m No Diabetic coma DL DBD 7.30 Nil Bilateral Nil PaO2 <300 mmHg EVLP 02 51 f Yes ICH DL DBD 4.31 Nil Clear Nil PaO2 <300 mmHg EVLP 03 33 m Yes ICH DL DBD 7.14 Nil Left diffuse Purulent PaO2 <300 mmHg EVLP 04 15 f No Hanging DL DCD 5.10 Nil Bilateral Nil PaO2 <300 mmHg EVLP 05 52 m No ICH DL DBD 5.78 Nil Clear Mucopurulent PaO2 <300 mmHg EVLP 06 57 m Yes ICH DL DBD 7.54 Nil Right basal Mucopurulent oedema EVLP 07 52 f Yes ICH SL DBD 2.70 Nil Bibasal Purulent PaO2 <300 mmHg EVLP 08 20 m Yes Hanging SL DBD 3.18 Nil Right basal Nil PV PaO2 <225 mmHg + oedema EVLP 09 36 m No Brain tumour SL DBD 3.11 Yesa Left diffuse Nil PaO2 <300 mmHg EVLP 10 47 m No HBI DL DBD 6.74 Nil Clear Nil PaO2 <300 mmHg EVLP 11 45 f No TBI DL DBD 6.18 Nil Clear Nil PV PaO2 <225 mmHg + mild contusion RLL EVLP 12 56 m No ICH DL DBD 6.50 Nil Clear Purulent PaO2 <300 mmHg EVLP 13 18 f Yes ICH DL DCD 5.03 Nil Bilateral Mucopurulent oedema EVLP 14 53 m Yes ICH DL DBD 7.06 Nil Clear Purulent PaO2 <300 mmHg EVLP 15 48 f No ICH DL DBD 4.57 Nil Bibasal Purulent PaO2 <300 mmHg EVLP 16 38 m Yes Hanging DL DCD 6.98 Nil Left basal Nil oedema Shaded donor lungs were not transplanted after EVLP . BALF: bronchoalveolar lavage fluid; EVLP : ex vivo lung perfusion; ICH: intracranial haemorrhage; HBI: hypoxic brain injury secondary to variceal bleed; TBI: traumatic brain injury following a road traffic accident; DL: double lung perfusion; SL: single lung perfusion; DBD: organ donation after brain death; DCD: organ donation after circulatory death; TLC: predicted total lung capacity of perfused lung; PaO2: partial pressure of arterial oxygen; PV: selective pulmonary vein blood gas analyses; RLL: right lower lobe. a Sputum culture with moderate growth of Staphylococcus aureus, Haemophilus influenzae, and Streptococcus pneumoniae.
label	Table 1:
caption	Donor characteristics
p	Donor characteristics
table	Donor no. Age/ gender Smoker Donor cause of death Donor type TLC (L) Micro on offer Radiographic infiltrates Secretions EVLP indication EVLP 01 18 m No Diabetic coma DL DBD 7.30 Nil Bilateral Nil PaO2 <300 mmHg EVLP 02 51 f Yes ICH DL DBD 4.31 Nil Clear Nil PaO2 <300 mmHg EVLP 03 33 m Yes ICH DL DBD 7.14 Nil Left diffuse Purulent PaO2 <300 mmHg EVLP 04 15 f No Hanging DL DCD 5.10 Nil Bilateral Nil PaO2 <300 mmHg EVLP 05 52 m No ICH DL DBD 5.78 Nil Clear Mucopurulent PaO2 <300 mmHg EVLP 06 57 m Yes ICH DL DBD 7.54 Nil Right basal Mucopurulent oedema EVLP 07 52 f Yes ICH SL DBD 2.70 Nil Bibasal Purulent PaO2 <300 mmHg EVLP 08 20 m Yes Hanging SL DBD 3.18 Nil Right basal Nil PV PaO2 <225 mmHg + oedema EVLP 09 36 m No Brain tumour SL DBD 3.11 Yesa Left diffuse Nil PaO2 <300 mmHg EVLP 10 47 m No HBI DL DBD 6.74 Nil Clear Nil PaO2 <300 mmHg EVLP 11 45 f No TBI DL DBD 6.18 Nil Clear Nil PV PaO2 <225 mmHg + mild contusion RLL EVLP 12 56 m No ICH DL DBD 6.50 Nil Clear Purulent PaO2 <300 mmHg EVLP 13 18 f Yes ICH DL DCD 5.03 Nil Bilateral Mucopurulent oedema EVLP 14 53 m Yes ICH DL DBD 7.06 Nil Clear Purulent PaO2 <300 mmHg EVLP 15 48 f No ICH DL DBD 4.57 Nil Bibasal Purulent PaO2 <300 mmHg EVLP 16 38 m Yes Hanging DL DCD 6.98 Nil Left basal Nil oedema
tr	Donor no. Age/ gender Smoker Donor cause of death Donor type TLC (L) Micro on offer Radiographic infiltrates Secretions EVLP indication
th	Donor no.
th	Age/ gender
th	Smoker
th	Donor cause of death
th	Donor type
th	TLC (L)
th	Micro on offer
th	Radiographic infiltrates
th	Secretions
th	EVLP indication
tr	EVLP 01 18 m No Diabetic coma DL DBD 7.30 Nil Bilateral Nil PaO2 <300 mmHg
td	EVLP 01
td	18 m
td	No
td	Diabetic coma
td	DL DBD
td	7.30
td	Nil
td	Bilateral
td	Nil
td	PaO2 <300 mmHg
tr	EVLP 02 51 f Yes ICH DL DBD 4.31 Nil Clear Nil PaO2 <300 mmHg
td	EVLP 02
td	51 f
td	Yes
td	ICH
td	DL DBD
td	4.31
td	Nil
td	Clear
td	Nil
td	PaO2 <300 mmHg
tr	EVLP 03 33 m Yes ICH DL DBD 7.14 Nil Left diffuse Purulent PaO2 <300 mmHg
td	EVLP 03
td	33 m
td	Yes
td	ICH
td	DL DBD
td	7.14
td	Nil
td	Left diffuse
td	Purulent
td	PaO2 <300 mmHg
tr	EVLP 04 15 f No Hanging DL DCD 5.10 Nil Bilateral Nil PaO2 <300 mmHg
td	EVLP 04
td	15 f
td	No
td	Hanging
td	DL DCD
td	5.10
td	Nil
td	Bilateral
td	Nil
td	PaO2 <300 mmHg
tr	EVLP 05 52 m No ICH DL DBD 5.78 Nil Clear Mucopurulent PaO2 <300 mmHg
td	EVLP 05
td	52 m
td	No
td	ICH
td	DL DBD
td	5.78
td	Nil
td	Clear
td	Mucopurulent
td	PaO2 <300 mmHg
tr	EVLP 06 57 m Yes ICH DL DBD 7.54 Nil Right basal Mucopurulent oedema
td	EVLP 06
td	57 m
td	Yes
td	ICH
td	DL DBD
td	7.54
td	Nil
td	Right basal
td	Mucopurulent
td	oedema
tr	EVLP 07 52 f Yes ICH SL DBD 2.70 Nil Bibasal Purulent PaO2 <300 mmHg
td	EVLP 07
td	52 f
td	Yes
td	ICH
td	SL DBD
td	2.70
td	Nil
td	Bibasal
td	Purulent
td	PaO2 <300 mmHg
tr	EVLP 08 20 m Yes Hanging SL DBD 3.18 Nil Right basal Nil PV PaO2 <225 mmHg + oedema
td	EVLP 08
td	20 m
td	Yes
td	Hanging
td	SL DBD
td	3.18
td	Nil
td	Right basal
td	Nil
td	PV PaO2 <225 mmHg + oedema
tr	EVLP 09 36 m No Brain tumour SL DBD 3.11 Yesa Left diffuse Nil PaO2 <300 mmHg
td	EVLP 09
td	36 m
td	No
td	Brain tumour
td	SL DBD
td	3.11
td	Yesa
td	Left diffuse
td	Nil
td	PaO2 <300 mmHg
tr	EVLP 10 47 m No HBI DL DBD 6.74 Nil Clear Nil PaO2 <300 mmHg
td	EVLP 10
td	47 m
td	No
td	HBI
td	DL DBD
td	6.74
td	Nil
td	Clear
td	Nil
td	PaO2 <300 mmHg
tr	EVLP 11 45 f No TBI DL DBD 6.18 Nil Clear Nil PV PaO2 <225 mmHg + mild contusion RLL
td	EVLP 11
td	45 f
td	No
td	TBI
td	DL DBD
td	6.18
td	Nil
td	Clear
td	Nil
td	PV PaO2 <225 mmHg + mild contusion RLL
tr	EVLP 12 56 m No ICH DL DBD 6.50 Nil Clear Purulent PaO2 <300 mmHg
td	EVLP 12
td	56 m
td	No
td	ICH
td	DL DBD
td	6.50
td	Nil
td	Clear
td	Purulent
td	PaO2 <300 mmHg
tr	EVLP 13 18 f Yes ICH DL DCD 5.03 Nil Bilateral Mucopurulent oedema
td	EVLP 13
td	18 f
td	Yes
td	ICH
td	DL DCD
td	5.03
td	Nil
td	Bilateral
td	Mucopurulent
td	oedema
tr	EVLP 14 53 m Yes ICH DL DBD 7.06 Nil Clear Purulent PaO2 <300 mmHg
td	EVLP 14
td	53 m
td	Yes
td	ICH
td	DL DBD
td	7.06
td	Nil
td	Clear
td	Purulent
td	PaO2 <300 mmHg
tr	EVLP 15 48 f No ICH DL DBD 4.57 Nil Bibasal Purulent PaO2 <300 mmHg
td	EVLP 15
td	48 f
td	No
td	ICH
td	DL DBD
td	4.57
td	Nil
td	Bibasal
td	Purulent
td	PaO2 <300 mmHg
tr	EVLP 16 38 m Yes Hanging DL DCD 6.98 Nil Left basal Nil oedema
td	EVLP 16
td	38 m
td	Yes
td	Hanging
td	DL DCD
td	6.98
td	Nil
td	Left basal
td	Nil
td	oedema
table-wrap-foot	Shaded donor lungs were not transplanted after EVLP . BALF: bronchoalveolar lavage fluid; EVLP : ex vivo lung perfusion; ICH: intracranial haemorrhage; HBI: hypoxic brain injury secondary to variceal bleed; TBI: traumatic brain injury following a road traffic accident; DL: double lung perfusion; SL: single lung perfusion; DBD: organ donation after brain death; DCD: organ donation after circulatory death; TLC: predicted total lung capacity of perfused lung; PaO2: partial pressure of arterial oxygen; PV: selective pulmonary vein blood gas analyses; RLL: right lower lobe. a Sputum culture with moderate growth of Staphylococcus aureus, Haemophilus influenzae, and Streptococcus pneumoniae.
footnote	Shaded donor lungs were not transplanted after EVLP .
p	Shaded donor lungs were not transplanted after EVLP .
footnote	BALF: bronchoalveolar lavage fluid; EVLP : ex vivo lung perfusion; ICH: intracranial haemorrhage; HBI: hypoxic brain injury secondary to variceal bleed; TBI: traumatic brain injury following a road traffic accident; DL: double lung perfusion; SL: single lung perfusion; DBD: organ donation after brain death; DCD: organ donation after circulatory death; TLC: predicted total lung capacity of perfused lung; PaO2: partial pressure of arterial oxygen; PV: selective pulmonary vein blood gas analyses; RLL: right lower lobe.
p	BALF: bronchoalveolar lavage fluid; EVLP : ex vivo lung perfusion; ICH: intracranial haemorrhage; HBI: hypoxic brain injury secondary to variceal bleed; TBI: traumatic brain injury following a road traffic accident; DL: double lung perfusion; SL: single lung perfusion; DBD: organ donation after brain death; DCD: organ donation after circulatory death; TLC: predicted total lung capacity of perfused lung; PaO2: partial pressure of arterial oxygen; PV: selective pulmonary vein blood gas analyses; RLL: right lower lobe.
footnote	a Sputum culture with moderate growth of Staphylococcus aureus, Haemophilus influenzae, and Streptococcus pneumoniae.
label	a
p	Sputum culture with moderate growth of Staphylococcus aureus, Haemophilus influenzae, and Streptococcus pneumoniae.
table-wrap	Table 2: EVLP and transplant outcomes Donor no. Optimized donor PaO2:FiO2 (mmHg) PaO2:FiO2 after EVLP (mmHg) Ischaemic time (min)a EVLP time (min)b Total time ex vivo (min)c Transplant PGD 3 at 72 h 90-day survival EVLP 01 149 521 345 290 1020 Yes–BL Yes Yes EVLP 02 248 342 360 360 N/A No–emphysema EVLP 03 209 167 330 270 N/A No–gross oedema EVLP 04 222 426 450 360 870 Yes–SL No Yes EVLP 05 167 440 320 330 N/A No–heavy gram-stain in BALF EVLP 06 357 525 395 240 582 Yes–SL No Yes EVLP 07 178 557 315 255 1048 Yes–SL Yes Yes EVLP 08 372 392 340 240 N/A No–emphysema EVLP 09 293 330 430 300 N/A No–gross oedema EVLP 10 171 291 305 300 N/A No–gross oedema EVLP 11 526 452 340 330 N/A No–gross oedema EVLP 12 191 512 395 220 785 Yes–BL No Yesd EVLP 13 360 352 345 180 N/A No–gross oedema EVLP 14 165 516 270 190 496 Yes–BL No Yes EVLP 15 293 443 360 165 713 Yes–BL No Yes EVLP 16 527 322 335 180 N/A No–gross oedema Shaded donor lungs were not transplanted after EVLP. PaO2:FiO2: the ratio of partial pressure of arterial oxygen to fraction of inspired oxygen; EVLP: ex vivo lung perfusion; PGD: primary graft dysfunction at 72 h; SL: single lung transplant; BL: bilateral lung transplant; N/A: not applicable; BALF: bronchoalveolar lavage fluid. a Start of donor lung flush at procurement to start of pulmonary artery (PA) perfusion on EVLP circuit. b Start of PA perfusion to disconnection from EVLP circuit. c Start of donor lung flush at procurement to reperfusion in recipient. d Death due to H1N1 infection 11 months post-transplant.
label	Table 2:
caption	EVLP and transplant outcomes
p	EVLP and transplant outcomes
table	Donor no. Optimized donor PaO2:FiO2 (mmHg) PaO2:FiO2 after EVLP (mmHg) Ischaemic time (min)a EVLP time (min)b Total time ex vivo (min)c Transplant PGD 3 at 72 h 90-day survival EVLP 01 149 521 345 290 1020 Yes–BL Yes Yes EVLP 02 248 342 360 360 N/A No–emphysema EVLP 03 209 167 330 270 N/A No–gross oedema EVLP 04 222 426 450 360 870 Yes–SL No Yes EVLP 05 167 440 320 330 N/A No–heavy gram-stain in BALF EVLP 06 357 525 395 240 582 Yes–SL No Yes EVLP 07 178 557 315 255 1048 Yes–SL Yes Yes EVLP 08 372 392 340 240 N/A No–emphysema EVLP 09 293 330 430 300 N/A No–gross oedema EVLP 10 171 291 305 300 N/A No–gross oedema EVLP 11 526 452 340 330 N/A No–gross oedema EVLP 12 191 512 395 220 785 Yes–BL No Yesd EVLP 13 360 352 345 180 N/A No–gross oedema EVLP 14 165 516 270 190 496 Yes–BL No Yes EVLP 15 293 443 360 165 713 Yes–BL No Yes EVLP 16 527 322 335 180 N/A No–gross oedema
tr	Donor no. Optimized donor PaO2:FiO2 (mmHg) PaO2:FiO2 after EVLP (mmHg) Ischaemic time (min)a EVLP time (min)b Total time ex vivo (min)c Transplant PGD 3 at 72 h 90-day survival
th	Donor no.
th	Optimized donor PaO2:FiO2 (mmHg)
th	PaO2:FiO2 after EVLP (mmHg)
th	Ischaemic time (min)a
th	EVLP time (min)b
th	Total time ex vivo (min)c
th	Transplant
th	PGD 3 at 72 h
th	90-day survival
tr	EVLP 01 149 521 345 290 1020 Yes–BL Yes Yes
td	EVLP 01
td	149
td	521
td	345
td	290
td	1020
td	Yes–BL
td	Yes
td	Yes
tr	EVLP 02 248 342 360 360 N/A No–emphysema
td	EVLP 02
td	248
td	342
td	360
td	360
td	N/A
td	No–emphysema
tr	EVLP 03 209 167 330 270 N/A No–gross oedema
td	EVLP 03
td	209
td	167
td	330
td	270
td	N/A
td	No–gross oedema
tr	EVLP 04 222 426 450 360 870 Yes–SL No Yes
td	EVLP 04
td	222
td	426
td	450
td	360
td	870
td	Yes–SL
td	No
td	Yes
tr	EVLP 05 167 440 320 330 N/A No–heavy gram-stain in BALF
td	EVLP 05
td	167
td	440
td	320
td	330
td	N/A
td	No–heavy gram-stain in BALF
tr	EVLP 06 357 525 395 240 582 Yes–SL No Yes
td	EVLP 06
td	357
td	525
td	395
td	240
td	582
td	Yes–SL
td	No
td	Yes
tr	EVLP 07 178 557 315 255 1048 Yes–SL Yes Yes
td	EVLP 07
td	178
td	557
td	315
td	255
td	1048
td	Yes–SL
td	Yes
td	Yes
tr	EVLP 08 372 392 340 240 N/A No–emphysema
td	EVLP 08
td	372
td	392
td	340
td	240
td	N/A
td	No–emphysema
tr	EVLP 09 293 330 430 300 N/A No–gross oedema
td	EVLP 09
td	293
td	330
td	430
td	300
td	N/A
td	No–gross oedema
tr	EVLP 10 171 291 305 300 N/A No–gross oedema
td	EVLP 10
td	171
td	291
td	305
td	300
td	N/A
td	No–gross oedema
tr	EVLP 11 526 452 340 330 N/A No–gross oedema
td	EVLP 11
td	526
td	452
td	340
td	330
td	N/A
td	No–gross oedema
tr	EVLP 12 191 512 395 220 785 Yes–BL No Yesd
td	EVLP 12
td	191
td	512
td	395
td	220
td	785
td	Yes–BL
td	No
td	Yesd
tr	EVLP 13 360 352 345 180 N/A No–gross oedema
td	EVLP 13
td	360
td	352
td	345
td	180
td	N/A
td	No–gross oedema
tr	EVLP 14 165 516 270 190 496 Yes–BL No Yes
td	EVLP 14
td	165
td	516
td	270
td	190
td	496
td	Yes–BL
td	No
td	Yes
tr	EVLP 15 293 443 360 165 713 Yes–BL No Yes
td	EVLP 15
td	293
td	443
td	360
td	165
td	713
td	Yes–BL
td	No
td	Yes
tr	EVLP 16 527 322 335 180 N/A No–gross oedema
td	EVLP 16
td	527
td	322
td	335
td	180
td	N/A
td	No–gross oedema
table-wrap-foot	Shaded donor lungs were not transplanted after EVLP. PaO2:FiO2: the ratio of partial pressure of arterial oxygen to fraction of inspired oxygen; EVLP: ex vivo lung perfusion; PGD: primary graft dysfunction at 72 h; SL: single lung transplant; BL: bilateral lung transplant; N/A: not applicable; BALF: bronchoalveolar lavage fluid. a Start of donor lung flush at procurement to start of pulmonary artery (PA) perfusion on EVLP circuit. b Start of PA perfusion to disconnection from EVLP circuit. c Start of donor lung flush at procurement to reperfusion in recipient. d Death due to H1N1 infection 11 months post-transplant.
footnote	Shaded donor lungs were not transplanted after EVLP.
p	Shaded donor lungs were not transplanted after EVLP.
footnote	PaO2:FiO2: the ratio of partial pressure of arterial oxygen to fraction of inspired oxygen; EVLP: ex vivo lung perfusion; PGD: primary graft dysfunction at 72 h; SL: single lung transplant; BL: bilateral lung transplant; N/A: not applicable; BALF: bronchoalveolar lavage fluid.
p	PaO2:FiO2: the ratio of partial pressure of arterial oxygen to fraction of inspired oxygen; EVLP: ex vivo lung perfusion; PGD: primary graft dysfunction at 72 h; SL: single lung transplant; BL: bilateral lung transplant; N/A: not applicable; BALF: bronchoalveolar lavage fluid.
footnote	a Start of donor lung flush at procurement to start of pulmonary artery (PA) perfusion on EVLP circuit.
label	a
p	Start of donor lung flush at procurement to start of pulmonary artery (PA) perfusion on EVLP circuit.
footnote	b Start of PA perfusion to disconnection from EVLP circuit.
label	b
p	Start of PA perfusion to disconnection from EVLP circuit.
footnote	c Start of donor lung flush at procurement to reperfusion in recipient.
label	c
p	Start of donor lung flush at procurement to reperfusion in recipient.
footnote	d Death due to H1N1 infection 11 months post-transplant.
label	d
p	Death due to H1N1 infection 11 months post-transplant.
sec	Ex vivo lung perfusion outcomes The median partial pressure of oxygen on 100% inspired oxygen (PaO2:FiO2) at the end of EVLP was 516 (478–523) mmHg for transplanted and 342 (322–392) mmHg for non-transplanted donor lungs (Mann–Whitney U = 4, P = 0.02). The median PaO2:FiO2 improvement after EVLP was 321 (186–362) mmHg in transplanted donor lungs (Wilcoxon Z = −2.37, P = 0.03). No improvement was seen in lungs that failed assessment; the median PaO2:FiO2 change was 20 (−42–94) mmHg, (Wilcoxon Z = −0.18, P = 0.91) (Fig. 1). Figure 1: Physiological characteristics of transplanted and non-transplanted EVLP donor lungs. Box and whisker plots with boxes showing medians with interquartile range and whiskers at max and min. **P<0.05. EVLP; start time: point when the lung had been successfully rewarmed to 37 °C in the beginning of the EVLP; end time: point before start of cooling at the end of the EVLP. Of the nine lungs that failed EVLP reconditioning, six displayed worsening pulmonary oedema and two showed signs of persistent hyperinflation and raised PVR, suggesting diffuse injury and a degree of intrinsic emphysema. This suggestion was later verified by a pathologist from pulmonary biopsies. Lastly, one set of lungs was turned down due to heavy microbial contamination with multiple gram-negative organisms identified in the tracheal aspirate.
title	Ex vivo lung perfusion outcomes
p	The median partial pressure of oxygen on 100% inspired oxygen (PaO2:FiO2) at the end of EVLP was 516 (478–523) mmHg for transplanted and 342 (322–392) mmHg for non-transplanted donor lungs (Mann–Whitney U = 4, P = 0.02). The median PaO2:FiO2 improvement after EVLP was 321 (186–362) mmHg in transplanted donor lungs (Wilcoxon Z = −2.37, P = 0.03). No improvement was seen in lungs that failed assessment; the median PaO2:FiO2 change was 20 (−42–94) mmHg, (Wilcoxon Z = −0.18, P = 0.91) (Fig. 1). Figure 1: Physiological characteristics of transplanted and non-transplanted EVLP donor lungs. Box and whisker plots with boxes showing medians with interquartile range and whiskers at max and min. **P<0.05. EVLP; start time: point when the lung had been successfully rewarmed to 37 °C in the beginning of the EVLP; end time: point before start of cooling at the end of the EVLP.
figure	Figure 1: Physiological characteristics of transplanted and non-transplanted EVLP donor lungs. Box and whisker plots with boxes showing medians with interquartile range and whiskers at max and min. **P<0.05. EVLP; start time: point when the lung had been successfully rewarmed to 37 °C in the beginning of the EVLP; end time: point before start of cooling at the end of the EVLP.
label	Figure 1:
caption	Physiological characteristics of transplanted and non-transplanted EVLP donor lungs. Box and whisker plots with boxes showing medians with interquartile range and whiskers at max and min. **P<0.05. EVLP; start time: point when the lung had been successfully rewarmed to 37 °C in the beginning of the EVLP; end time: point before start of cooling at the end of the EVLP.
p	Physiological characteristics of transplanted and non-transplanted EVLP donor lungs. Box and whisker plots with boxes showing medians with interquartile range and whiskers at max and min. **P<0.05. EVLP; start time: point when the lung had been successfully rewarmed to 37 °C in the beginning of the EVLP; end time: point before start of cooling at the end of the EVLP.
p	Of the nine lungs that failed EVLP reconditioning, six displayed worsening pulmonary oedema and two showed signs of persistent hyperinflation and raised PVR, suggesting diffuse injury and a degree of intrinsic emphysema. This suggestion was later verified by a pathologist from pulmonary biopsies. Lastly, one set of lungs was turned down due to heavy microbial contamination with multiple gram-negative organisms identified in the tracheal aspirate.
sec	Transplant outcomes In the seven donor lungs that were transplanted, perfusion and ventilation parameters remained stable or improved during assessment. The data on changes in lung compliance, airway resistance, PVR, perfusate flow, and ventilation and perfusion pressures are shown in Fig. 1. None of the recipients required extracorporeal life support. Two patients (29%) suffered from PGD (Grade 3) at 72 h post-transplant. However, both patients responded well to standard management. All seven recipients survived to hospital discharge. One recipient died of severe pneumonia at 11 months post-transplant due to an influenza infection unrelated to the EVLP procedure. The other six (86%) remain alive and well >2 years after transplant (Table 2).
title	Transplant outcomes
p	In the seven donor lungs that were transplanted, perfusion and ventilation parameters remained stable or improved during assessment. The data on changes in lung compliance, airway resistance, PVR, perfusate flow, and ventilation and perfusion pressures are shown in Fig. 1.
p	None of the recipients required extracorporeal life support. Two patients (29%) suffered from PGD (Grade 3) at 72 h post-transplant. However, both patients responded well to standard management. All seven recipients survived to hospital discharge. One recipient died of severe pneumonia at 11 months post-transplant due to an influenza infection unrelated to the EVLP procedure. The other six (86%) remain alive and well >2 years after transplant (Table 2).
sec	Analysis of perfusate and bronchoalveolar lavage fluid Assessed lungs were divided into two groups based on decision after EVLP: transplanted (n = 7) or non-transplanted (n = 9). Decision on transplant suitability was based on strict criteria summarized in Supplementary Table S2. Samples were analysed retrospectively for protein expressions. For an initial general assessment of tissue injury, we measured LDH levels. LDH was found in substantially lower levels in perfusate from lungs meeting criteria for transplant (T) after EVLP compared to those that did not meet the criteria for transplant (NT) lungs. Suitability for transplant was evident after 2 h of perfusion: mean (T) = 0.150 U (SD 0.030, n = 14) and mean (NT) = 0.223 U (SD 0.082, n = 18) [t(30) = 2.77, P = 0.03] (Fig. 2). The LDH perfusate levels remained significantly different when assessed over all sample time points: mean (T) = 0.149 U (SD 0.033, n = 24) and mean (NT) = 0.259 U (SD 0.088, n = 40) [t(62) = 5.45, P < 0.001]. No difference in LDH levels was seen in BALF samples collected pre- and post-perfusion. Figure 2: Protein expressions during EVLP in perfusate from transplanted and non-transplanted donor lungs. Protein levels expressed as medians with interquartile ranges. *P<0.05. EVLP: ex vivo lung perfusion; LDH: lactate dehydrogenase; IL: interleukin; TNF-α: tumour necrosis factor alpha; HMGB-1: high-mobility group box-1; Time: hours of ex vivo lung perfusion after start of pulmonary artery perfusion. To assess the integrity of the pulmonary vascular compartment, we measured Syndecan-1 levels as a marker of endothelial glycocaylx disruption. Syndecan-1 levels were significantly lower in perfusate samples from transplanted donor lungs: mean (T) = 111 (pg/ml) (SD 98, n = 24) compared to mean (NT) = 281 (pg/ml) (SD 219, n = 40) [t(62) = 4.14, P < 0.001] (Fig. 3). Figure 3: Protein expressions in perfusate and BALF from transplanted and non-transplanted EVLP donor lungs. Interleaved scatter plots with molecular marker levels expressed in (pg/ml) and lines representing means. P<0.1. P<0.05. P<0.001. BALF: bronchoalveolar lavage fluid; EVLP: ex vivo lung perfusion; IL: interleukin; TNF-α: tumour necrosis factor alpha; LDH: lactate dehydrogenase; HMGB-1: high-mobility group box-1. We noted a similarly consistent pattern towards lower release of proinflammatory cytokines into the circulating perfusate from transplanted compared to non-transplanted lungs. The average level of IL-8 was nearly eight times higher in perfusate samples from non-transplanted lungs: mean (T) = 165 (pg/ml) (SD 155, n = 24) compared to mean (NT) = 1310 (pg/ml) (SD 1510, n = 35) [t(57) = 2.27, P = 0.06]. Levels of the anti-inflammatory cytokine IL-10 increased noticeably over the first 2 h of perfusion in the transplanted group, with high initial fold changes in both perfusate and BALF (Fig. 4). Perfusate levels then diminished and were less than half those in non-transplanted lungs when measured over the full assessment: mean (T) = 9 (pg/ml) (SD 11, n = 24) compared to mean (NT) = 23 (pg/ml) (SD 32, n = 40) [t(57) = 2.02, P = 0.07]. No difference in levels of IL-10 in BALF was seen between the groups. Figure 4: Protein median fold changes in perfusate and BALF from EVLP donor lungs. Bar graphs showing median fold changes with interquartile range. BALF: bronchoalveolar lavage fluid; EVLP: ex vivo lung perfusion; IL: interleukin; TNF-α: tumour necrosis factor alpha; LDH: lactate dehydrogenase; HMGB-1: high-mobility group box-1. A panel of three damage-associated molecular patterns, IL-33, HMGB-1 and S100A9, was used to assess the extent of cellular injury in the donor lung. Lower levels of IL-33 and HMGB-1 were found in perfusate from transplanted donor lungs. The average IL-33 levels in perfusate were mean (T) = 3 (pg/ml) (SD 5, n = 24) and mean (NT) = 9 (pg/ml) (SD 10, n = 40) [t(62) = 2.93, P = 0.02]. HMGB-1 was highly expressed in both perfusate and BALF. The HMGB-1 levels in perfusate were mean (T) = 8318 (pg/ml) (SD 4474, n = 24) and mean (NT) = 10 545 (pg/ml) (SD 4,313, n = 40) [t(62) = 2.10, P = 0.07]. No difference in lavage fluid levels of IL-33 or HMGB-1 was seen between the groups. We noted a consistent pattern towards increasing perfusate protein levels and a more pronounced separation between transplanted and non-transplanted lungs over time of perfusion (Fig. 2). This pattern was most noticeable for the investigated inflammatory cytokines. The only protein marker not showing this pattern was S100A9, which appeared lower in the non-transplanted group. A multivariate analysis model was fitted to assess the predictive value of combining two or more protein markers. The optimal model was established by leave-one-out cross-validation and is shown in Fig. 5 as a scatterplot of IL-1β and IL-8 perfusate levels from the 16 assessments. The shaded prediction region was derived from the regression analysis with covariates IL-1β, IL-8 and (IL-8)2 after 2 h of perfusion. The model suggests that the closer to the centre of the area, the higher was the probability of a donor lung being found suitable for transplant after EVLP assessment. This early IL-1β–IL-8 signature remained equally as robust when applied to the full set of samples (Fig. 5B). An ROC curve showing the potential benefit of the model in predicting EVLP outcome compared to random is shown in Fig. 5C, with a sensitivity of 0.81 and specificity of 0.92 at 2 h of perfusion. Figure 5: (A) Early IL-1β-IL-8 signature in EVLP perfusate after 2 h of perfusion. (B) IL-1β-IL-8 signature in EVLP perfusate after full assessment. (C) ROC curve. (A) and (B) Scatter plots of IL-1β and IL-8 levels in perfusate samples from transplanted and non-transplanted EVLP donor lungs. The shaded area is the prediction region from the optimal logistic regression model with covariates IL-1β, IL-8 and (IL-8)2. (C) Receiver operating characteristic (ROC) curve of the model as a predictor of EVLP outcome. Calculated sensitivity (y axis) is plotted against the 1-specificity formula (x axis) of the logistic regression function 2 h into perfusion and after full assessment. EVLP: ex vivo lung perfusion; IL: interleukin. The modelled proteins were lastly assessed for correlations to post-transplant outcome measures. Donor IL-1β levels in the perfusate at time of transplant decision showed a significant negative correlation to recipient PaO2:FiO2 24 h post-transplant (r = −093, P = 0.03) (Fig. 6). No other correlations remained significant after corrections for multiple testing in this cohort. Figure 6: Association between donor IL-1β level in EVLP perfusate at the time of transplant decision and recipient P/F ratio 24 h post lung transplantation. Pearson correlation plot. *P<0.05. EVLP: ex vivo lung perfusion; IL: interleukin; PaO2:FiO2: partial pressure of oxygen on 100% inspired oxygen.
title	Analysis of perfusate and bronchoalveolar lavage fluid
p	Assessed lungs were divided into two groups based on decision after EVLP: transplanted (n = 7) or non-transplanted (n = 9). Decision on transplant suitability was based on strict criteria summarized in Supplementary Table S2. Samples were analysed retrospectively for protein expressions.
p	For an initial general assessment of tissue injury, we measured LDH levels. LDH was found in substantially lower levels in perfusate from lungs meeting criteria for transplant (T) after EVLP compared to those that did not meet the criteria for transplant (NT) lungs. Suitability for transplant was evident after 2 h of perfusion: mean (T) = 0.150 U (SD 0.030, n = 14) and mean (NT) = 0.223 U (SD 0.082, n = 18) [t(30) = 2.77, P = 0.03] (Fig. 2). The LDH perfusate levels remained significantly different when assessed over all sample time points: mean (T) = 0.149 U (SD 0.033, n = 24) and mean (NT) = 0.259 U (SD 0.088, n = 40) [t(62) = 5.45, P < 0.001]. No difference in LDH levels was seen in BALF samples collected pre- and post-perfusion. Figure 2: Protein expressions during EVLP in perfusate from transplanted and non-transplanted donor lungs. Protein levels expressed as medians with interquartile ranges. **P<0.05. EVLP: ex vivo lung perfusion; LDH: lactate dehydrogenase; IL: interleukin; TNF-α: tumour necrosis factor alpha; HMGB-1: high-mobility group box-1; Time: hours of ex vivo lung perfusion after start of pulmonary artery perfusion.
figure	Figure 2: Protein expressions during EVLP in perfusate from transplanted and non-transplanted donor lungs. Protein levels expressed as medians with interquartile ranges. **P<0.05. EVLP: ex vivo lung perfusion; LDH: lactate dehydrogenase; IL: interleukin; TNF-α: tumour necrosis factor alpha; HMGB-1: high-mobility group box-1; Time: hours of ex vivo lung perfusion after start of pulmonary artery perfusion.
label	Figure 2:
caption	Protein expressions during EVLP in perfusate from transplanted and non-transplanted donor lungs. Protein levels expressed as medians with interquartile ranges. **P<0.05. EVLP: ex vivo lung perfusion; LDH: lactate dehydrogenase; IL: interleukin; TNF-α: tumour necrosis factor alpha; HMGB-1: high-mobility group box-1; Time: hours of ex vivo lung perfusion after start of pulmonary artery perfusion.
p	Protein expressions during EVLP in perfusate from transplanted and non-transplanted donor lungs. Protein levels expressed as medians with interquartile ranges. **P<0.05. EVLP: ex vivo lung perfusion; LDH: lactate dehydrogenase; IL: interleukin; TNF-α: tumour necrosis factor alpha; HMGB-1: high-mobility group box-1; Time: hours of ex vivo lung perfusion after start of pulmonary artery perfusion.
p	To assess the integrity of the pulmonary vascular compartment, we measured Syndecan-1 levels as a marker of endothelial glycocaylx disruption. Syndecan-1 levels were significantly lower in perfusate samples from transplanted donor lungs: mean (T) = 111 (pg/ml) (SD 98, n = 24) compared to mean (NT) = 281 (pg/ml) (SD 219, n = 40) [t(62) = 4.14, P < 0.001] (Fig. 3). Figure 3: Protein expressions in perfusate and BALF from transplanted and non-transplanted EVLP donor lungs. Interleaved scatter plots with molecular marker levels expressed in (pg/ml) and lines representing means. P<0.1. P<0.05. **P<0.001. BALF: bronchoalveolar lavage fluid; EVLP: ex vivo lung perfusion; IL: interleukin; TNF-α: tumour necrosis factor alpha; LDH: lactate dehydrogenase; HMGB-1: high-mobility group box-1.
figure	Figure 3: Protein expressions in perfusate and BALF from transplanted and non-transplanted EVLP donor lungs. Interleaved scatter plots with molecular marker levels expressed in (pg/ml) and lines representing means. P<0.1. P<0.05. **P<0.001. BALF: bronchoalveolar lavage fluid; EVLP: ex vivo lung perfusion; IL: interleukin; TNF-α: tumour necrosis factor alpha; LDH: lactate dehydrogenase; HMGB-1: high-mobility group box-1.
label	Figure 3:
caption	Protein expressions in perfusate and BALF from transplanted and non-transplanted EVLP donor lungs. Interleaved scatter plots with molecular marker levels expressed in (pg/ml) and lines representing means. P<0.1. P<0.05. **P<0.001. BALF: bronchoalveolar lavage fluid; EVLP: ex vivo lung perfusion; IL: interleukin; TNF-α: tumour necrosis factor alpha; LDH: lactate dehydrogenase; HMGB-1: high-mobility group box-1.
p	Protein expressions in perfusate and BALF from transplanted and non-transplanted EVLP donor lungs. Interleaved scatter plots with molecular marker levels expressed in (pg/ml) and lines representing means. P<0.1. P<0.05. **P<0.001. BALF: bronchoalveolar lavage fluid; EVLP: ex vivo lung perfusion; IL: interleukin; TNF-α: tumour necrosis factor alpha; LDH: lactate dehydrogenase; HMGB-1: high-mobility group box-1.
p	We noted a similarly consistent pattern towards lower release of proinflammatory cytokines into the circulating perfusate from transplanted compared to non-transplanted lungs. The average level of IL-8 was nearly eight times higher in perfusate samples from non-transplanted lungs: mean (T) = 165 (pg/ml) (SD 155, n = 24) compared to mean (NT) = 1310 (pg/ml) (SD 1510, n = 35) [t(57) = 2.27, P = 0.06]. Levels of the anti-inflammatory cytokine IL-10 increased noticeably over the first 2 h of perfusion in the transplanted group, with high initial fold changes in both perfusate and BALF (Fig. 4). Perfusate levels then diminished and were less than half those in non-transplanted lungs when measured over the full assessment: mean (T) = 9 (pg/ml) (SD 11, n = 24) compared to mean (NT) = 23 (pg/ml) (SD 32, n = 40) [t(57) = 2.02, P = 0.07]. No difference in levels of IL-10 in BALF was seen between the groups. Figure 4: Protein median fold changes in perfusate and BALF from EVLP donor lungs. Bar graphs showing median fold changes with interquartile range. BALF: bronchoalveolar lavage fluid; EVLP: ex vivo lung perfusion; IL: interleukin; TNF-α: tumour necrosis factor alpha; LDH: lactate dehydrogenase; HMGB-1: high-mobility group box-1.
figure	Figure 4: Protein median fold changes in perfusate and BALF from EVLP donor lungs. Bar graphs showing median fold changes with interquartile range. BALF: bronchoalveolar lavage fluid; EVLP: ex vivo lung perfusion; IL: interleukin; TNF-α: tumour necrosis factor alpha; LDH: lactate dehydrogenase; HMGB-1: high-mobility group box-1.
label	Figure 4:
caption	Protein median fold changes in perfusate and BALF from EVLP donor lungs. Bar graphs showing median fold changes with interquartile range. BALF: bronchoalveolar lavage fluid; EVLP: ex vivo lung perfusion; IL: interleukin; TNF-α: tumour necrosis factor alpha; LDH: lactate dehydrogenase; HMGB-1: high-mobility group box-1.
p	Protein median fold changes in perfusate and BALF from EVLP donor lungs. Bar graphs showing median fold changes with interquartile range. BALF: bronchoalveolar lavage fluid; EVLP: ex vivo lung perfusion; IL: interleukin; TNF-α: tumour necrosis factor alpha; LDH: lactate dehydrogenase; HMGB-1: high-mobility group box-1.
p	A panel of three damage-associated molecular patterns, IL-33, HMGB-1 and S100A9, was used to assess the extent of cellular injury in the donor lung. Lower levels of IL-33 and HMGB-1 were found in perfusate from transplanted donor lungs. The average IL-33 levels in perfusate were mean (T) = 3 (pg/ml) (SD 5, n = 24) and mean (NT) = 9 (pg/ml) (SD 10, n = 40) [t(62) = 2.93, P = 0.02]. HMGB-1 was highly expressed in both perfusate and BALF. The HMGB-1 levels in perfusate were mean (T) = 8318 (pg/ml) (SD 4474, n = 24) and mean (NT) = 10 545 (pg/ml) (SD 4,313, n = 40) [t(62) = 2.10, P = 0.07]. No difference in lavage fluid levels of IL-33 or HMGB-1 was seen between the groups.
p	We noted a consistent pattern towards increasing perfusate protein levels and a more pronounced separation between transplanted and non-transplanted lungs over time of perfusion (Fig. 2). This pattern was most noticeable for the investigated inflammatory cytokines. The only protein marker not showing this pattern was S100A9, which appeared lower in the non-transplanted group.
p	A multivariate analysis model was fitted to assess the predictive value of combining two or more protein markers. The optimal model was established by leave-one-out cross-validation and is shown in Fig. 5 as a scatterplot of IL-1β and IL-8 perfusate levels from the 16 assessments. The shaded prediction region was derived from the regression analysis with covariates IL-1β, IL-8 and (IL-8)2 after 2 h of perfusion. The model suggests that the closer to the centre of the area, the higher was the probability of a donor lung being found suitable for transplant after EVLP assessment. This early IL-1β–IL-8 signature remained equally as robust when applied to the full set of samples (Fig. 5B). An ROC curve showing the potential benefit of the model in predicting EVLP outcome compared to random is shown in Fig. 5C, with a sensitivity of 0.81 and specificity of 0.92 at 2 h of perfusion. Figure 5: (A) Early IL-1β-IL-8 signature in EVLP perfusate after 2 h of perfusion. (B) IL-1β-IL-8 signature in EVLP perfusate after full assessment. (C) ROC curve. (A) and (B) Scatter plots of IL-1β and IL-8 levels in perfusate samples from transplanted and non-transplanted EVLP donor lungs. The shaded area is the prediction region from the optimal logistic regression model with covariates IL-1β, IL-8 and (IL-8)2. (C) Receiver operating characteristic (ROC) curve of the model as a predictor of EVLP outcome. Calculated sensitivity (y axis) is plotted against the 1-specificity formula (x axis) of the logistic regression function 2 h into perfusion and after full assessment. EVLP: ex vivo lung perfusion; IL: interleukin.
figure	Figure 5: (A) Early IL-1β-IL-8 signature in EVLP perfusate after 2 h of perfusion. (B) IL-1β-IL-8 signature in EVLP perfusate after full assessment. (C) ROC curve. (A) and (B) Scatter plots of IL-1β and IL-8 levels in perfusate samples from transplanted and non-transplanted EVLP donor lungs. The shaded area is the prediction region from the optimal logistic regression model with covariates IL-1β, IL-8 and (IL-8)2. (C) Receiver operating characteristic (ROC) curve of the model as a predictor of EVLP outcome. Calculated sensitivity (y axis) is plotted against the 1-specificity formula (x axis) of the logistic regression function 2 h into perfusion and after full assessment. EVLP: ex vivo lung perfusion; IL: interleukin.
label	Figure 5:
caption	(A) Early IL-1β-IL-8 signature in EVLP perfusate after 2 h of perfusion. (B) IL-1β-IL-8 signature in EVLP perfusate after full assessment. (C) ROC curve. (A) and (B) Scatter plots of IL-1β and IL-8 levels in perfusate samples from transplanted and non-transplanted EVLP donor lungs. The shaded area is the prediction region from the optimal logistic regression model with covariates IL-1β, IL-8 and (IL-8)2. (C) Receiver operating characteristic (ROC) curve of the model as a predictor of EVLP outcome. Calculated sensitivity (y axis) is plotted against the 1-specificity formula (x axis) of the logistic regression function 2 h into perfusion and after full assessment. EVLP: ex vivo lung perfusion; IL: interleukin.
p	(A) Early IL-1β-IL-8 signature in EVLP perfusate after 2 h of perfusion. (B) IL-1β-IL-8 signature in EVLP perfusate after full assessment. (C) ROC curve. (A) and (B) Scatter plots of IL-1β and IL-8 levels in perfusate samples from transplanted and non-transplanted EVLP donor lungs. The shaded area is the prediction region from the optimal logistic regression model with covariates IL-1β, IL-8 and (IL-8)2. (C) Receiver operating characteristic (ROC) curve of the model as a predictor of EVLP outcome. Calculated sensitivity (y axis) is plotted against the 1-specificity formula (x axis) of the logistic regression function 2 h into perfusion and after full assessment.
p	EVLP: ex vivo lung perfusion; IL: interleukin.
p	The modelled proteins were lastly assessed for correlations to post-transplant outcome measures. Donor IL-1β levels in the perfusate at time of transplant decision showed a significant negative correlation to recipient PaO2:FiO2 24 h post-transplant (r = −093, P = 0.03) (Fig. 6). No other correlations remained significant after corrections for multiple testing in this cohort. Figure 6: Association between donor IL-1β level in EVLP perfusate at the time of transplant decision and recipient P/F ratio 24 h post lung transplantation. Pearson correlation plot. **P<0.05. EVLP: ex vivo lung perfusion; IL: interleukin; PaO2:FiO2: partial pressure of oxygen on 100% inspired oxygen.
figure	Figure 6: Association between donor IL-1β level in EVLP perfusate at the time of transplant decision and recipient P/F ratio 24 h post lung transplantation. Pearson correlation plot. **P<0.05. EVLP: ex vivo lung perfusion; IL: interleukin; PaO2:FiO2: partial pressure of oxygen on 100% inspired oxygen.
label	Figure 6:
caption	Association between donor IL-1β level in EVLP perfusate at the time of transplant decision and recipient P/F ratio 24 h post lung transplantation. Pearson correlation plot. **P<0.05. EVLP: ex vivo lung perfusion; IL: interleukin; PaO2:FiO2: partial pressure of oxygen on 100% inspired oxygen.
p	Association between donor IL-1β level in EVLP perfusate at the time of transplant decision and recipient P/F ratio 24 h post lung transplantation. Pearson correlation plot. **P<0.05. EVLP: ex vivo lung perfusion; IL: interleukin; PaO2:FiO2: partial pressure of oxygen on 100% inspired oxygen.
sec	DISCUSSION In this proof-of-concept study, the feasibility of using inflammation and tissue injury markers in both perfusate and BALF from human donor lungs undergoing EVLP to validate the decision to use or decline them for clinical transplant was evaluated. All 16 donor lungs in this study were assessed consecutively with intent for transplant following a strict protocol with predefined criteria for EVLP and transplant suitability to reduce the risk of bias in decision making. All protein analyses were done retrospectively and had no impact on the transplant decision. Our results demonstrate a difference in protein expression in perfusate from lungs that were transplanted compared with those that were declined. This difference became more apparent for each hour of perfusion, and BALF samples showed a comparable but weaker signal from the airway compartment. The difference in donor lungs was effectively demonstrated by a logistic regression model combining inflammatory cytokines IL-1β and IL-8 after 2 h of perfusion. The model revealed a clear separation of the transplanted lungs from those declined. Importantly, the model remained equally as robust when applied to perfusate samples from all time points. Furthermore, perfusate IL-1β concentrations in transplanted lungs demonstrated a clear negative correlation to early recipient oxygenation post-transplant. Even though our results are noteworthy, overinterpretation of this regression model should be avoided at this stage. Further investigations are required to assess its biological plausibility. A few declined lungs displayed a very low IL-1β/IL-8 signal—below that of lungs deemed reconditioned and transplanted with good outcomes after the EVLP assessment. If this finding indicates that those lungs were wrongfully declined and could safely have been transplanted, that EVLP-assessed lungs deemed suitable for transplant truly express an intermediate level of inflammation, or that this is a signal specific to this relatively small cohort needs further assessment in follow-up studies. The measurement of LDH levels in perfusate had the potential to significantly predict EVLP outcome as early as 2 h into perfusion. Although real-time ligand binding assays for inflammatory cytokines and other tissue injury markers may soon become accessible [22], LDH has the benefit of being a widely recognized marker of tissue injury in serum and BALF and rapidly available as a point-of-care test [23, 24]. The Brisbane lung transplant group have described measures of endothelial dysfunction as potentially useful in EVLP [25, 26]. Our results strengthen that belief. Syndecan-1, released by disruption of the endothelial glycocalyx, may be a discriminatory marker in EVLP perfusate and should be further evaluated. Increasing levels of inflammatory cytokines in perfusate and tissue have been shown during experimental lung perfusion in porcine models and a small preclinical human lung study, which might reflect procedure-mediated inflammation [27–29]. Our findings during full clinical EVLP support recent observations made by Machuca and colleagues [30] from Toronto and demonstrate the feasibility of identifying biomarkers that might improve EVLP assessment. IL-8, proposed as one of the best markers of EVLP transplant outcome in the Machuca study, has consistently shown potential in previous studies of donor lung injury from our group [12, 13]. The potential of IL-8 to discriminate successful EVLP in our study was enhanced by combining it with IL-1β. Our observations add significantly to previous observations because we have included an assessment of BALF inflammatory and tissue injury markers compared to EVLP perfusate measures. Even though BALF samples in this study did not add any apparent clinical value to perfusate sampling, we feel it is too early to draw any firm conclusions, and we will continue to assess a broad spectrum of markers and lung compartment samples in future studies. In addition, because protein levels are measured in a fixed volume of 2 litres of perfusate solution, we corrected protein levels for the size of lung tissue perfused. This potential confounder has been disregarded in previous studies. The pTLC of the donor lung ranged from 4.31 to 7.54 litres among the double lungs perfused, and the smallest single lung had a pTLC of only 2.70 litres. Unadjusted marker levels are therefore likely to differ widely between donor lungs regardless of the degree of lung injury present. A limitation of our study is its relatively small sample size, with 16 human EVLP assessments included, yet this number is sufficient to demonstrate the feasibility of using perfusate markers to discriminate successful EVLP. Outcomes were almost universally good, and this study does not identify predictors of early lung injury. Larger studies, with a cohort of patients doing less well, are needed to address that particular question. Caution is, however, required in overinterpretation of moderate to severe PGD (Grade 2–3) after transplant as a reason for donor lungs to be declined after EVLP, because many recipients with PGD 2–3 will recover and have satisfactory early outcomes. Truly marginal donor lungs subjected to EVLP for reconditioning purposes form by nature a subpopulation that is likely to have a higher inflammatory burden than standard donor lungs. We believe that biomarker profiles of these lungs could help support the clinical decision to use or decline the donor lungs after EVLP and allow continued safe transplant activity. Furthermore, this approach offers the potential to attenuate the inflammatory response or encourage recovery of vascular integrity in the donor lung before implantation in a stratified medicine approach, which may improve early outcomes after lung transplant and help to safely maximize lung use from the existing donor pool. The results of this study need further evaluation in a larger validation cohort of lungs exposed to ex vivo perfusion but demonstrate the feasibility of identifying an early perfusate signature potentially predictive of successful EVLP reconditioning.
title	DISCUSSION
p	In this proof-of-concept study, the feasibility of using inflammation and tissue injury markers in both perfusate and BALF from human donor lungs undergoing EVLP to validate the decision to use or decline them for clinical transplant was evaluated. All 16 donor lungs in this study were assessed consecutively with intent for transplant following a strict protocol with predefined criteria for EVLP and transplant suitability to reduce the risk of bias in decision making. All protein analyses were done retrospectively and had no impact on the transplant decision. Our results demonstrate a difference in protein expression in perfusate from lungs that were transplanted compared with those that were declined. This difference became more apparent for each hour of perfusion, and BALF samples showed a comparable but weaker signal from the airway compartment.
p	The difference in donor lungs was effectively demonstrated by a logistic regression model combining inflammatory cytokines IL-1β and IL-8 after 2 h of perfusion. The model revealed a clear separation of the transplanted lungs from those declined. Importantly, the model remained equally as robust when applied to perfusate samples from all time points. Furthermore, perfusate IL-1β concentrations in transplanted lungs demonstrated a clear negative correlation to early recipient oxygenation post-transplant.
p	Even though our results are noteworthy, overinterpretation of this regression model should be avoided at this stage. Further investigations are required to assess its biological plausibility. A few declined lungs displayed a very low IL-1β/IL-8 signal—below that of lungs deemed reconditioned and transplanted with good outcomes after the EVLP assessment. If this finding indicates that those lungs were wrongfully declined and could safely have been transplanted, that EVLP-assessed lungs deemed suitable for transplant truly express an intermediate level of inflammation, or that this is a signal specific to this relatively small cohort needs further assessment in follow-up studies.
p	The measurement of LDH levels in perfusate had the potential to significantly predict EVLP outcome as early as 2 h into perfusion. Although real-time ligand binding assays for inflammatory cytokines and other tissue injury markers may soon become accessible [22], LDH has the benefit of being a widely recognized marker of tissue injury in serum and BALF and rapidly available as a point-of-care test [23, 24].
p	The Brisbane lung transplant group have described measures of endothelial dysfunction as potentially useful in EVLP [25, 26]. Our results strengthen that belief. Syndecan-1, released by disruption of the endothelial glycocalyx, may be a discriminatory marker in EVLP perfusate and should be further evaluated.
p	Increasing levels of inflammatory cytokines in perfusate and tissue have been shown during experimental lung perfusion in porcine models and a small preclinical human lung study, which might reflect procedure-mediated inflammation [27–29]. Our findings during full clinical EVLP support recent observations made by Machuca and colleagues [30] from Toronto and demonstrate the feasibility of identifying biomarkers that might improve EVLP assessment.
p	IL-8, proposed as one of the best markers of EVLP transplant outcome in the Machuca study, has consistently shown potential in previous studies of donor lung injury from our group [12, 13]. The potential of IL-8 to discriminate successful EVLP in our study was enhanced by combining it with IL-1β.
p	Our observations add significantly to previous observations because we have included an assessment of BALF inflammatory and tissue injury markers compared to EVLP perfusate measures. Even though BALF samples in this study did not add any apparent clinical value to perfusate sampling, we feel it is too early to draw any firm conclusions, and we will continue to assess a broad spectrum of markers and lung compartment samples in future studies. In addition, because protein levels are measured in a fixed volume of 2 litres of perfusate solution, we corrected protein levels for the size of lung tissue perfused. This potential confounder has been disregarded in previous studies. The pTLC of the donor lung ranged from 4.31 to 7.54 litres among the double lungs perfused, and the smallest single lung had a pTLC of only 2.70 litres. Unadjusted marker levels are therefore likely to differ widely between donor lungs regardless of the degree of lung injury present.
p	A limitation of our study is its relatively small sample size, with 16 human EVLP assessments included, yet this number is sufficient to demonstrate the feasibility of using perfusate markers to discriminate successful EVLP. Outcomes were almost universally good, and this study does not identify predictors of early lung injury. Larger studies, with a cohort of patients doing less well, are needed to address that particular question. Caution is, however, required in overinterpretation of moderate to severe PGD (Grade 2–3) after transplant as a reason for donor lungs to be declined after EVLP, because many recipients with PGD 2–3 will recover and have satisfactory early outcomes.
p	Truly marginal donor lungs subjected to EVLP for reconditioning purposes form by nature a subpopulation that is likely to have a higher inflammatory burden than standard donor lungs. We believe that biomarker profiles of these lungs could help support the clinical decision to use or decline the donor lungs after EVLP and allow continued safe transplant activity. Furthermore, this approach offers the potential to attenuate the inflammatory response or encourage recovery of vascular integrity in the donor lung before implantation in a stratified medicine approach, which may improve early outcomes after lung transplant and help to safely maximize lung use from the existing donor pool.
p	The results of this study need further evaluation in a larger validation cohort of lungs exposed to ex vivo perfusion but demonstrate the feasibility of identifying an early perfusate signature potentially predictive of successful EVLP reconditioning.
sec	SUPPLEMENTARY MATERIAL Supplementary material (Tables 1 and 2) is available at EJCTS online.
title	SUPPLEMENTARY MATERIAL
p	Supplementary material (Tables 1 and 2) is available at EJCTS online.
sec	Fundi ng This work was also partly funded by the National Institute for Health Research Blood and Transplant Research Unit (NIHR BTRU) in Organ Donation and Transplantation at the University of Cambridge in collaboration with Newcastle University and in partnership with NHS Blood and Transplant (NHSBT). The views expressed are those of the author(s) and not necessarily those of the NHS, the NIHR, the Department of Health or NHSBT. The funding organizations had no role in the collection of data, its analysis or interpretation, and had no right to approve or disapprove publication of the finished manuscript. This study was supported by a research grant awarded by the United Kingdom Cystic Fibrosis Trust as a result of a donation made by the Robert Luff Foundation for research on lung transplants. Conflict of interest: none declared.
title	Fundi ng
p	This work was also partly funded by the National Institute for Health Research Blood and Transplant Research Unit (NIHR BTRU) in Organ Donation and Transplantation at the University of Cambridge in collaboration with Newcastle University and in partnership with NHS Blood and Transplant (NHSBT). The views expressed are those of the author(s) and not necessarily those of the NHS, the NIHR, the Department of Health or NHSBT. The funding organizations had no role in the collection of data, its analysis or interpretation, and had no right to approve or disapprove publication of the finished manuscript. This study was supported by a research grant awarded by the United Kingdom Cystic Fibrosis Trust as a result of a donation made by the Robert Luff Foundation for research on lung transplants.
p	Conflict of interest: none declared.
sec	Supplementary Material Supplementary Data Click here for additional data file.
title	Supplementary Material
label	Supplementary Data
caption	Click here for additional data file.
p	Click here for additional data file.

projects that have annotations to this span

Unselected / annnotation		Selected / annnotation
testtesttest (1)

TAB JSON ListView MergeView

PMC:5400024 / 9870-10005 JSONTXT

Document structure show

projects that have annotations to this span

PMC:5400024 / 9870-10005 JSON TXT