The role of ex vivo lung perfusion in lung transplantation

Abstract

Whilst lung transplantation is a viable solution for end-stage lung disease, donor shortages, donor lung inflammation and perioperative lung injury remain major limitations. Ex vivo lung perfusion has emerged as the next frontier in lung transplantation to address and overcome these limitations, with multicentre clinical trials ongoing in the UK, rest of Europe and North America. Our research seeks to identify the poorly understood cellular and molecular mechanisms of primary graft dysfunction through the development of an isolated perfused lung model of transplantation and investigation of the role of pulmonary inflammation in this paradigm.

Introduction

Advances in modern medicine have vastly improved the morbidity and mortality of chronic lung disease. However, transplantation remains the definitive treatment when other therapeutic avenues are no longer effective. The lung transplantation process is limited by numerous factors, which can be broadly divided into three categories. Availability of organs is the primary hurdle, with demand far outstripping supply. Variable graft quality is the next challenge with up to 80% of organs being rejected at the time of harvest.¹ Finally, primary graft dysfunction (PGD) in the recipient contributes to considerable morbidity and mortality.²

Organ availability

Lung transplantation remains the gold standard therapy for end-stage lung disease from a variety of aetiologies, most commonly chronic obstructive pulmonary disease, cystic fibrosis and pulmonary fibrosis. However, its success is limited by numerous factors, primarily organ availability. Despite significant efforts to expand the donor organ pool, availability is still exceeded greatly by demand, in part due to poor donation consent rates and previously inconsistent organ assessment and donor management programmes.

Organ quality

Once donors become available, graft quality is very variable as the lung is highly susceptible to injury in the last days of life. There is inevitable (for example in intensive care) exposure to events such as aspiration, ventilator-associated lung injury and pneumonia, sepsis and brainstem death-related pulmonary oedema. As a result, up to 80% of potential lungs are deemed unsuitable at the point of harvest, significantly reducing the number ultimately deemed appropriate for transplantation.¹

Morbidity and mortality post-transplant

Of those patients receiving a graft, significant proportions develop PGD and many die as a result.² Furthermore, the average length of survival of a successful lung recipient is only five years, primarily due to the development of bronchiolitis obliterans. This is significantly less than other transplanted organs (e.g. heart, renal, liver), which survive on average 10 years.³

PGD is a form of acute lung injury (ALI) occurring within 72 h of lung transplantation. It presents with acute hypoxia, bilateral pulmonary infiltrates and impaired pulmonary compliance (without alternate cause), accompanied by diffuse alveolar damage microscopically, not dissimilar to the criteria ascribed to ALI and acute respiratory distress syndrome.⁴ Previous terms for it allude to its potential mechanisms including ‘re-implantation oedema’, ‘ischaemia-reperfusion injury’ and ‘early graft dysfunction’. PGD is thought to occur in 10–25% of patients and is the leading cause of death (>40%) at 30 days post-transplantation.^2,5 A variety of factors have been implicated in the pathogenesis of PGD, which can be broadly divided into pre-transplant (i.e. donor factors) and peri-transplant (i.e. ischaemia-reperfusion (IR)).

A large amount of research into the pathogenesis of PGD focuses on the effect of the peri-transplantation exposure of lungs to IR. A combination of mechanisms for this has been postulated, and considerable effort has been made to ameliorate any injury at this stage. Preventative measures include the use of preserving solutions such as Perfadex, storage at 4℃ and more recently devices such as the Organ Care System, which enable effective lung assessment, preservation and transportation.⁶

However, pre-transplant donor factors are now receiving increasing attention in the study of PGD pathogenesis. Donor lungs are exposed to a variety of insults, inevitable to the circumstances of becoming a donor. These factors are thought to contribute to, at one extreme, the lungs being deemed unsuitable for transplantation, and at the other, an undetectable subclinical injury that may manifest itself as PGD once in the recipient. Suggested causes include ventilator-associated lung injury and pneumonia, sepsis, acid aspiration and the autonomic storm, hypertension, neurogenic pulmonary oedema and systemic inflammation, manifested secondary to post-traumatic brain injury.⁷ These may induce, or contribute to, undetected donor lung inflammation, which is likely to be a previously unrecognised principle determinant in the development of PGD.⁸

IR injury

Lung ischaemia is not only a major feature in the lung transplantation paradigm but also relevant during other low pulmonary perfusion states including cardiopulmonary bypass, severe hypotension, circulatory arrest and pulmonary embolism.⁹ Following reperfusion, injury is manifested by microvascular and alveolar damage, hypoxaemia and pulmonary oedema.⁹ Ischaemia, taken from the Greek to restrict blood, is a unique process in the lungs. In other organs, ischaemia leads inevitably to hypoxia, with reoxygenation likewise occurring at reperfusion. However, the lung parenchymal oxygen supply is derived from its dual blood supply (bronchial and pulmonary) and oxygen derived from the alveolar space directly, with the latter maintained during the transplantation process. Thus, in contrast to solid organs, it has been shown that adenosine triphosphate levels can remain normal in a model of ventilated ischaemia.¹⁰ However, regardless of oxygen availability, significant reduction or absence of blood flow and resultant lack of vascular mechanotransduction has been reported to induce the generation of reactive oxygen species (ROS) in endothelia, macrophages and other immune cells.^9,11 Oxidants are produced by both ventilated and hypoxic models of lung ischaemia, although predictably only occurring on reintroduction of oxygen in the latter case.¹² These highly reactive and unstable ROS interact with an array of molecules, most commonly the lipid components of cell membranes, inducing cell lysis and permeability changes.⁸ In addition to inducing ROS production in endothelial cells, loss of endothelial shear stress also stimulates nitric oxide synthase, nuclear factor kappa B and cell-adhesion molecule upregulation.^8,11,13,14

Leukocytes play a prominent role in IR, enhancing inflammation and injury. In addition to the resident inflammatory cell populations of the lungs (e.g. interstitial/alveolar macrophages, dendritic cells), there is also a rapidly recruitable population of intravascular leukocytes. This is due to the narrow diameter of the pulmonary capillaries where leukocytes readily sequester when activated, and remain, initially due to loss of deformability. Furthermore, IR-induced upregulation of pulmonary endothelial cell adhesion molecule expression may facilitate leukocyte extravasation, contributing to microvascular injury,¹⁵ with selective blockade of associated molecules (platelet-activating factor, ICAM1, CD18 and P-selectin) shown to confer protection to IR.^15–18 Pro-inflammatory cytokines in bronchoalveoalar lavage fluid (BALF) in animal models of lung transplantation (tumor necrosis factor (TNF), interleukin 2 (IL-2) and interferon γ (IFN-γ))¹⁹ and in human lung tissue supernatants (TNF, IL-8, -10, -12 and -18 and IFN-γ) during ischaemia and reperfusion have been documented.²⁰ Moreover, real-time polymerase chain reaction to quantify mRNA in donor lungs prior to transplant identified a similar pattern of cytokine expression associated with survival post-transplantation. This technique demonstrated that raised levels of TNF, IL6, IL8 and IL1β were associated with greater risk of mortality post-transplantation, whilst IL10 and IFNγ were protective,²¹ thus potentially enabling identification of subclinically inflamed lungs. IL8 levels in BALF and lung tissue have also been negatively correlated to graft function and outcome post-transplantation.^20,22

Leukocyte recruitment and activation within the pulmonary vasculature is key to the pathogenesis of ALI. Neutrophils are regarded as the central players and likely culprits for transplant-associated ALI, evidenced by their increased margination, activation and migration into the lung alveoli in animal models of brain injury.^23,24 Furthermore, high levels of the neutrophil chemoattractant IL-8 in the BALF of human lung transplants have been shown to correlate with incidence of PGD and survival.²² However, plasma monocyte chemoattractant protein 1 has also been found to be significantly raised in patients with PGD, implicating a specific role for monocytes in this process.²⁵

Improving organ availability: Ex vivo lung perfusion

As transplant waiting lists greatly exceed organ donors, it is essential to maximise donor lung availability by improving early recognition of donor lung injury and developing novel strategies to reduce it. Varied attempts to address organ shortage are ongoing, one of which is the active management of cadaveric donors. In recent years, protocolised regimes including resuscitation, aggressive fluid management and invasive monitoring have been shown to successfully improve organ quality, acceptance rate and thus numbers available for transplant.²⁶ Furthermore, non-heart beating donors are now also being increasingly utilised as a further source of potential lungs grafts.²⁷

Complementing such optmisation procedures in the donor is the recent advent of the post-harvest strategy for maximising lung availability, termed ex vivo lung perfusion (EVLP), which is now becoming widely adopted. EVLP is a novel method of assessing, recovering and repairing injured donor lungs deemed borderline or unsuitable for transplantation.^28–31 EVLP has also provided an enhanced ability to investigate the pathological mechanisms associated with PGD.

EVLP represents just one recent step in the advancement of the lung transplantation process over the last century. In the 1930s, the successful ex vivo preservation of whole organs was first described by Carrel and Lindbergh³² after their invention of a ‘perfusion pump’. Following on from this, the surgical technique of organ transplantation was first attempted in animals in the 1940s and 1950s. Whilst survival was initially severely limited by organ rejection, the first human lung was transplanted only a decade later in 1963 by James Hardy in Mississippi, USA.³³ The recipient is reported to have only survived for 18 days.³⁴ It was not until 20 years later, assisted by improved surgical techniques, the use of cardiopulmonary bypass and immunosuppressive agents that the technique was regarded as successful. Since then, survival rates have increased remarkably to 83% at one year and 54% at five years in the USA, where ∼1800 lung transplants were undertaken in 2011.³⁵

Professor Steen of Lund University, Sweden, initially pioneered EVLP, for the purpose of assessment of lung function. His group established the technique whereby borderline or even rejected lungs could be assessed and potentially reconditioned for transplantation. This technique allows monitoring, functional assessment and crucially, therapeutic interventions to occur in the isolated organ, proving especially relevant to lungs sourced from non-heart beating donors.³⁶ Preliminary work suggests that this strategy might include clearance of inflammatory intravascular and intraalveolar leukocytes, treatment of pulmonary oedema and pulmonary emboli, infection and aspiration-pneumonitis.^37–41 Another potential mechanism for repair during EVLP is the application of gene therapy, specifically anti-inflammatory agents and stem cell therapy.^42,43

Steen’s group went on to report the use of EVLP to enable the successful transplantation of a human lung from a previously unsuitable donor in 2007.⁴⁴ Several groups have now successfully transplanted EVLP-reconditioned lungs worldwide and several pro- and retrospective clinical trials are ongoing, producing encouraging results.

In recent years, Toronto Hospital, home of the first successful lung transplants in the 1960s, has fostered a successful EVLP programme and contributed significantly to development of the technique. They adapted an acellular perfusate for their experimental and clinical studies and have demonstrated excellent results with this method of EVLP.²⁹ Moreover, they established that maintaining the organs at 37℃ in the EVLP system confers improved lung function as compared to standard cold ischaemic storage.⁴⁵

In addition, this group published the first prospective clinical trial comparing 20 high-risk lungs accepted for EVLP versus 116 standard lung transplants. They were able to clearly show a significant improvement of PO₂:FiO₂ ratio (335 mmHg to 443 mmHg, p < 0.001) over the 4 h of EVLP in the high-risk lungs. Following transplantation of the EVLP lungs, they reported an incidence of grade 2 or 3 PGD of 15% (vs. 30% in control lungs, p = 0.11). Crucially, they also showed no difference between the two groups of lung recipients in intensive care unit (ICU) stay, hospital stay, airway complications or 30-day mortality.⁴⁶

To date, this group have now performed over 50 EVLP procedures, and their method has now been adopted as part of a large multi-centre trial in the USA (the NOVEL Lung Trial).⁴⁷ At this year’s International Society of Heart and Lung Transplantation (ISHLT) meeting, the NOVEL group presented their one-year outcomes of 42 EVLP versus 42 standard criteria lungs. They report no significant difference in outcomes between these groups, including PGD at any stage, ICU and hospital stay, and 30-day and one-year survival.⁴⁸

In addition, the EVLP groups from Toronto, Vienna and Paris presented their combined four-year data at the ISHLT meeting in 2013.⁴⁹ Collectively they performed 125 EVLPs of which 103 went on to be transplanted. Of these, they report a 4% 30-day mortality and 85% survival at 12 months.

These results have also been replicated throughout Europe. The EVLP group in Italy, led by Gattinoni, has also recently published their experience from 2011 to 2013.⁵⁰ Of 36 sets of lungs from brain death donors during this period, seven sets (with initial PO₂:FiO₂ ratios <300 mmHg) were placed on EVLP and were successfully transplanted. Even though lung allocation scores were significantly greater in these lungs, post-transplant PGD, 30-day mortality and survival were not significantly different to standard transplant patients.

The Gothenburg group described similar results with 11 patients who received single or double transplants from EVLP-treated lungs over an 18-month period. Again no significant difference was reported in PGD when compared to standard lung transplant patients at 72 h and with all EVLP patients being alive at hospital discharge.⁵¹

There are also promising single-centre reports from the UK. The Harefield group recently described successful transplantation of six pairs of EVLP lungs over a two-year period. They found no differences in 30-day, three-month and six-month survival, in comparison to their standard transplant patients.³¹ The Newcastle team has also reported six successful EVLP transplants, with all surviving except one, who died 11 months post-procedure of unrelated influenza infection.⁵²

Beyond these single-centre experiences, all UK lung transplant centres have introduced EVLP to their practice and are part of the unique DEVELOP-UK prospective multicentre trial, which is designed to assess the clinical and cost effectiveness of this therapy in the UK setting.

EVLP represents the next frontier in transplant medicine. It has the potential to simultaneously address and overcome the numerous limiting factors that up until now have significantly limited the success of lung transplantation. Ongoing research and development of this promising technique will likely play a central role in the future of lung transplant medicine, ultimately improving the mortality and morbidity of those patients for whom transplantation has become their last resort.

Current research

Our own research work focuses on identification of the cellular and molecular mechanisms responsible for PGD, which remain poorly understood. We have previously shown that subclinical endotoxaemia induces monocyte margination and subsequent activation within the pulmonary vasculature, contributing to the development of acute lung injury in animal models (microbial challenge, ventilator-induced lung injury).^53–55 We therefore hypothesised that endotoxaemia-induced margination of inflammatory leukocytes in donor lungs could enhance the response to IR, exacerbating an acute lung injury in the recipient. To this end, we have developed a murine-isolated perfused lung model to investigate the effect of subclinical endotoxaemia and the development of IR injury. Thus far, this work has allowed us to map leukocyte pulmonary margination and retention, activation status and the production of inflammatory chemokines and cytokines, and their relationship to development of pulmonary injury. Our preliminary findings confirm the importance of latent pulmonary inflammation on enhancement of the IR-induced inflammatory response and suggest that margination and retention of inflammatory leukocytes, including monocytes, could be a key determinant in PGD development post-transplantation.^56,57 Such model-based findings may be highly relevant to further optimisation and refinement of the existing EVLP-based strategies.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Dr K Tatham is supported by funding from the Wellcome Trust. Dr N Marczin is co-applicant on the NIHR multicentre EVLP (DEVELOP-UK) trial and has been supported by NIHR, the Cystic Fibrosis Trust and Royal Brompton and Harefield NHS Foundation Trust.