Lung Transplantation State Of The Art Biology Essay

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

The first human single lung transplant dates back to 1963. Although surgery went well, the recipient died on the 18th postoperative day from renal failure. In 1971, Derom et al from the University of Ghent reported the first single LTx successful in the medium-term, namely the 10-month survival of a 23-year old man who underwent a right LTx. During the postoperative period, the patient recovered completely from two major acute rejection crises, but ultimately died because of a pneumonia which, possibly, was superimposed on a chronic rejection process (2). As will be discussed later on, it was already clear that the enduring exposure of the graft to the external environment would constitute an additional obstacle. In 1976, the immunosuppressive properties of cyclosporin were discovered, which announced the modern era of solid organ transplantation. Cyclosporin is a fungal polypeptide that inhibits calcineurin which is involved in T-cell activation and IL-2 production (3). This made it possible for the Toronto Group to make a major leap forward. In 1983, after 12 years of no medium-term survivors, they performed a landmark single LTx in establishing this procedure as an accepted last resort treatment for a variety of end-stage pulmonary diseases: the recipient, a 58-year-old male with end-stage pulmonary fibrosis, was discharged six weeks after LTx and returned to normal life. He survived eight years with normal pulmonary function before dying from renal failure. (4). With this case as a precedent, remarkable progress has been achieved. Moreover, since 1990, with the advent of the bilateral sequential technique, i.e. a sequential implantation technique through a transverse thoracosternotomy incision, double LTx has gained popularity and from the early 2000s onwards, it began to emerge as more common than single LTx (1;5). Besides, in a retrospective publication dating from 1988, Reitz et al from Stanford University report 250 successful heart-lung transplantations performed in their group since 1981, with the two longest living recipients surviving over 5 years post-transplantation. However, a high operative mortality of 25% due to multisystem failure is mentioned (6). Nevertheless, despite these historical successes, (heart-)lung transplantation still experiences worse survival compared to transplantation of other solid organs (1-year survival for e.g. kidney transplantation is 95-97% vs. 79% for LTx) (7). As such, LTx remains to be the subject of extensive research.

State of affairs in Belgium

Belgium is one of the world leaders in terms of number of deceased organ donors. This is because our country uses the so-called opt-out legislative system for determining voluntary consent, which means that anyone who has not refused is a donor. Concretely, the lung recovery rate is about 35% and with 8.3 LTx per million population per year, Belgium is amongst the world leaders (8). The procedure is currently performed in 4 university hospitals, namely in Brussels, Antwerp and Leuven, which is front runner with over 700 LTx's from 1991 up to now. Around 450 recipients are at present still alive. In 2012, X LTx's were performed at the Leuven Lung Transplant Unit, which makes it among the top 10 of world's largest LTx centers.


The majority of adult LTx's are performed for emphysema due to chronic obstructive pulmonary disease (COPD) (34%) or α1-antitrypsin deficiency (A1ATD) (6%). Moreover, provided that patients suffering from emphysema, which is a very common condition, get maximal medical care, they survive for a long time on the transplant waiting list. Idiopathic lung disease (ILD) accounts for 23% of all adult LTx's and cystic fibrosis (CF) for 17%. A wide variety of miscellaneous diseases comprise the remaining percentage of conditions requiring LTx (9;10).

It is noteworthy that the age of lung donors has increased over the years, reaching an average of 38.0 years, according to the 2012 report of the ISHLT registry. This may lead to a higher total number of LTx. Likewise, the mean age of recipients continues to increase as well, with a mean of 56 years and a permanently growing percentage of recipients aged older than 65 (9).

Furthermore, the experience of the Leuven Lung Transplant Unit learns that 1% of LTx's performed, concern retransplantations, with aantal dit jaar/ sinds de start vn LTx in leuven/ op het total aantal LTx's. These data show that retransplantation is a rather uncommon procedure.

Congenital heart disease, pulmonary arterial hypertension (PAH) and CF are the main indication for adult heart-lung transplantations. Compared to the 3519 LTx's performed in 2010 that were reported to the International Society for Heart and Lung Transplantation (ISHLT) Registry, heart-lung transplantation is still rather uncommon with 94 reported procedures in 2010 (9).


Overall 1-, 3-, 5- and 10-year survival among lung transplant recipients is 79, 64, 53 and 30% respectively. In ISHLT experience, survival depends on several factors:

Due to the continuous advancements in the field of lung transplantation, overall survival has consistently improved by era.

Late survival is superior in bilateral transplant recipients versus recipients of a single lung. In contrast with the early days of LTx when mainly single LTx's were performed for the sake of technical feasibility, the proportion of bilateral transplant procedures has risen overall for each of the four major indications for LTx (COPD, ILD, A1ATD and PAH) since 1994. Like this, bilateral transplantation accounted for 74% of transplant procedures in 2010 across all groups and diagnoses. In addition, it is believed that early postoperative management is made much simpler in double transplant recipients.

Survival rates differ significantly by recipient age, with short and long-term survival rates being lower among recipients in age groups >49 years. Long-term survival seems to decrease the survival rates in patients >65: 1-year survival is 73% in recipients >65 compared with 79% for those <50, whereas 5-year survival is 39% in those aged >65 vs. 51 to 56% in those <50. Moreover, these age-related effects may actually be more prominent after adjustment for era, because most transplants in older patients were done in more recent eras and survival has continuously been improving.

In ISHLT Registry experience, there is a significant difference in survival by transplant indication. Among patients surviving at least 1 year, diagnoses of CF and AATD emphysema had significantly better survival at 5 and 10 years after transplantation compared to COPD and ILD, most likely due to the older age of patients with more comorbidities among COPD and ILD patients.

Regarding the quality of life in recipients, an improvement is seen after the transplant and usually becomes evident after 3-6 months. Mobility, energy, sleep, activities of daily living dependency level and dyspnea were reported to improve after LTx (9;10).


Primary graft dysfunction (PGD)

PGD is clinically defined by the presence of hypoxemia, non-cardiogenic pulmonary edema and infiltrates on radiographic imaging and is the major cause of morbidity in the earliest period after transplant, i.e. within 72 hours of surgery. It affects 10% to 30% of all recipients. Prolonged hospitalization and increased short- and long-term mortality are associated with PGD. The exact pathogenesis is unclear; however, lung ischemia reperfusion injury (IRI) is thought to be a driving force in its development (11). IRI is an inflammatory insult that, in the case of organ transplantation, is initiated/enhanced by donor brain death, ventilator induced lung injury, aspiration events and cold preservation (12). Subsequent implantation with resulting reoxygenation after reperfusion drives the development of injury (13). Furthermore, according to a retrospective review by Toyoda et al., extracorporeal membrane oxygenation (ECMO), i.e. a technique for providing both cardiac and respiratory support to critically ill patients who can no longer survive otherwise, which serves as a bridge to LTx, significantly increases the risk of PGD. This in turn leads to higher rates of post-transplant ECMO and longer median hospital stay (14). Physiological changes linked with ischemia-reperfusion are widespread capillary leakage and consequent impaired gas exchange. Induction of inflammatory mediators such as chemokines, cytokines and oxygen radicals is one of the crucial mechanisms underlying aforementioned symptoms (15). Furthermore, IRI plays an important role in acute rejection of the graft, as it leads to the release of endogenous innate immune activators that stimulate toll-like receptors (TLRs) and activate the innate system within minutes after graft implantation. These substances that activate the innate immune system have remained unidentifiable. A substantial role seems to be accredited to alveolar macrophages as orchestrators of innate immune responses in the lung. Also the share of natural killer cells and neutrophils should not be disregarded. These findings pinpoint the importance of the innate immune system in organ rejection (13).

In the light of the aforementioned consequences of IRI, treatment, which includes diuresis and maximal ventilator support, should not be postponed or omitted. It includes. In most cases, this condition resolves over 24-48 hours (10).


The lung graft, in contrast to all other transplanted organs with the exception of the small bowel, is in permanent contact with the external environment. As such, various infectious agents can easily gain access to the graft and cause harm (16).

Cytomegalovirus (CMV) is the most prevalent postoperative infection. CMV-negative recipients receiving CMV-positive donor lungs are at the greatest risk of developing life-threatening disease from primary infection. The underlying reasons for this association are not known, but may include immunogenic or fibrotic effects of CMV activation in the allograft (9;10).

In the early post-transplant period (first 30 days), bacterial infections are most common and remain the primary cause of mortality. Generally, the donor's and host's resident bacteria, complemented by organisms populating the intensive care unit at the time, are involved (9;10).

Fungal infections, mostly represented by Candida albicans and Aspergillus, are a major problem after lung transplantation and occur early and late following transplantation (9).


In comparison with other types of solid organ transplantation, the incidence of acute and chronic rejection is much higher in consequence of lung transplantation. As described by the Registry of the ISHLT, up to 55% of lung transplant recipients are treated for acute allograft rejection in the first year after transplantation (16;17). Below, mechanisms of acute rejection, clinical presentation, diagnosis, histology, risk factors, therapy and other rejection related subjects will be discussed.

Sophisticated mechanisms that recognize non-self from self permit organisms to balance between responding to infectious agents on the one hand and yet tolerating their own cells on the other. As such, the advanced interplay of the innate and adaptive immune system causes a strong response to organ allograft reception. This alloimmune reaction is mainly driven by T cell recognition of foreign major histocompatibility complexes (MHC), which are in humans also referred to as Human Leukocyte Antigen (HLA). The MHC regulates the immune response by presenting antigenic peptides to T cells. In transplantation, there are 2 pathways for presenting allogenic MHC to recipient T cells: in the direct pathway, allogenic MHC is directly presented to recipient T cells by donor macrophages or dendritic cells in the graft. In contrast, using the indirect pathway, alloantigens get processed and presented to recipient T cells by recipient macrophages or dendritic cells when donor antigen-presenting cells (APCs) die out or are destroyed. The clinical importance of this mechanism lies in the fact that the enormous diversity of HLA polymorphisms creates an environment of quick recognition of the graft as non-self on the basis of HLA differences with the recipient. Subsequent recruitment and activation of recipient lymphocytes (mostly effector T cells) to the lung allograft can result in injury to and loss of function of the organ (17).

When symptomatic, the clinical presentation of acute rejection can include dyspnea, cough or sputum production. Pulmonary function testing such as spirometry can contribute to clinical evaluation and radiographic imaging is useful in tracing specific causes of symptoms or decreased pulmonary function. Ground-glass opacities, septal thickening, volume loss and pleural effusions on CT scans suggest acute rejection. For a solid diagnosis of acute rejection, transbronchial biopsies (TBBs) are the gold standard. At the Leuven Lung Transplant Unit, routine biopsies are performed at day 1, 21, 90, 180, 360, 540 and 720, or when clinically indicated. As such, clinically silent acute rejection can be identified in time (16;17).

Acute lung rejection is defined by perivascular mononuclear cell infiltrates. This process may be accompanied by lymphocytic bronchiolitis (18). For grading the histological appearance of acute rejection, the ISHLT has draught a classification system:

A-grade acute cellular rejection (ACR) of the lung allograft. Perivascular mononuclear infiltrates with or without interstitial mononuclear cells (mainly T cells, occasionally B cells or eosinophils) are thought to represent the typical acute lung allograft rejection (Figure 1A). Increasing thickness of the mononuclear cell cuff around vessels with increasing mononuclear invasion into the interstitial and alveolar space is what determines the A-grade (17).

B-grade airway inflammation or lymphocytic bronchiolitis (LB). Similar lymphocytic processes as described in ACR have also been defined in the bronchi and bronchioles (Figure 1B). The possibility of coexistent infections and other confounding factors have made it difficult to interpret the involvement of small airways as a manifestation of acute rejection in the past. Nevertheless, the expression of MHC antigens by bronchial epithelium during episodes of acute rejection or insufficient immunosupression has led to increasing acceptance that findings of LB could be considered as a marker of acute rejection (19). Recently, Verleden et al. found a significant correlation between exposure to particulate air pollution (with particulate matter < 10 µm) and LB diagnosed 2-3 days afterwards. This indicates that acute exposure to air pollution increases the risk for acute rejection, as LB is seen as a marker for it (20).

Figure 1. (A) A-grade acute cellular rejection with extensive perivascular mononuclear infiltrate (H&E staining; x40). (B) Lymphocytic bronchiolitis with dense peribronchiolar mononuclear infiltrate (H&E staining; x40) (17).

In addition to ACR, some lung transplant recipients appear to show a humoral response to the graft. Antibody-mediated allograft rejection is, although less frequent than ACR, gaining awareness as a concept within lung transplantation. In the acute phase, a humoral response is mounted to donor MHC antigens, although other endothelial or epithelial antigens expressed in the lung may become antibody targets as well. In the presence of the appropriate cytokine and co-stimulatory factors, B cells receive help from indirectly activated T cells for antibody class switching and affinity maturation. Bharat et al hypothesize that PGD in the immediate post-lung transplant period influences allograft rejection due to inflammation and upregulation of MHC on the allograft. This leads to increased alloantigen presentation and production of antidonor HLA-antibodies (17;21;22).

There are some factors, both alloimmune-dependent and -independent, which are thought to predict the development of acute rejection. Like this, it is generally believed that the intensity of host alloimmune response is related to recipient recognition of differences with the donor antigens, and that this process drives acute lung allograft rejection. As such, the more HLA mismatching between donor and recipient, the greater the risk of acute lung rejection. Multiple factors in the recipient him- or herself can also predispose for acute rejection, such as polymorphisms in diverse genes, as for example in the gene for TLR4 (17).

In the light of long-term outcomes, every conceivable effort should be made to prevent acute rejection. The association between these two entities will be discussed later on. Thereby, induction and maintenance immunosupression regimens appear to have a favorable effect on survival. Induction therapy with either a cytolytic agent or an interleukine-2 receptor blocker has been shown to reduce early rejection rates. In clinical practice, multiple maintenance immunosuppressive regimens exist, but calcineurin inhibitors are used in all patients, with cyclosporin and tacrolimus as the most known representatives of this group of immunosuppressives. The most used purine synthesis antagonist is mycophenolate mofetil. Also azathioprine is used commonly. To arrest an episode of acute cellular rejection and the damage it gives rise to, recipients are treated with steroids such as methylprednisolone and corticosteroids. The dose depends on the severity of the episode (9;10;17).

Chronic Lung Allograft Dysfunction (CLAD)

Bronchiolitis Obliterans Syndrome

Bronchiolitis Obliterans Syndrome (BOS) is the more common name for chronic rejection (12;23). This phenomenon was first described in the mid-1980s in a small cohort of heart-lung transplant recipients at Stanford University (24). Below, diagnosis, incidence, histopathological manifestation, risk factors and treatment will be discussed.

Diagnosis and clinical presentation

Diagnosing a lung transplant recipient with BOS is a clinical matter which is based on pulmonary function testing. Because the histological features of BOS, which will be described in-depth further on, are often variable and not unequivocal, the ISHLT considers the forced expiratory volume in 1 second (FEV1) the most reliable and consistent indicator of BOS, i.e. a persistent decline in FEV1 of <80% compared to the best post-operative value. Using this criterion, clinicians should first exclude other causes of pulmonary dysfunction, eg. infection, ACR, recurrent or progressive native disease. Herein, radiographic imaging, which is systematically performed, plays a key role. moreover, high resolution computed tomography (HRCT) imaging may be useful for confirming and refining the diagnosis (12;23).

Together with infection and various types of cancer, BOS is one of the leading causes of morbidity and late mortality (> 1 year) after heart-lung transplantation. Although it is rare within the first year after LTx, data from the ISHLT describes BOS in 48% of recipients by 5 years after LTx and in 76% by 10 years (9;12) (Figure 2).

ishlt freedom of bos curve.PNG

Figure 2. Freedom from BOS in lung recipients for follow-up between April 1994 and June 2011, conditioned on surviving 2 weeks (9).

BOS is a very heterogeneous condition. Although it affects all LTx recipients irrespective of donor and recipient's characteristics, type of transplantation and pretransplant disease, the time to onset and the rate of progression of the disease are very variable between patients (16) (Figure 3). In a British progressive study, 204 LTx recipients were followed (25). 56% exhibited a sudden drop in FEV1, i.e. acute BOS onset, with an abrupt, severe decline in lung function over a few weeks, whereas 18% followed a smooth linear decline over months to years, i.e. chronic onset. Moreover, there is a considerable association between the time to onset of BOS and post-BOS survival, as recipients with chronic onset BOS have a mean survival of twice that of patients experiencing acute onset, i.e. 58 months vs. 29 months respectively (12;16;25).

fig Knoop Estenne Edited.png

Figure 3. Changes in FEV1 over time elapsed since transplantation in three patients with BOS. Cut-off for diagnosis is put at 80% of best post-operative value. The figure illustrates that the pattern of functional change is highly variable among patients, as the two recipients designated by an open or filled circle show acute BOS onset while the patient represented by a square experiences chronic onset (16).

Obliterative Bronchiolitis, the histopathological correlate of BOS

The histological manifestation of BOS is referred to as obliterative bronchiolitis (OB), which is originally an inflammatory process that later on advances into having a relatively acellular fibrosing course affecting the bronchioles. The process is initialized by injury to the epithelium lining the bronchiolar lumen. This leads to infiltration of lymphocytes which can extend as far as the mucosa, submucosa and into the peribronchiolar parenchyma, causing ulcerations and necrosis. This course of events is known as lymphocytic bronchiolitis. Meanwhile, an inflammatory reaction in the airway lumen develops as well, joined by the recruitment and proliferation of (myo)fibroblasts in an attempt to repair the damage done to the bronchiolar structures. This causes submucosal bronchiolar fibrosis. Loose edematous intraluminal granulation tissue containing numerous lymphocytes and macrophages and, occasionally plasma cells and neutrophils, may lead to subtotal (eccentric or concentric) or total obliteration of the airway lumen (12;26-28). It was already stated in the mid-1980s by investigators of Stanford University (at the time one of the frontrunners on heart-LTx) that concentric sheets of collagen may surround the central core of necrotic debris, lymphocytes and macrophages. As such, OB resembles a plug of collagenized scar tissue in these bronchioles. Thin capillaries occasionally supply these intraluminal masses of collagen (24;28;29). In an advanced state, OB can include a spectrum from partial to completely acellular fibrotic obliteration where a scar is the only remnant of the airway lumen. Recognition is then only possible based on its location adjacent to an artery and by staining residual circumferential smooth muscle (12;28) (Figure?). The pathological findings described above are time-dependent, meaning that OB lesions with certain characteristics can evolve into lesions with other features throughout time. Generally, three main stages can be recognized when describing OB:

Inflammatory active OB. Lesion characterized by peribronchiolar infiltration of mononuclear cells, i.e. macrophages and lymphocytes.

Fibrotic active OB. Recruitment and proliferation of (myo)fibroblasts lead to the intraluminal accumulation of granulation tissue.

Inactive OB. The complete fibrotic obliteration of the lumen, without significant inflammation or active fibrosis. The smooth muscle layer is at least partially disrupted and intraluminal masses of collagen can be discerned.

Note that the entire spectrum of above mentioned expressions of OB can be present at one given time point within the same patient (30).

OB is a histologic diagnosis, i.e. the pathological term 'obliterative bronchiolitis' should be reserved for histologic specimens showing dense fibrosis within the small airways. However, TBBs, the gold standard for diagnosing acute rejection, may not be very helpful in this case, especially in the early stages, as the presence of LB or intraluminal granulation tissue is not sufficient to speak of OB. Moreover, a study performed by Chamberlain et al showed a low sensitivity of TBBs for identifying BOS (31). In 105 LTx recipients surviving three months or more, one TBB was performed. Sensitivity, which measures the proportion of actual positives which are correctly identified as such and specificity, which measures the proportion of negatives which are correctly identified, were determined and were 17,1 and 94,5%, respectively. Hence, because of the low sensitivity, it is very difficult to diagnose a recipient as suffering from BOS based on histological findings. Therefore, the disorder is frequently diagnosed clinically, as mentioned above (12;31;32).

Diagnostic imaging is very important in evaluating CLAD in recipients. Herein, high resolution computed tomography (HRCT) scans play the lead role. With slices that are only 1-2 mm thick (compared with conventional CT in which slices are 10 mm thick) in combination with a high-spatial-frequency reconstruction algorithm, HRCT can resolve objects of 0,5 mm diameter and is thus ideally suited for use in the lungs (33;34). HRCT scans can visualize typical signs suggestive of OB. Somewhat paradoxically, the scans of recipients diagnosed with chronic rejection have evidence of proximal bronchiectasis and the bronchial tree shows alternating areas of dilatation and constriction. The bronchiectasis is often accompanied by inflammation within the bronchial wall, which is also often thickened, and submucosal scarring. Distal to these areas, obliteration of the bronchioles is regularly observed. This then frequently leads to air trapping, due to obstructive nature of the scars. (12;28;35;36).

Mechanisms involved in the development of BOS

Although alloimmune reactivity remains the cornerstone in the development of BOS, recently, several novel mechanisms such as antibody-mediated rejection and autoimmunity have been proposed to contribute to the fibroproliferative cascade of events leading to OB. Together with alloimmune T cell reactivity, they will be briefly discussed below (37).

As mentioned before, an alloimmune reaction of the host brings about an intense immune response, causing inflammation. The specific mechanisms that lead to fibrotic obliteration of allograft airways during BOS, may involve interactions between the subpopulations of CD4+ effector T cells, namely TH1-, TH2- and TH17 cells. The TH1 cells secrete in particular IL-2 and IFN-γ and are as such responsible for classic cell-mediated functions like the activation of cytotoxic CD8+ T lymphocytes. The TH2 subset secretes various cytokines such as IL-4, Il-5, IL-6 and IL-10 and functions more effectively as a helper for B-cell activation and thus humoral response (12). The TH17 lineage is one of the sources of IL-17, which is suspected of orchestrating neutrophilic influx via inducing the production of several chemokines by, notably IL-8. This causes immune dysregulation which results in a fibrotic process with infiltration and proliferation of fibroblast, and epithelial cell injury. This course of events will ultimately culminate in airway obliteration. As IL-23 is known as a key cytokine in TH17 differentiation, a substantial part of the developmental mechanism of BOS may be ascribed to the IL-23/IL-17 axis (38;39).

Furthermore, it has been demonstrated that before the evolution to OB, a significant decrease in the number of blood vessels supplying the small airways takes place. Indeed, acute cellular rejection is a vascular process and is, in addition, widely accepted as a risk factor for developing BOS. This suggests that allograft airway microvascular injury causes local ischemia that contributes to loss of epithelia integrity and the upregulation of inflammatory mediators. The compromise of the microvascular blood supply may also affect reparative responses in the small airways. Moreover, local ischemia will stimulate hypoxia inducible factors which stimulate angiogenesis, a conditio sine quae non for inflammation and fibrotic obliteration. The end result is the scarring by which OB is characterized (12;40). In contrast to the decrease in microvasculature, there have been reports of an increased large airway vascularity in human lung allografts (41;42). Interestingly, no significant differences could be demonstrated between stable LTx recipients and patients suffering from BOS, leading to the hypothesis that the increase in airway vascularity is probably due to the nature of transplant surgery. Early allograft ischemia, possibly a result of the latter, might stimulate angiogenesis leading to airway narrowing. A precedent for this hypothesis can be found in asthma, where the inflammatory response increases the vascular response and as such encroaches the airway lumen. The same happens in the lung allograft, as bronchoalveolar lavage (BAL) and airway T cell infiltration and acute rejection scores were positively correlated with increased airway vascularity. Because it is known that even stable LTx recipients have persistent airway allograft inflammation, the changes seen in the airway vasculature could therefore be considered secondary to this acute inflammatory process. Like this, the vascular enlargement may assist in inflammatory processes by facilitating the influx of inflammatory cells into the airway wall. Ultimately, increased vascularity, embodied by enlarged vessels, promotes airway narrowing and thus contributes to airflow limitation distinctive of BOS (40;41).

Also humoral, or antibody-mediated rejection, has been recognized to play a role in the pathogenesis of BOS. HLA or donor-specific antibodies are correlated with an increased risk for BOS as they have been shown to damage the allograft airway epithelium (12). By performing in vitro studies, Jaramillo and colleagues demonstrated that anti-HLA antibodies can cause activation of airway epithelial cells, which then stimulate fibroblast proliferation and eventually undergo apoptosis (43). As such, they could potentially contribute to the generation of OB lesions. Non-donor-specific antibodies targeted at the bronchial epithelium and bronchial wall microvasculature have been detected and described among lung allograft recipients as well. This could lead to activation of the complement cascade with complement deposits leading to endothelial cell injury, production of proinflammatory molecules and recruitment of inflammatory cells (17;37).

Recently, also autoimmunity was described as a mediator of BOS. Both immune (as described above) and non-immune injury (e.g. infection) may reveal previously unexposed, or sequestered, self-antigens to the recipient's immune system, possibly triggering autoimmune responses. The exposed antigens can then sustain rejection even in the absence of persistent alloimmunity. Type V collagen [col (V)] was the first native protein to be proposed as a self-antigen (12;37). Through the experimental work of several research groups, increasing evidence arises for the central role of col (V) in autoimmunity, ultimately resulting in BOS. Like this, Wilkes et al prospectively monitored responses to col (V) in LTx recipients and observed a significantly higher risk for developing BOS in those patients with elevated col (V)-specific cell-mediated immunity (44;45). Furthermore, rats fed col (V) before receiving a lung allograft, were rendered tolerant to the protein and, even without administration of immunosuppressive drugs, did not develop acute or chronic rejection (46). Intriguingly, anti-col(V) cell-mediated immunity was associated with the upregulation of IL-17. Precedents for that outcome can be found in common, similar autoimmune diseases like rheumatoid arthritis and inflammatory bowel disease (12;39).

Risk factors for BOS

Alloimmune-dependent risk factors for BOS

Acute cellular rejection (ACR). Several studies pose ACR as the dominant risk factor for the development of BOS, as multiple or severe episodes of ACR increase the risk of BOS (12). However, the relationship between ACR and BOS is rather counterintuitive as the former involves blood vessels and the latter obliterates airways. Furthermore, the severity of ACR was a risk factor in univariable analysis, but was not an independent risk factor for either BOS or death in multivariable analysis (47).

Lymphocytic bronchiolitis (LB). Recently, Glanville and co-workers suggested that actually the intensity of LB was more decisive to the later development of BOS than ACR, taking into account the fact that the relationship between ACR and BOS was not significant in multivariable analysis. This suggests that it is the concomitant grade of LB that determines the outcome when ACR is diagnosed. The axiomatic association between ACR and chronic rejection is thus due to the fact that early studies didn't take LB (correctly) into consideration (12;47).

Human leukocyte antigen (HLA) mismatches. Up to the present, HLA compatibility has not played a role in the allocation process of donor lungs. Recently, using the data of the Collaborative Transplant Study upon LTx, 5-year graft outcome in 8020 deceased donor LTx's performed between 1989 and 2009, was linked to the amount of HLA mismatch between donor and recipient (48). The key finding was that the risk of graft failure decreased as the number of HLA mismatches decreased. Surprisingly, lung transplants with perfect matches, i.e. zero HLA mismatches, showed an extremely poor outcome with half of all grafts failing during the first year (48). However, disregarding the latter, it may be worth the effort to consider (at least partial) donor-recipient HLA matching, as nowadays, three or more HLA mismatches were present in about 95% of LTx recipients (12). Unfortunately, this is difficult to accomplish: because the size of the patient pool is much smaller than compared to other solid organ transplantations, obtaining good HLA matches is inevitably also reduced because of the extensive genetic polymorphisms of the HLA system (48).

Nonalloimmune-dependent risk factors and BOS

Infection. Nonimmunologic inflammatory conditions may also trigger chronic rejection. CMV infection has been well described as a potential risk factor in the development of chronic rejection; however, others found only a marginal or no relationship at all. CMV shares nucleic acid sequence homology with MHC class I and as such, due to CMV infection, proinflammatory cytokines and allogeneic responses may be upregulated. Also lower respiratory tract infections due to community acquired respiratory viruses (CARV) such as parainfluenza, respiratory syncytial virus, metapneumovirus, coronavirus, rhinovirus, influenza virus and paramyxovirus have been associated with chronic rejection. Therefore, adequate prevention of infections is of utmost importance, as this may reduce the incidence of chronic rejection (16;23;49).

Primary graft dysfunction (PGD). From a recent retrospective single-centre study, it appears that LTx recipients who developed PGD had an increased risk of BOS. This was demonstrated using a multivariable model, suggestion that the association was independent of ACR, LB and CARV infections. Moreover, the increased risk of BOS was directly related to the severity of PGD (12;23;50).

Gastroesophageal reflux (GER). GER is common in the pre-LTx population due to end-stage lung disease and may be exacerbated after LTx. The mechanism by which GER contributes to BOS remains quite unclear, but aspiration, the entry of secretions or foreign material into the trachea and lungs, may contribute to airway injury as BAL fluid reveals the presence of bile acids and pepsin. As a standard treatment with medication such as proton pump inhibitors only affects acid reflux, surgery is recommended in the case of persistent post-LTx GER. This is associated with greater freedom of BOS and improved lung function, although this needs confirmation in a clinical trial (12;16;23).

Air pollution. Nawrot et al. found that chronic exposure to (traffic-related) air pollution predisposes to BOS (51). Patients living within 171m of a major road were 2,06 times more likely to develop BOS than patients living farther away. Also considering the results of the above mentioned publication of Verleden et al. (21), these two studies collectively highlight the vulnerability of the lung graft for environmental factors such as air pollution, which appears to be a risk factor for both acute and chronic rejection. Therefore, in contrast with most other solid organ transplantations, both short-term morbidity and long-term mortality after LTx is influenced by the constant exposure of the graft to the external environment.

BOS as a heterogeneous condition

Following the 2001 diagnostic criteria of Estenne et al., BOS can be characterized by:

Neutrophilic airway inflammation, as can be discerned performing a bronchoalveolar lavage (BAL),

largely irreversible and persistent obstructive pulmonary function decline, and

a fibroproliferative narrowing of the bronchioles (52).

However, this definition needs revision and rephrasing, as thorough clinical observation and patient characteristic (for instance the greatly diverging survival between patients) reviewing have encouraged several research groups to set up new thinking frames in which BOS is no longer the homogenous entity it was for decades. Based on the existence of multiple phenotypes within the BOS patient population, which will be discussed below, the term chronic lung allograft dysfunction (CLAD) has been introduced. Although the content of this term is still under constant evolution, it may better explain the heterogeneity of chronic events following LTx.

Azithromycin therapy revealed a dichotomy in BOS

In 2008, Vanaudenaerde et al. came across a dichotomy in BOS after the introduction of the macrolide antibiotic azithromycin (30). As treatment of BOS, which will be discussed in greater detail later on, was, and still remains to be disappointing, lung function improvement in subsets of patients after the administration of azithromycin raised new hope. Not only does the macrolide antibiotic arrest further lung function decline, an effect that is comparable to the outcome of other therapies, it also seems to reverse the drop in lung function. However, that outcome is only seen in about 40% of the patients, according to several studies (53-55). The beneficial effect in LTx recipients suffering from BOS is believed to be, although probably not completely, attributable to its anti-inflammatory properties. In 2006, Verleden et al performed a 3-month follow-up study on azithromycin therapy for patients diagnosed with BOS (56). Besides a mean increase in FEV1 of 13% taking all study participants into consideration, a significant decrease of both BAL neutrophils and BAL IL-8 was reported. Vanaudenaerde et al. further investigated this finding using human airway smooth muscle cells and demonstrated that azithromycin inhibits the IL-17-induced IL-8 production in these cells and as such counteracts neutrophil activation and recruitment (57). Following these findings, the existence of at least two phenotypes within BOS was proposed: neutrophilic reversible allograft dysfunction (NRAD) and fibroproliferative BOS (fBOS) (30). A subsequent study performed by the same group strengthens this line of thinking via insights acquired by measuring in BAL the expression of a wide range of proteins reflecting different processes in patients suffering from CLAD (58). When compared to stable patients, significant protein variations were measured in the NRAD group, whereas none of the examined proteins showed a different expression profile in the fBOS group. This demonstrates not only yet again the fact that, within BOS, there are divergent conditions, but also that all previous data concerning neutrophilia and the expression pattern of relevant proteins needs to be looked at extra critical, as most studies performed up till now have not made the differentiation between NRAD and fBOS, which could have biased the results (58).

Restrictive allograft syndrome (RAS): a novel form of CLAD

As azithromycin is an effective treatment for NRAD, the focus is now shifting towards irreversible BOS. Like this, it becomes increasingly clear that the subdivision of BOS into NRAD and fBOS is not the endpoint. Recently, the Toronto group proposed to take restrictive pulmonary function into consideration as well (59). After performing a retrospective review involving 478 patients who received bilateral LTx in the Toronto LTx program from 1996 to 2009, they concluded, based upon pulmonary function tests, histopathology and radiology, that also irreversible CLAD is a heterogeneous condition. On the one hand, there is fBOS which is characterized by a stable total lung capacity (TLC), where restrictive allograft syndrome (RAS) is defined as irreversible CLAD with a decline in TLC of <90% compared to baseline (59). This restrictive phenotype has a typical histopathological hallmark, namely diffuse alveolar damage and extensive fibrosis in the alveolar interstitium, visceral pleura and interlobular septa, with scattered OB lesions. Massive infiltration of myofibroblasts was demonstrated in the peripheral lung tissue of RAS patients. In contrast, BOS lungs show, beside diffuse OB lesions, relatively intact peripheral lung tissue without myofibroblasts. RAS can occur at any time after LTx and it accounts for approximately 25 to 35% of CLAD over time (59;60). Moreover, both groups report worse survival in RAS patients, which was less than half of that of BOS patients (figure 3). This indicates a significant negative impact of RAS on lung allograft survival (59;60).

Figure 3. Survival of patients who developed BOS versus patients that developed RAS after onset of CLAD in a retrospective review of 478 patients between 1996 and 2009 (59).

Because acute rejection and infection were demonstrated in the acute phase of RAS in some cases, the authors pose that in RAS, an insult to the lung allograft leads to acute, uncontrollable inflammation followed by fibrosis. The episodes were treated with high dose steroids although the efficacy remains to be ascertained as pulmonary function may improve, remain stable or continue to decline (61).

Despite the efforts of several groups, many questions remain unsolved. It is for instance unclear whether fBOS and RAS are merely a time-dependent manifestation of the same condition or if different pathophysiological mechanisms are active. Further characterization of the various manifestations of CLAD is indispensable to gather in-depth knowledge of this heterogeneous condition.

Treatment of BOS

Although almost three decades have passed since BOS was first described, little progress has been made in preventing and treating it. Up till now, BOS remains a pervasive process with devastating consequences. Given the link between ACR and BOS, every episode of acute rejection should be persistently treated as described above. Moreover, most LTx recipients routinely continue to receive a triple-drug maintenance immunosuppressive regimen consisting of a calcineurin inhibitor, a purine synthesis antagonist and corticosteroids. Baseline immunosupression is often switched when BOS is first diagnosed. However, there is no consensus about the exact approach, so no recommendations can be made about a general treatment of BOS, let alone with reference to routine management. Therefore, it is worth the effort to take a closer look to the use of other immunosuppressant therapies in novel ways (3).

Aerosolized cyclosporine. The topical delivery of cyclosporine to the airway does not additionally improve the rate of ACR but caused a reduction in BOS and improved survival. In addition, this new method of administering cyclosporine may result in an improvement of the pulmonary function of the LTx recipient (12;23;62).

Alemtuzumab. This humanized anti-CD52 antibody depletes CD4+ lymphocytes which significantly improved the histological grade of BOS in >50% of a small cohort of patients. However, alemtuzumab had no influence on pulmonary function (12;23;63).

Azithromycin. Macrolide antibiotic that improves lung function in a subset of patients suffering from CLAD, as mentioned extensively before.

Total lymphoid irradiation. In LTx recipients suffering from fBOS, so were azithromycin no longer shows a beneficial effect on FEV1, total lymphoid irradiation (TLI), may stabilize the decline of the FEV1, without producing major side effects. It is known that TLI acts on T cells and as such, it may also affect TH17 cells. Like this, TLI interferes with the allogeneic response to the graft. Therefore, it may represent an interesting treatment option and serve as a bridge to retransplantation (64).

Extracorporeal photopheresis is a treatment whereby isolated white blood cells are treated with photoactivable drugs which are then activated with ultraviolet light. Its effectiveness has been shown in several T-cell mediated diseases, as it induces lymphocyte apoptosis. In 2012, Jaksch et al. performed a prospective study to evaluate the efficacy of extracorporeal photopheresis (ECP) in patients suffering from BOS (65). The large cohort consisted of 1012 LTx recipients (transplanted between 1989 and 2010) whereof 194 developed BOS and received established treatment. Of them, 51 patients received additional ECP. The results were quite promising as 61% showed improvement or stabilization of FEV1 after 3, 6 or 12 months of therapy. As a result of the observed improvement in lung function, there was a significant survival benefit as well (65).

Montelukast. As LTx recipients suffering from fBOS are not responsive to azithromycin, and only a temporary arrest in FEV1 decline can be achieved, the search continues for a appropriate treatment for these patients. Therefore, Verleden et al. performed a pilot study with montelukast, a leukotriene receptor antagonist widely used in the treatment of asthma (66). As such, they demonstrated that the addition of montelukast in patients with fBOS resulted in a slowing of FEV1 decline. Treatment with montelukast grossly has the same effect as TLI and ECP, but the latter are very time consuming and quite expensive treatment options. Moreover, they are not available in every hospital were LTx is performed. In contrast, treating LTx recipients with montelukast is very easily feasible, cheap and available to everyone (66).

So far, it seems that LTx can be considered merely a treatment for many end-stage pulmonary disorders, and BOS is seen as the final frontier for it to become an actual cure. The definitive strategy for managing BOS nowadays, is retransplantation. Facts and figures about retransplantation performed by the Leuven Lung Transplant Unit can be found earlier in this overview of the literature. In 2007, Kawut et al. performed a large retrospective cohort study of patients who underwent LTx between 2001 and 2006 in the USA (67). One should bear in mind that retransplantation candidates already shoulder a high burden of medical complications due to chronic immunosuppressive treatment, so it is not surprising that survival after lung retransplantation is not as good as after the initial LTx procedure (figure 4). Most of the time, this is due to the higher prevalence of comorbidities. In addition, retransplant recipients seem to have a higher risk of BOS. Besides, retransplantation early after the initial LTx poses a particularly high mortality risk (67). Moreover, society should not lose sight of the ethical considerations concerning this practice: initial LTx candidates are facing a waiting list and accordingly, substantial thought should be given before restarting the clock again (23;68).

Figure 4. Kaplan-Meier survival estimate of initial LTx recipients and retransplant recipients who underwent these respective procedures between 2001 and 2006 (67).

BOS after allogeneic hematopoietic stem cell transplantation: pulmonary chronic graft-versus-host disease

Allogeneic hematopoietic stem cell transplantation (HSCT)

The intensive research efforts made in the 1950s and early 1960s concerning the infusion of bone marrow, was triggered by the clinical observations of severe bone marrow suppression or myelotoxicity of radiation among nuclear bomb survivors at Hiroshima and Nagasaki. The first reports of clinically successful application of bone marrow transplantation occurred in the late 1960s and early 1970s, when the procedure was mainly applied in patients with severe combined immunodeficiency disorders and advanced acute leukemias. Nowadays, the term 'bone marrow transplantation' has become obsolete and is replaced by hematopoietic stem cell transplantation, as hematopoietic stem cells can be derived from a variety of sources, such as the peripheral and umbilical cord blood. However, in steady-state, the concentration of hematopoietic stem cells and myeloid progenitor cells is rather low, so administering hematopoietic growth factors aims to recruit these cells. Furthermore, it appeared that allogeneic HSCT using peripheral blood mobilized stem cells instead of bone marrow retrieved ones, imposes a greater risk for the occurrence and severity of chronic GVHD, which will be discussed in-depth later on (69). All things considered, peripheral blood hematopoietic stem cells are presently used in about 65% of allogeneic HSCT. Consequently, the following definition for HSCT was proposed: 'The process and intravenous infusion of hematopoietic stem and progenitor cells to restore normal hematopoiesis and/or treat malignancy.' Furthermore, the term allogeneic refers to the source of the stem cells as being a genetically non-identical donor from the same species. Allogeneic HSCT has become a common treatment option for several hematologic malignancies, with as main example acute myeloid leukemia (AML) accounting for 33% of transplants (70). AML is a clonal stem cell malignancy resulting in the accumulation of immature leukemic blasts in the bone marrow and peripheral blood and, as such, interferes with normal hematopoiesis (71). In addition, allogeneic HSCT is a standard treatment for many immunodeficiency states (70).

Graft-versus-host disease (GVHD)

In addition to graft rejection, which occurs when recipient immunologically competent cells destroy the transplanted cells from donor origin, an event taking place in 5-11% of allogeneic HSCT, the recipient's integrity is at risk as well. When Lorenz et al. carried out their pioneer experiments, performing bone marrow transplantation in mice, they observed that this intervention could cure radiation aplasia, i.e. the bone marrow becoming incapable of performing hematopoiesis. However, they also found that allogeneic bone marrow transplantation caused a lethal secondary disease, characterized by wasting, diarrhea and skin lesions. This disease is now commonly known as graft-versus-host disease (GVHD) and constitutes the most important barrier to allogeneic HSCT. GVHD develops when graft immunocompetent cells recognize major histocompatibility complex (MHC) molecules and mount an immune attack against the cells in the recipient. GVHD is described as either acute, i.e. within the first 100 days after transplantation, or chronic, generally presenting after the first 100 days post-transplant (70;72). The skin is generally the first and most commonly affected organ in acute GVHD. Prominent palmar erythema, i.e. redness of the skin, may be the earliest manifestation and can, in severe cases eventually progress into erythroderma, i.e. skin inflammation with exfoliation (figure 5). Although the skin may be the only target organ in acute GVHD, also the liver, gastrointestinal tract and, rarely, other organs may be involved (73).

Figure 5. (A) erythema and (B) erythroderma caused by acute graft-versus-host disease (73).

The focus will now shift towards chronic GVHD, which, with an incidence of 60-80% post allogeneic HSCT, remains a staggering burden for the patient due to the impact on overall health, functional ability and quality of life. In 20% of the cases, there is no history of prior acute GVHD, although alloreactive donor T cells play a central role in both time-dependent manifestations of GVHD (70;74). This was for instance shown by a meta-analysis on allogeneic HSCT performed with bone marrow-harvested stem cells in comparison with allogeneic HSCT using peripheral blood mobilized stem cell products, which contain a greater dose of donor T cells than stem cells garnered from the bone marrow. It appeared that allogeneic HSCT using peripheral blood mobilized stem cells, imposes a greater risk for the occurrence and severity of chronic GVHD (69). The clinical manifestations of the syndrome are diverse, but the most commonly affected organs are the skin, eyes, mouth, and liver. Less frequently, also pulmonary complications, among which bronchiolitis obliterans syndrome is predominant, have been described (74). This will be the focus of the remaining part of this overview of the literature.

Bronchiolitis obliterans syndrome after allogeneic hematopoietic stem cell transplantation

Definition and epidemiology

For a long time, there was no consensus about concerning the diagnose of BOS post allogeneic HSCT, for which there were at least ten distinct clinical definitions. This led to much variation in the estimated prevalence of BOS: 2-3% according to most studies, or 6% among patients with chronic GVHD, but others suspected the prevalence of BOS to be as high as 10-20% (75). Eventually, in 2005, the National Institutes of Health (NIH), i.e. the US' medical research agency, a part of the US Department of Health and Human Services, proposed consensus diagnostic criteria for BOS following allogeneic HSCT:

FEV1 < 75% predicted;

FEV1/FVC ratio < 0,7;

evidence of air trapping, small airway thickening, or bronchiectasis on HRCT or residual volume > 120% predicted or pathological confirmation;

absence of respiratory tract infection (76).

However, even with the NIH consensus criteria, the diagnosis of BOS remains a challenge. This was recently shown in a retrospective evaluation of a cohort of proven BOS patients (n=22), of which only 18% met the consensus definition for a clinical diagnosis of BOS. Restrictive lung disease in patients with BOS after allogeneic HSCT may, at least partially, account for this, as it would lead to a false normalization of the FEV1/FVC ratio (77;78). Taken together, these findings highlighted the need for thoughtful evaluation of the consensus criteria. While lung biopsy is definitive for diagnosis, the complication rate of 13% (e.g. pneumothorax but also death) inevitably causes scepticism toward this approach (77). Recently, some recommendations were made to augment the diagnostic accuracy of the consensus criteria. A strategy applied by the lung transplantation community is to look at the magnitude of decline of FEV1 compared to pre-transplant values rather than using a strict FEV1 threshold. This is because the original indication for LTx may have caused already a decreased in FEV1. Such an approach, i.e. close monitoring of the decline over time instead of relating post-transplant FEV1 values to predicted outcomes, may present the greatest opportunity for an earlier, more sensitive and specific diagnosis of BOS after allogeneic HSCT (78). In 2011, Au et al. performed a study in which, amongst others, aforementioned recommendation was used to assess the prevalence of BOS in a cohort of allogeneic HSCT recipients (75). As such, they identified 5,5% of the entire study population as suffering from BOS. Amongst patients who developed chronic GVHD, this value increased to 14%. Those values are now generally adopted. Furthermore, in the majority of patients, BOS developed between 300-600 days post-transplantation. The median time from transplantation to meeting the NIH diagnostic criteria, was 439 days (with a range between 274 and 1690 days) (75).

Risk factors

The study of Au et al. revealed, after multivariate analysis, two significant risk factors for the development of BOS after allogeneic HSCT (75).

Low circulating IgG levels. This association may be explained by increased susceptibility for infection, which is, in turn, a known risk factor for BOS after LTx. However, the general mechanism in which low circulating IgG levels give rises to BOS following allogeneic HSCT is currently unknown (75).

Chronic GVHD. As has been shown in several previous studies, the risk of developing BOS is strongly associated with the manifestation of chronic GVHD at another site. This robustly supports the long-accepted hypothesis that BOS, as it is also observed following LTx, represents an alloimmune reaction in the lung (75;77;79). This will be discussed more thoroughly below.


Biopsy specimens of recipients diagnosed with BOS after allogeneic HSCT, show the same characteristics as were observed in histopathological samples of patients suffering from BOS after LTx. Also here, early- and more advanced-type lesions can be discerned, which may imply that post-allogeneic HSCT BOS has a time-dependent course as well. Examination of the samples demonstrates obliteration of the airway lumen of terminal bronchioles due to the formation of granulation tissue between the epithelium and smooth muscle cell layer (figure 6A). The latter contains inflammatory cell infiltrates consisting of neutrophils and mononuclear cells. Diffuse alveolar damage may be present. This stage is referred to as early BOS. In the chronic phase, fibrotic obliteration of the small airway lumen can be noted, either as fibrotic active OB, showing the proliferation of fibroblasts and myofibroblasts or as inactive OB lesions, i.e. collagen scars (figure 6B) (77;78;80).

Figure 6. (A) Inflammatory active OB lesion showing granulation infiltrates between the epithelium and smooth muscle layer (H&E staining). (B) chronic phase post-allogeneic HSCT BOS with collagen scarring (H&E staining) (78;80).

Different phenotypes within post-allogeneic HSCT BOS

In a recent retrospective study, Bergeron et al. reviewed the data of the 77 allogeneic HSCT recipients diagnosed with BOS in the Parisian Saint Louis hospital between 1999 and 2010 (81). They discovered, based on pulmonary function testing, the existence of two distinct phenotypes within the BOS cohort. 53 patients had an FEV1/FVC < 0,7 and were as such corresponding to the NIH criteria for BOS as a manifestation of chronic GVHD, i.e. typical BOS patients. The remaining 24 patients however, showed besides a decrease in FEV1 to < 80% predicted, a concomitant decline in FVC to < 80% predicted resulting in a FEV1/FVC ratio > 0,7 (81). Indeed, this phenomenon was already mentioned by Williams et al. who suggested that restrictive lung disease may lead to a false normalization of the FEV1/FVC ratio (78). Moreover, HRCT of typical BOS patients revealed significantly less centrilobular nodules, i.e. nodules as small as 1-2 mm, mostly centered 5-10 mm from the pleural surface and not occurring in relation to interlobular septa, than were discernible on HRCT of atypical BOS patients. In addition, regardless of the phenotype, FEV1 only changed slightly during the follow-up after diagnosis of BOS. Also, no association was found between the FEV1 value and the outcome of the BOS patients. Taken together, these data may suggest that a change in pulmonary function testing is merely a late feature of BOS and that the decline in FEV1 is a sudden event rather than a progressive fact in most patients. Therefore, the authors call attention to the necessity of identifying new markers in order to early diagnose BOS, instead of the NIH consensus criteria that seem insufficiently capable of doing so (81).

In 2011, von der Thüsen et al. published a study representing four patients who underwent bone marrow transplantation and subsequently developed pleuroparenchymal fibroelastosis, i.e. a histological pattern of interstitial fibrosis and pleural thickening visible on HRCT (82). All patients were also diagnosed with BOS and, strikingly, presented with recurrent pneumothoraces. These may be caused by the increased tension in the healthy native lung parenchyma due to adjacent stiff fibrotic tissue. The etiology of the fibroelastosis remains elusive, as inflammation does not appears to be a dominant feature in the biopsies. However, due to concomitant presence of BOS and the delayed onset of the fibroelastosis, the latter may be a late complication of GVHD. Following the report of von der Thüsen et al. and given the fact that BOS after bone marrow transplantation is reminiscent of BOS post LTx, Ofek and colleagues hypothesize that pleuroparenchymal fibroelastosis might represent an important histological correlate of RAS following LTx (83). Performing a retrospective review consisting of 493 patients who underwent LTx between 1996 and 2009 and of which for 16 out of the 47 recipients who developed RAS there was histological data available, they found that all lungs showed varying degrees of pleural fibrosis and hardly one did not show pleuroparenchymal fibroelastosis. Given the fact that pleuroparenchymal fibroelastosis as found following both LTx and allogeneic HSCT, as well as the observation of auto-antibodies in some of these patients, suggests a contributing immunological mechanism. With these findings as a bridge between the late complications of LTx and bone marrow transplantation, also restrictive changes should be kept in mind in the management of pulmonary chronic GVHD (82;83).

Treatment and prognosis

Although treatment options for post-allogeneic HSCT BOS are similar to those applied in BOS after LTx, the prognosis is worse, with a very poor overall survival rate of 44% at 2 years and 13% at 5 years (77). In the same study as mentioned above, Bergeron et al., performing a multivariate analysis, also identified several prognostic factors for the development of BOS (81). Manifestation of acute GVHD, extensive chronic GVHD, diagnosis ≤ 1 year post-transplantation and treatment of BOS with prednisone, i.e. a corticosteroid drug, are associated with poor survival. Unfortunately, it is still unclear how systemic steroid treatment can lead to worse outcome (81). However, the administration of corticosteroids could go hand in hand with an early diagnosis, as patients whose BOS was clinically recognized without delay, very quickly were put on an augmented immunosuppressive regimen. As such, these patients might suffer from a more aggressive form of BOS whereas recipients that longer remain clinically unrecognized, experience a relatively milder and more stable manifestation of BOS (78).