Chronic lung allograft dysfunction: Restrictive allograft syndrome

INTRODUCTION — Chronic lung allograft dysfunction (CLAD) remains a major cause of morbidity and mortality following lung transplantation [1]. In the early years of heart-lung transplantation, it was noted that many recipients experienced declining lung function, usually with obstructive physiology and without evidence of a specific cause. Large tissue biopsies and autopsies frequently revealed histologic bronchiolitis obliterans (BO). The clinical scenario was therefore called bronchiolitis obliterans syndrome (BOS). The term BOS was used for all idiopathic drops in lung function after lung transplantation until 2010 to 2011, when the broader term CLAD was first introduced [2]. Between 2010 and 2019, the terms BOS and CLAD were often used interchangeably, although there was a growing recognition of a new CLAD phenotype characterized by restriction.

In May 2019, the International Society of Heart and Lung Transplantation (ISHLT) published guidelines that defined CLAD as the umbrella term, encompassing multiple phenotypes, including BOS and restrictive allograft syndrome (RAS) [1]. It is important to note that pre-2010 studies (and likely some 2010 to 2019 studies) included patients with RAS among patients labeled as BOS.

The epidemiology, clinical presentation, evaluation, diagnosis, treatment, and prognosis of RAS will be discussed here, while placing it in the context of CLAD. Other types of lung allograft dysfunction, such as BOS, acute antibody-mediated rejection, and acute cellular rejection, are discussed separately. (See "Chronic lung allograft dysfunction: Bronchiolitis obliterans syndrome" and "Evaluation and treatment of antibody-mediated lung transplant rejection" and "Evaluation and treatment of acute cellular lung transplant rejection".)

DEFINITIONS

Chronic lung allograft dysfunction (CLAD) — CLAD is an umbrella term describing a significant decline in lung function after lung transplantation in the absence of other identifiable causes [1]. CLAD is defined as a persistent and irreversible ≥20 percent drop in forced expiratory volume in one second (FEV₁) compared with the posttransplant baseline, which is itself defined as the average of the two maximal posttransplant FEV₁ values that are at least three weeks apart. If alternate diagnoses are identified at the time of CLAD onset (eg, acute rejection or infection, which represent important CLAD risk factors) but allograft dysfunction does not resolve after treatment, CLAD onset is defined at the first drop in FEV₁. If the alternate condition is considered irreversible (eg, pneumonectomy, new airway stenosis), a new FEV₁ baseline can be set [1]. A classification scheme is in use that takes into account diagnostic uncertainty early in the CLAD course (figure 1):

●Possible CLAD – New allograft dysfunction with an FEV₁ decline by ≥20 percent of posttransplant baseline, lasting <3 weeks.

●Probable CLAD – Identified by two FEV₁ values that are ≥20 percent below the posttransplant baseline and are at least three weeks, but ≤3 months, apart.

●Definite CLAD – Consistent lung allograft dysfunction with FEV₁ decline by ≥20 percent of posttransplant baseline lasting >3 months, as documented by FEV₁ measurements.

CLAD severity is determined based on the change in FEV₁ from baseline [1]:

●CLAD Stage 0 – FEV₁ >80 percent of baseline

●CLAD Stage 1 – FEV₁ 66 to 80 percent of baseline

●CLAD Stage 2 – FEV₁ 51 to 65 percent of baseline

●CLAD Stage 3 – FEV₁ 36 to 50 percent of baseline

●CLAD Stage 4 – FEV₁ ≤35 percent of baseline

CLAD phenotypes — The 2019 International Society for Heart and Lung Transplantation guidelines define four phenotypes of CLAD (table 1), which should be identified at the time of CLAD onset, based on the observed physiologic and radiographic patterns [1]. Measurements of FEV₁, forced vital capacity (FVC), total lung capacity (TLC), and a chest computed tomography (CT) are required for an adequate phenotype classification.

Bronchiolitis obliterans syndrome (BOS) — BOS is the predominant phenotype of CLAD and presents clinically as obstructive lung disease detected as a decline in FEV₁ from the posttransplant baseline, associated with a FEV₁/FVC <70 percent, with no restriction and no persistent fibrotic-like opacities. (See "Chronic lung allograft dysfunction: Bronchiolitis obliterans syndrome", section on 'Diagnosis'.)

Restrictive allograft syndrome (RAS) — The RAS phenotype is defined as CLAD with a restrictive defect (TLC <90 percent of the posttransplant baseline, defined as the average of the two TLC values measured at, or near, the same time as the two baseline FEV₁ measurements), persistent fibrotic-like opacities, and no obstruction. While the imaging assessment remains quite subjective, opacities consistent with RAS are currently defined as those that (1) look like parenchymal or pleural fibrosis, (2) are likely to cause restrictive physiology, and (3) are persistent.

For a precise assignment of a CLAD phenotype, patients need a TLC and chest CT to be performed around the time of CLAD onset. When these tests are not available, an FVC decrease to ≤80 percent of the posttransplant baseline can be used as a surrogate of restriction, although air trapping due to obstruction may also cause such a drop. Restriction by FVC along with relevant opacities identified on chest CT has been shown to be useful to identify patients with RAS with 80 percent sensitivity and >95 percent specificity [3]. Although less sensitive and specific, a chest radiograph can be used instead of a chest CT to identify opacities. If pulmonary function tests and/or imaging are not available, adequate CLAD phenotyping cannot be performed based on the international criteria.

Mixed phenotype — CLAD with the combination of obstruction and restriction, in the presence of persistent fibrotic-like opacities [4].

Undefined CLAD — A category characterized by either the presence of obstruction and persistent opacities without restriction or by obstruction with concurrent restriction but without persistent opacities. Eleven percent of 174 bilateral lung transplant recipients with CLAD were found to have an undefined phenotype in one single-center study [5].

Unclassified CLAD — When the International Society of Heart and Lung Transplantation (ISHLT) criteria are applied stringently to a cohort of CLAD patients, some patients may remain unclassified [1,5], up to 15 percent in one study of bilateral lung transplant recipients. It is possible that these patients have some other underlying process that cannot be identified with our current clinical tools.

Phenotyping CLAD in single lung transplant recipients — Most studies that assessed CLAD phenotyping have been performed in bilateral lung transplant recipients, where pulmonary function tests reflect both allografts. However, in single-lung transplant recipients, where PFTs are a combined representation of the allograft and the native lung, interpretation of clinical data is more challenging [6]. One recent study found that 22 percent of 105 single-lung transplant recipients with CLAD had undefined phenotype [7]. Another study diagnosed 3 percent of 67 single-lung transplant recipients with CLAD with undefined phenotype and 28.3 percent remained unclassified [6]. The latter study of single lung transplant recipients showed a poor interrater agreement for CLAD phenotypes but a significantly better agreement on the presence of chest CT opacities [6], suggesting that radiologic criteria are more reliable in these patients.

PATHOLOGY — The hallmark of RAS is the presence of parenchymal fibrosis. Several patterns of parenchymal fibrosis have been described in RAS. The classic pattern is that of pleuroparenchymal fibroelastosis (picture 1), characterized by preservation of alveolar septa and filling of the alveoli with collagen and other extracellular matrix proteins, often associated with pleural thickening and fibrosis [8-12]. This pattern is also frequently seen in the mixed phenotype CLAD [4]. A pattern of nonspecific interstitial pneumonia (NSIP) has also been described [8,9]. Concurrent bronchiolitis obliterans (BO) lesions are usually found within pathologic samples from RAS patients [8,9]. (See "Chronic lung allograft dysfunction: Bronchiolitis obliterans syndrome", section on 'Histopathology'.)

More inflammatory pathological findings have been identified prior to RAS onset or in early RAS phenotype. These patterns may represent precursor lesions that later evolve into parenchymal fibrosis and include diffuse alveolar damage [13,14] and acute fibrinous and organizing pneumonia [9,15].

Other pathological findings that have been described in conjunction with RAS-related fibrosis include acute cellular rejection, vascular fibrosis, and epithelial damage [8,9].

EPIDEMIOLOGY — CLAD affects approximately 50 percent of lung transplant recipients at five years [16]. The prevalence of RAS among patients with CLAD depends somewhat on the exact definition of RAS and has been reported to be 18 to 30 percent depending on the definition and patient population [5,17,18].

Many studies of RAS epidemiology were done before the 2019 consensus definition [1]. Therefore, the term RAS used in these studies may not have strictly adhered to the ISHLT definition; frequently the diagnosis was based solely on pulmonary function tests without considering imaging findings. For example, in two case series monitoring bilateral lung transplant recipients, 30 percent of new-onset CLAD cases were categorized as RAS based on either ≥10 percent drop in total lung capacity (TLC) [17] or ≥20 percent decrease in forced vital capacity (FVC) from baseline at the time of CLAD onset [18]. In a separate two-center cohort study of single and bilateral lung transplant recipients, the overall prevalence of RAS based on ≥20 percent decrease in FVC was 36 percent, underscoring the possibility that single-lung transplant recipients are more likely to have FVC loss due to native lung pathology [19]. Similarly, 28 percent of patients with CLAD (approximately one-third of whom were single-lung transplant recipients) had RAS using the criteria of either ≥10 percent TLC drop or increased forced expiratory volume in one second (FEV₁)/FVC ratio [20].

Other studies have combined PFT information with CT imaging. In the previously described two-center retrospective cohort study, in patients with FVC loss ≥20 percent and an evaluable CT scan, characteristic RAS features of septal thickening/reticulation or the combination of either ground-glass or consolidative opacities with pleural abnormality (thickening or small effusion) were noted in 60 percent of bilateral recipients and 62 percent of single-lung recipients at the onset of CLAD [19]. However, in a separate study of 268 lung transplant recipients with CLAD, only 18 percent had RAS at presentation defined by CT changes in conjunction with either TLC or FVC loss [4].

A study that stringently applied the 2019 International Society of Heart and Lung Transplantation (ISHLT) guidelines to a single-center cohort of bilateral lung transplant recipients identified RAS in 9.2 percent and mixed phenotype in 5.2 percent of patients at the time of CLAD onset [5]. Using these ISHLT guidelines in single-lung transplant recipients determined the presence of RAS in 19 percent and mixed phenotype in 3 percent of patients with CLAD [6]. Both of these studies found that the identification of persistent fibrotic-like opacities on chest CT alone was as good or better at predicting prognosis than the combination of CT opacities and pulmonary function tests. Persistent CT changes may also appear in patients with stable spirometry and appear to portend a similar prognosis to FVC loss [19,21].

It is important to note that patients can progress from one CLAD phenotype to another, with development of fibrosis on chest CT demonstrating the most prognostic significance. We therefore recommend ongoing monitoring of patients with chest CT scans after CLAD onset.

In one study, 22.5 percent of patients with CLAD transitioned from one phenotype to another. A non-RAS/mixed to a RAS/mixed phenotype transition was seen in 7 percent of CLAD patients. The development of fibrotic-like opacities on chest CT was considered the main predictor of worse survival [22]. In a separate study, 11 percent of patients with BOS at CLAD presentation eventually progressed to a mixed phenotype with both obstructive and restrictive physiology [4]. Similarly, another analysis showed that 13 percent of CLAD patients who initially had preserved lung volumes on chest CT later developed significant restriction, which was associated with worse survival.

ETIOLOGY AND RISK FACTORS — Many studies have not distinguished between RAS and bronchiolitis obliterans syndrome (BOS) in assessment of specific etiologies and predictors. Additional risk factors of CLAD in general are discussed separately. (See "Chronic lung allograft dysfunction: Bronchiolitis obliterans syndrome".)

The contribution of specific risk factors to development of RAS is described here, although all of the studies were published prior to the ISHLT CLAD phenotyping guidelines [1] and used less stringent diagnostic criteria.

●Alloimmunity – Alloimmunity is considered to be the most important risk factor of CLAD and is thought to play a major role in development of both BOS and RAS. Acute cellular rejection has been described as a key independent risk factor of CLAD overall [23]. In three studies that assessed risk factors of BOS and RAS separately, acute cellular rejection was associated with both RAS and BOS in one study [17], with only RAS in another [24], and with neither in the third study [25]. Antibody-mediated rejection is generally defined by the presence of donor-specific antibodies (DSA) with concurrent allograft dysfunction and injury and represents another manifestation of alloimmunity. Presence of DSA was strongly associated with both BOS and RAS in at least two studies [25,26]. Specific features of DSA including persistence, DQ antigen specificity and complement-binding capacity were particularly associated with CLAD, although this analysis did not distinguish between phenotypes [26]. Tissue-bound intragraft DSA appear to be more abundant in RAS compared with BOS, providing further evidence that augmented alloimmunity likely plays a role in RAS pathogenesis [27].

●Infections – Infections are thought to augment the risk of CLAD through direct lung injury or through potentiation of alloimmunity. However, specific associations between infections and RAS remain poorly understood. Overall infections and pseudomonal colonization were associated with both BOS and RAS in one study [24].

●Recurrent lung inflammation and injury – The pathogenesis of RAS is thought to involve a greater magnitude and possibly a more distal localization of the recurrent and chronic inflammation and injury that takes place in most lung allografts. Studies showing diffuse alveolar damage [13,14] and acute fibrinous and organizing pneumonia [9,15,28] as pathologic entities that precede RAS or appear in conjunction with RAS, particularly when present beyond the early posttransplant period, support this notion. Several studies have also shown increased inflammation in the context of RAS as compared with BOS, based on bronchoalveolar lavage (BAL) eosinophilia [24,29]. Translational research studies have identified elevations of multiple inflammatory and injurious proteins [30-32] as well as markers of epithelial cell death [33] in RAS compared with BOS.

●Native lung disease – Native lung disease may play a role as a predictor of later CLAD phenotype: Chronic obstructive pulmonary disease and interstitial lung disease patients had a higher risk of RAS compared with cystic fibrosis in one study [25], although this effect was not detected in other studies [17,24].

CLINICAL PRESENTATION — The presenting symptoms of RAS have not been catalogued in detail. Some patients present with an asymptomatic decline in lung function detected on routine posttransplant surveillance. Symptomatic patients report dyspnea, mainly on exertion, and/or a cough that is typically nonproductive. RAS frequently has an acute onset of respiratory symptoms, sometimes accompanied by fever. Multiple such acute exacerbations may occur with only partial, if any, recovery and stepwise loss of lung function [34]. Bronchiolitis obliterans syndrome (BOS) tends to have similarly nonspecific symptoms, but with a more insidious onset. (See "Chronic lung allograft dysfunction: Bronchiolitis obliterans syndrome".).

Physical examination may identify tachypnea, hypoxemia, and crackles on auscultation.

EVALUATION — Clinical concern for CLAD generally arises when routine monitoring identifies a decline in the patient's forced expiratory volume in one second (FEV₁) ≥10 percent compared with prior values (potential CLAD (figure 1)) [1] (see 'Chronic lung allograft dysfunction (CLAD)' above). Evaluation is directed at characterizing whether the pattern associated with the decline in FEV₁ is obstructive or restrictive, assessing the severity and persistence of decline, examining chest imaging for new abnormalities, and excluding alternate causes for these findings.

Clinical evaluation — The history should include questions to characterize new respiratory symptoms and associated nonrespiratory symptoms (eg, fever, sore throat, myalgias) that might suggest upper or lower respiratory tract infection, and a review of new or ongoing medications that could be associated with pneumonitis, such as sirolimus or everolimus. (See "Maintenance immunosuppression following lung transplantation" and "Viral infections following lung transplantation".)

Patients should be asked about chest or abdominal pain that could cause inadequate inspiration with resultant restriction on pulmonary function testing.

The physical examination should look, in particular, for rash, focal wheeze, evidence of fluid overload, and/or cardiac dysfunction. Marked weight gain since transplant may be an alternate cause of restriction if the development of obesity clearly tracks with the decline in lung function.

Laboratory studies — The main role of laboratory testing is to identify evidence of allograft rejection and exclude processes in the differential diagnosis, such as infection, drug reaction, heart failure, or fluid overload due to impaired liver or kidney function. Testing may include complete blood count and differential, kidney and liver function tests, brain natriuretic peptide, a respiratory viral panel, cytomegalovirus polymerase chain reaction, and donor-specific anti-human leukocyte antigen (HLA) antibodies. Concurrent donor-specific anti-HLA antibodies can be found in patients with RAS, but laboratory studies are often normal. Peripheral blood eosinophilia should raise consideration of a drug reaction or fungal infection.

Calcineurin inhibitor levels should be reviewed to ensure they are in the target range.

Pulmonary function in RAS — The stringent consensus definition of RAS includes a 10 percent drop in total lung capacity (TLC), compared with the posttransplant baseline, at the time of CLAD onset. The TLC baseline is defined as the average of TLC measurements obtained at the same time as, or temporally close to, the baseline FEV₁ measurements. The International Society for Heart and Lung Transplantation (ISHLT) guidelines recommend measuring TLC in all lung transplant recipients at three and six months posttransplant and then annually, in order to establish the baseline value.

If TLC measurements are not available, a diagnosis of RAS is suggested when the FVC has dropped more than 20 percent from baseline and the FEV₁/FVC ratio is normal or increased. However, one must be cautious in using these criteria for RAS diagnosis as other physiologic changes (ie, air trapping due to obstruction) may reduce the FVC [3,18-20].

The accuracy of CLAD phenotyping is limited when lung volumes are not available due to either center-specific protocols or patients’ inability to perform such measurements. Additionally, single lung transplant patients pose a particular challenge for assessment of spirometric and volumetric changes over time due to the contribution of the native lung and possible progression of native lung disease [6]. Imaging may therefore be particularly important in single lung transplant recipients with suspected CLAD.

Imaging — The primary goal of imaging at the time of suspected RAS onset is to evaluate the presence and characteristics of parenchymal and pleural changes. Additionally, imaging should rule out other causes of restriction such as large pleural effusions, ascites, or diaphragmatic elevation.

Persistent fibrotic-like opacities are a key component of the RAS diagnosis. These opacities are characteristically parenchymal or pleural and are consistent with possible fibrosis and deemed likely to cause the restrictive changes, as opposed to bronchiectatic changes often seen with bronchiolitis obliterans syndrome (BOS) (image 1) [35]. The opacities in RAS have been described as reticulations, ground-glass opacities, or consolidations, with upper lobe predominance per some reports, although this localization has not been seen in all case series [17,19,36]. These opacities should be persistent for at least three months, to differentiate them from reversible infection-related changes [35].

The characterization of opacities in RAS is subjective. Developing more objective and quantitative approaches in the future will be important towards standardizing the diagnosis of RAS [37]. Furthermore, several studies have shown that computer-based strategies, involving different levels of machine learning, can be used for quantification of volume and density changes on chest computed tomography (CT) scans that correlate with CLAD phenotype and/or prognosis [38-43]. Incorporation of such computerized technologies to radiologic assessments may be helpful in future approaches for CLAD diagnosis.

The ISHLT guidelines recommend a surveillance chest CT be done at minimum six months posttransplant as a baseline for evaluation of future changes for CLAD phenotyping [1]. A more frequent CT scan surveillance program is used by some transplant centers. Continued chest CT monitoring after CLAD onset may identify potential worsening or de novo development of fibrotic-like opacities [22].

Bronchoscopy — Bronchoscopy, bronchoalveolar lavage, and transbronchial biopsies are performed as part of the evaluation of potential CLAD. However, we generally avoid transbronchial biopsies in patients with established RAS with marked restriction, due to the higher procedural risks.

Typically, bronchoscopy in RAS reveals normal airways, with minimal and nonpurulent secretions. The presence of purulent secretions suggests infection or aspiration, although these may coexist with and/or trigger RAS. Bronchoalveolar lavage (BAL) cultures are usually negative (unless there is concurrent infection). BAL neutrophilia or eosinophilia may be present in RAS [29,31], although these should also prompt consideration of alternate diagnoses such as infection or drug reaction.

Transbronchial biopsies are useful to rule out concurrent acute cellular rejection, especially if presenting within the first two years posttransplant. Pathologic findings of diffuse alveolar damage or organizing pneumonia are often found in RAS. However, acute fibrinous and organizing pneumonia or pleuroparenchymal fibroelastosis are rarely identified on transbronchial biopsy, likely due to inadequate sampling and heterogeneous distribution of the pathology. Transbronchial biopsies may also rarely reveal recurrence of native lung disease, such as sarcoidosis.

Cardiac evaluation — If pulmonary edema is clinically suspected, echocardiography may reveal cardiac dysfunction or pericardial effusion. Further investigations such as cardiac magnetic resonance imaging or cardiac catheterization with pressure measurements may be required for suspected constrictive pericarditis.

Gastrointestinal and swallowing investigations — If recurrent aspiration, reflux, or gastroparesis are clinically suspected, further consultation and testing may be helpful. (See "Physiologic changes following lung transplantation", section on 'Oropharyngeal dysphagia, gastroesophageal reflux, and gastroparesis'.)

DIAGNOSIS — The diagnosis of CLAD due to RAS is based on a ≥20 percent decline in forced expiratory volume in one second (FEV₁) that is persistent for three months after the first value is obtained and associated with restriction (total lung capacity [TLC] <90 percent of the posttransplant baseline, which is the average of the two TLC values measured at, or near, the same time as the two baseline FEV₁ measurements), chest computed tomography (CT) showing persistent fibrotic-like opacities, and no obstruction (FEV₁/forced vital capacity [FVC] ≥0.7) (table 1) [1] (see 'Definitions' above). Other potential causes of lung function decline should be excluded. (See 'Differential diagnosis' below.)

DIFFERENTIAL DIAGNOSIS — The differential diagnosis of RAS includes other phenotypes of CLAD (table 1) (see 'CLAD phenotypes' above), primarily the mixed phenotype, which is characterized by reduced TLC, radiologic opacities, and also obstruction with forced expiratory volume in one second (FEV₁)/forced vital capacity (FVC) <0.7. The clinical importance of distinguishing RAS from mixed phenotype is unclear at this time, although this may change with the advent of more specific therapies in the future.

The differential diagnosis of RAS further includes non-CLAD causes of restricted ventilation such as obesity, pleural disease, heart failure, cancer, drug toxicity, ascites, diaphragmatic dysfunction, and progression of native lung disease in single lung recipients. Evaluation for these processes includes physical examination for muscle weakness, ascites, or weight gain, chest imaging looking for pleural effusion, evidence of new interstitial lung disease, or airway occlusion, and bronchoscopy with bronchoalveolar lavage and transbronchial lung biopsy. (See "Chronic lung allograft dysfunction: Bronchiolitis obliterans syndrome".)

PREVENTION — The general approach in CLAD prevention is to focus on aggressive treatment of known risk factors (see 'Etiology and risk factors' above), although no direct evidence exists that treatment of risk factors actually decreases or delays CLAD. There are no known specific preventive measures for RAS, and we recommend applying preventive measures to all lung transplant patients through optimization of immunosuppression and reduction of allograft injury. (See "Chronic lung allograft dysfunction: Bronchiolitis obliterans syndrome", section on 'Prevention'.)

TREATMENT — Current therapies for RAS are generally ineffective.

Our approach — The initial approach, which matches the approach to treatment of bronchiolitis obliterans syndrome (BOS), is to treat concurrent processes, such as acute cellular rejection, antibody-mediated rejection, infection, or gastroesophageal reflux disease. For patients with a new diagnosis of possible, probable, or definite CLAD, we typically initiate oral azithromycin, 250 mg three times weekly. If bacterial infection is also suspected, we may precede this with a five-day treatment course of azithromycin (500 mg on day 1 and 250 mg days 2 to 5). In addition, we review the maintenance immunosuppressive regimen for adequacy and ensure that serum levels of the various immunosuppressive agents are appropriate. We may switch maintenance immunosuppression to alternate agents in the same drug class. Augmented immunosuppression is often used, although it is less likely to be of benefit in the absence of documented acute cellular rejection or antibody-mediated rejection and may be detrimental by increasing the risk of infections (see "Chronic lung allograft dysfunction: Bronchiolitis obliterans syndrome", section on 'Treatment of BOS'). We do not use total lymphoid irradiation or extracorporeal photopheresis as employed by some centers [44,45] due to the paucity of data supporting efficacy of these interventions.

Azithromycin is used for all forms of CLAD, based on studies showing better lung function and longer time to CLAD in patients initiated on this therapy at the time of transplant, as well as improved lung function with treatment initiated at the time of CLAD onset [46-49]. However, these studies included all CLAD patients and did not make the distinction between BOS and RAS, so the specific effect of azithromycin in RAS is not known. (See "Chronic lung allograft dysfunction: Bronchiolitis obliterans syndrome", section on 'Azithromycin'.)

Retransplantation — Given the disappointing results with most therapeutic approaches to date, retransplantation is an important consideration for many patients with RAS. Unfortunately, RAS is associated with worse post-retransplant survival compared with BOS [50]. Thus, a delicate balance is needed between referring patients with RAS for retransplantation early, while recognizing the poor retransplantation outcomes for RAS. This higher risk profile of RAS lung transplant candidates needs to be considered when added to other risks such as being on a ventilator or on extracorporeal life support.

Future directions — Given the extensive parenchymal fibrosis in RAS, antifibrotic agents (eg, pirfenidone, nintedanib) are now being assessed as potential therapies. Preliminary case reports and case series show potential benefit [51-53], but results from ongoing trials will be important to draw more firm conclusions. mTOR inhibitors have been used for putative antifibrotic properties, although the scant observational evidence does not appear encouraging [54].

PROGNOSIS — Several studies have shown that the presence of restriction and lung opacities at the time of CLAD onset portend a poor long-term prognosis. While RAS patients have, overall, a worse survival compared with patients with bronchiolitis obliterans syndrome (BOS), few studies to date have assessed the differences in outcome between the newly, more precisely defined phenotypes [5,7]. Studies of CLAD prognosis based on phenotype include the following:

●Several studies suggest that lung parenchymal opacities on chest imaging portend a worse prognosis [17,19]. As an example, among 156 lung allograft recipients with CLAD, those with RAS had a median survival of 541 days, compared with a median survival of 1421 days in all other CLAD patients [17]. Additionally, development of opacities may precede the drop in lung function and may predict later development of RAS [55,56].

●In a separate series of 71 recipients with CLAD, those with RAS had a median survival of 8 months compared with 35 months in those with non-RAS CLAD [20]. A subsequent study demonstrated that a forced vital capacity (FVC) ≤80 percent at the time of CLAD onset was also associated with a decreased three-year survival of 9 percent compared with 48 percent in all other CLAD patients [18].

●A study of 53 patients with RAS showed that lower lobe predominant or diffuse opacities on chest computed tomography (CT), increased neutrophils or eosinophils in the bronchoalveolar lavage (BAL), history of lymphocytic bronchiolitis on transbronchial biopsy, and identification of a specific trigger for CLAD represented predictors of shorter survival after RAS onset [57].

●For bilateral lung recipients phenotyped based on the 2019 International Society of Heart and Lung Transplantation (ISHLT) classification, the median allograft survival post-CLAD onset was 372 days for recipients with RAS and 500 days for those with BOS [5]. In a separate study of single-lung transplant recipients with CLAD, median survival times after CLAD onset were 833 days for BOS, 738 for RAS, and 555 for mixed phenotype. Patients with undefined CLAD phenotype had a higher median survival of 1563 days, suggesting other mechanisms of decreased graft function in this group [7].

●A multicenter study of 39 bilateral lung transplant patients with RAS and 31 with mixed phenotype according to 2019 ISHLT classification showed a graft survival of 89 percent at six months, 79 percent at nine months and 61 percent at 12 months after CLAD diagnosis. Graft survival did not differ between RAS and mixed phenotypes [58].

The rate of progression of allograft decline after RAS onset can vary. A fulminant drop, a step-wise worsening, and a gradual worsening have all been described, although the prevalence of each of these patterns is unclear [35]. While most of the studies have focused on the phenotype at CLAD onset, patients may progress from one phenotype to another as the process evolves [22]. The appearance of fibrotic-like opacities portends a worse prognosis [22]. Also, more specifically, recipients with progression from BOS or RAS to the mixed phenotype appear to have a subsequent survival comparable to that after RAS onset [4]. Increased susceptibility of CLAD patients to superimposed infections and further rejection may account for some of these differences. More data are needed to better understand the natural progression of the different CLAD phenotypes over time.

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Lung transplantation".)

SUMMARY AND RECOMMENDATIONS

●CLAD definition – Chronic lung allograft dysfunction (CLAD) is defined as a persistent and irreversible decline in forced expiratory volume in one second (FEV₁) by ≥20 percent of the posttransplant baseline, in the absence of other diagnostic explanations (figure 1). (See 'Chronic lung allograft dysfunction (CLAD)' above.)

●CLAD subtypes – Bronchiolitis obliterans syndrome (BOS) and restrictive allograft syndrome (RAS) are the two main recognized phenotypes of CLAD, with BOS being the most common (≥70 percent) (table 1).

•Bronchiolitis obliterans syndrome (BOS) – BOS is characterized by an obstructive pattern on pulmonary function tests (PFTs) and the absence of radiographic opacities. (See 'Definitions' above.)

•Restrictive allograft syndrome (RAS) – RAS is characterized by a restrictive pattern on PFTs with a decrease in total lung capacity (TLC) to ≤90 percent of the posttransplant baseline (defined as the average of the two TLC values measured at, or near, the same time as the two baseline FEV₁ measurements), absence of obstruction (ie, FEV₁/forced vital capacity [FVC] >0.7), and presence of persistent fibrotic-like opacities on chest computed tomography. (See 'Restrictive allograft syndrome (RAS)' above.)

•Other subtypes – BOS and RAS can be present together as a mixed phenotype. Occasional patients have "undefined" CLAD and present with obstruction and persistent opacities without restriction or obstruction with concurrent restriction but without persistent opacities. Some patients with CLAD may remain unclassified. (See 'CLAD phenotypes' above.)

●Epidemiology – CLAD affects approximately 50 percent of lung transplant recipients at five years, and of those, RAS affects approximately 10 to 30 percent. (See 'Epidemiology' above.)

●Clinical presentation of RAS – RAS can present with an asymptomatic decline in lung function detected on routine posttransplant surveillance or with dyspnea, mainly on exertion, and/or a cough that is typically nonproductive. RAS frequently has an acute onset of respiratory symptoms, sometimes accompanied by fever. In comparison, BOS has similarly nonspecific symptoms, but typically an insidious onset. (See 'Clinical presentation' above.)

●Evaluation – CLAD is generally suspected when routine monitoring identifies a decline in the patient's FEV₁ ≥10 percent compared with prior values or when a patient reports new or worsening dyspnea on exertion or cough. Evaluation is directed at characterizing whether the pattern associated with the decline in FEV₁ is obstructive or restrictive, assessing the severity and persistence of decline, examining chest imaging for new abnormalities, and excluding alternate causes for the change in lung function. (See 'Evaluation' above.)

●Diagnosis – The diagnosis of CLAD due to RAS is based on the demonstration of a persistent, restrictive lung function decline and presence of compatible opacities on chest imaging, as described in the definition (table 1). Other potential causes of lung function decline, such as infection, drug-induced pneumonitis, heart failure, obesity, pleural disease, diaphragmatic dysfunction, cancer, and progressive lung disease in the native lung in single lung recipients, must be excluded. (See 'Definitions' above and 'Differential diagnosis' above.)

●Treatment – Current therapies for RAS are generally ineffective. Therefore, the therapeutic focus should be on addressing concurrent and reversible causes of allograft dysfunction and dyspnea. (See 'Treatment' above.)

•Pharmacotherapy – For patients with a new diagnosis of CLAD with RAS and no evidence of infection, we suggest oral azithromycin (250 mg three times weekly) rather than observation alone (Grade 2C) based on evidence of benefit in the CLAD population as a whole. In addition, we review the maintenance immunosuppressive regimen and ensure that serum levels of the various immunosuppressive agents are appropriate (table 2). Augmented immunosuppression is often used but is rarely of benefit. (See 'Our approach' above.)

•Retransplantation – Because of the often-aggressive course of allograft decline in RAS patients, the option of retransplantation should be discussed early with a prompt referral for retransplant assessment where appropriate. (See 'Retransplantation' above.)

●Prognosis – The rate of progression after RAS onset varies and may be fulminant, incremental, or gradual. The overall prognosis of RAS tends to be worse than that of BOS, with a median life expectancy of 8 to 18 months after RAS onset; the survival of mixed phenotype RAS-BOS is comparable to RAS alone. (See 'Prognosis' above.)