Published ahead of print on April 5, 2007, doi:10.1164/rccm.200610-1485OC
© 2007 American Thoracic Society doi: 10.1164/rccm.200610-1485OC
Pepsin, a Biomarker of Gastric Aspiration in Lung AllograftsA Putative Association with Rejection1 Applied Immunobiology and Transplantation Research Group, Institute of Cellular Medicine, and 2 Epithelial Research Group, Institute for Cell and Molecular Biosciences, University of Newcastle-upon-Tyne, Newcastle-upon-Tyne, United Kingdom; and 3 North West Lung Research Centre, University of Manchester, Manchester, United Kingdom Correspondence and requests for reprints should be addressed to Chris Ward, Ph.D., Institute of Cellular Medicine, School of Clinical Medical Sciences, William Leech Building, Newcastle University, Framlington Place, Newcastle-upon-Tyne NE2 4HH, UK. E-mail: chris.ward{at}ncl.ac.uk
Rationale: Human lung transplantation is a therapeutic option for selected patients with advanced cardiopulmonary disease, but long-term survival is limited by chronic rejection. Persistent acute rejection and gastric aspiration have been implicated as risk factors but there is little or no evidence to date that they are associated. Objectives: We have tested the hypothesis that pepsin, a marker of gastric aspiration, is present in lung transplant recipients, and that high levels are associated with biopsy-diagnosed acute rejection and/or bronchiolitis obliterans syndrome. Methods: Levels of bronchoalveolar lavage (BAL) pepsin were measured by ELISA in 36 lung transplant recipients, 4 normal volunteers, and 17 subjects with unexplained chronic cough.
Measurements and Main Results: Our primary finding was that, compared with control subjects, BAL pepsin levels were elevated in stable lung transplant recipients, subjects with acute rejection, and subjects with bronchiolitis obliterans syndrome. Our secondary finding was that the highest levels were found in recipients with acute vascular rejection grade
Conclusions: We have shown that elevated levels of pepsin, a biomarker of gastric aspiration, are consistently identified in the BAL of lung allografts. The highest levels were seen in patients with
Key Words: lung allograft • gastroesophageal reflux, GER • pepsin • rejection
Human lung transplantation is a therapeutic option for selected patients with advanced cardiopulmonary disease, but long-term survival is limited by the development of obliterative bronchiolitis, recognized physiologically as bronchiolitis obliterans syndrome (BOS) (1). The pathophysiology is poorly understood, but the frequency and severity of episodes of acute vascular rejection are consistently associated with an increased risk of developing BOS (2, 3). Most lung allograft recipients experience at least one episode of acute rejection, frequently in the early stages of the transplant, and this is recognized by clinical symptoms, or from standardized biopsy grading after surveillance bronchoscopy and transbronchial biopsies (4). The biopsy grading includes vascular and airway components, and it is assumed that the latter is a marker of airway rejection, an immune process mediated by immune cells, including alloreactive T cells. It is also increasingly recognized that nonalloimmune mechanisms may contribute to chronic airway injury in lung allografts, and gastroesophageal reflux (GER) has been implicated as a potential cause, not only post–lung transplantation but also in other airway and lung diseases (5–9). It is notable that several open studies of fundoplication surgery by the Duke University group have shown a survival benefit and a delayed onset of BOS in lung transplantation (10–13). We have recently shown for the first time that pepsin, a marker of gastric aspiration, is present in lung allograft recipients without airflow limitation (14). Our previous study was in a small number of patients and was not powered to study possible associations between acute rejection and evidence of gastric aspiration. We have therefore tested the hypothesis that pepsin is present in the lungs of allograft recipients and that high levels are associated with the presence of acute rejection and/or BOS in a prospective study. These data have been presented in part at the European Respiratory Society Annual Congress 2006, the United European Gastroenterology Week 2006, and the British Thoracic Society Winter Meeting 2006, and have been published in abstract form (15–17).
Study approvals were obtained by the local research ethics committees for Newcastle and North Tyneside and Manchester, with separate applications for studies in lung allografts, normal volunteer control subjects and volunteer subjects with unexplained chronic cough (disease controls). After informed consent was received, 36 lung allograft recipients were recruited before undergoing either routine or symptom-driven transbronchial biopsy (TBB) and bronchoalveolar lavage (BAL), with the research sample intercalated within this procedure. Research samples were taken at least 1 week after any preceding bronchoscopic procedure. All patients were receiving standard long-term maintenance immunosuppressive therapy comprising a calcineurin inhibitor, a cell cycle inhibitor, and prednisolone. The patients were not formally investigated for GER, but none reported symptoms suggestive of GER. In particular, there were no reports of heartburn, retrosternal ache, sour taste in the mouth, or pain on swallowing. Thirty patients were treated with acid suppression therapy (Table 1), which is a common empirical therapy in this patient population after transplantation.
Four normal, nonsmoking control subjects (one female; median age, 39 yr; range, 32–46 yr) were recruited from volunteer hospital staff for a research BAL. These patient samples are the same as reported in previous work (14). Seventeen subjects with normal lung function, under investigation for unexplained chronic cough (nine females; median age, 54 yr; range, 25–81 yr) were recruited as "disease controls" for the study. Patients were investigated by an algorithm similar to that outlined in the European Respiratory Society guidelines (18), with history, subsequent investigations, and treatment trials targeted sequentially at the sites of cough afferents. Two patients were diagnosed with cough variant asthma after positive bronchial provocation testing with histamine and subsequent response of symptoms to inhaled corticosteroids. One patient was diagnosed with bronchiectasis after high-resolution computed tomography scanning of the thorax. In four patients, no cause for cough could be found despite investigation into possible pulmonary, ear, nose, and throat (ENT), and gastroesophageal etiology and appropriate treatment trials. These patients were classified as having "idiopathic" or unexplained cough. The remainder of the patient group had a diagnosis of GER disease (GERD). This was based on symptomatic relief of cough with antireflux treatment trials and return of symptoms when medication was withdrawn (n = 2) or objective evidence of excessive reflux from either standard 24-h pH (19) (n = 2) or impedance/pH monitoring (20) (n = 6). Of the patients who had impedance/pH monitoring, four also had grade 1 esophagitis at gastroscopy.
Bronchoscopy, BAL, and TBB BAL was standardized to a 3 x 60 ml procedure. The BAL sample was split, and clinical microbiology was assessed in a standardized fashion, which included the use of selective agars and extended culture for bacterial, fungal, and Legionella spp. Differential cell counts were made on Giemsa-stained cyto-centrifuge preparations. Cell-free BAL supernatants were prepared by centrifugation (10 min, 1,500 rpm, 4°C); aliquots were snap-frozen by immersion in liquid N2, and stored at –80°C before ELISA.
Pepsin/Pepsinogen ELISA We have evaluated our assay for potential cross-reactivity with other BAL analytes, including neutrophil degranulation products such as neutrophil elastase. These were shown not to interfere in our assay. We have previously identified a potential for fibrinogen, an acute-phase inflammatory marker, to interfere with our pepsin ELISA, and although theoretically possible, any such interference would have led us to underestimate the levels of BAL pepsin in lung allografts (21).
TBB Processing
Statistical Analysis
Possible Confounding Variables
A model including demographic variables.
Patient demographics are summarized in Table 1. BAL microbiology and TBB pathological rejection assessments are summarized only for the lung transplant recipients. In the lung transplant recipients, 12 patients had stable lung function with no evidence of clinically significant acute rejection (6 had grade A0, 6 had grade A1) or BOS, or clinical evidence of infection (stable lung transplant recipients). One patient in this group had asymptomatic infection with Pseudomonas aeruginosa. Twelve patients had biopsy-proven acute rejection (A2 or greater) without BOS and 12 patients were diagnosed as having BOS.
BAL Data
BAL Pepsin
Comparison of lung transplant recipients with stable allograft function, acute rejection, and BOS.
When compared with normal and chronic cough control subjects (both GERD and no-GERD), BAL pepsin levels were significantly raised in the grade A2 or greater acute rejection group (vs. normal, p = 0.004, and vs. both chronic cough groups, p < 0.001). Grades of airway inflammation were highest in patients with acute rejection: median B score 2 (range, 0–2) versus 0 in stable transplant patients (range, 0–1), p = 0.009. The BOS patients had a median score of 1 (range, 0–1) (Table 1). Pepsin levels were not significantly different in patients treated with a maintenance dose of acid suppression therapy (median, 8.7; range, 0–51.7 ng/ml) compared with patients who were untreated (median, 6.0; range, 0–15.6 ng/ml; p = 0.3).
Potential Confounding Variables
A model including demographic variables.
We have recently shown for the first time that pepsin, a marker of gastric aspiration, is present in lung allograft airways (14). The primary finding of our present study confirmed this in a larger, independent cohort, with more control data. Our secondary finding showed that the pepsin levels were highest in lung allograft recipients with histologically verified grade A2 or greater acute rejection, and that subjects with acute rejection had the highest grades of airway inflammation. The potential significance is that aspiration and acute rejection have long been implicated risk factors for BOS, and our study provides evidence that supports the importance of alloimmune and nonalloimmune injury to lung allografts (22). Our study contributes to an emerging paradigm shift in lung transplant rejection indicating that nonalloimmune causes of injury may be important in addition to the alloimmune mechanisms that have been the traditional focus of research and therapeutic intervention (22). Accumulating clinical evidence suggests that GER with sequential aspiration may be an important source of injury and this is underlined by a recent pediatric lung transplantation study from Great Ormond Street, United Kingdom, which showed that all subjects had evidence of GER post-transplant, except one individual who had received a previous fundoplication (23). This observational study, in a limited number of patients, reported that all subjects with acute rejection had moderate to severe GERD and that no episodes of acute cellular rejection occurred post–Nissen fundoplication (23). The implication from this, as well as our finding of higher levels of pepsin in adult patients with grade A2 or greater acute rejection, is that nonalloimmune injury may contribute to a pathology previously attributed solely to alloimmunity. Our finding that pepsin levels were elevated in all transplant recipients compared with control subjects suggests that aspiration may be an ongoing source of injury. It has been previously shown that other nonalloimmune injuries, such as ischemia–reperfusion, may further predispose the transplanted lung to alloimmune injury (24), and our study supports the importance of both alloimmune and nonalloimmune damage to overall injury in lung transplantation. It is of interest that, both in the present study and in our previous work (14), we found evidence of gastric aspiration even in patients who were treated with maintenance proton pump inhibitors. Approximately 12 to 20% of patients with GERD are resistant to acid suppression therapy (25), and, in addition, it is important to recognize that proton pump inhibitors do not prevent reflux per se but rather act to cut down on acidic reflux. Mildly acidic, neutral, or alkaline reflux, which will still contain pepsin is not the target of proton pump inhibitor therapy but may still be an important source of aspiration injury that requires a biomarker approach to monitor. It is therefore understandable that these allograft patients will continue to reflux and aspirate when on maintenance acid suppression therapy. The lack of any pepsin in the BAL samples from patients with chronic cough, even when 10 of 17 were diagnosed with GERD, can be explained by the fact that a diagnosis of GERD does not mean that patients are refluxing out of the esophagus and hence aspirating. Even if the refluxate reaches the upper airway, it is almost certainly cleared by a hyperactive cough reflex (26). In contrast, lung allograft recipients are regarded as being especially vulnerable to aspiration because they have gastroparesis, impaired cough (27), and an impaired mucociliary clearance (28). Overall, we believe that our work is consistent with the findings of the Duke University group, which has repeatedly shown that early fundoplication is necessary to achieve clinical benefit in lung transplantation (11). Clinical research in human patients such as ours is often limited to studies of association and consideration of animal models is warranted. Recent work in a rat model of lung transplantation from the Duke group has shown that allografts challenged with chronic aspiration demonstrated severe grade 4 acute rejection with significant monocyte infiltration, fibrosis, and lung destruction (29). Aspiration was also associated with increases in CD8+ T cells and this "proof of principle" study indicates that aspiration may indeed lead to pathological changes previously attributed to T-cell, alloimmune-based mechanisms (30, 31). For this study, we defined acute rejection as grade A2 or greater because this is the level to trigger altered clinical management at our center. However, we do appreciate that there has been debate on the importance of minimal acute rejection (A1) as a risk factor for BOS (32, 33). Our current data indicate that, in grade A1 rejection, pepsin levels were not raised. This could support the concept that aspiration injury is additive to alloimmunity, and where present, results in higher grades of injury. This is speculative, however, and our suggestion of a possible association between acute rejection and aspiration requires future studies that may, in particular, better define the role of A1 rejection. Our present study did not show higher levels of pepsin associated with BOS. A potential reason for this is that subjects with BOS in this cross-sectional study were significantly further out post-transplant compared with the other groups, at a time when gastroparesis specifically associated with perioperative vagal damage and early post-transplant events would be less marked. Further insights into this would be provided through longitudinal assessments of gastric aspiration and its relationship with allograft dysfunction. Our present findings and previous work indicate that a biomarker approach to studying gastric aspiration in lung allografts is informative and practicable. There is also increasing evidence to indicate that duodenal–gastroesophageal reflux and aspiration can also occur in lung allografts using bile salts as a biomarker (34), and that this associated with the development of BOS (35). One practical reason why such biomarker approaches may be important is that calcineurin inhibitors and other drugs that may be augmented to treat presumed alloimmune rejection have adverse effects on gastric motility and may in fact predispose patients to aspiration. We see the present study as one that highlights an issue relevant to lung transplantation and other areas of respiratory medicine but that will benefit from more extensive longitudinal data collections, ideally including other objective data on GERD and potential aspiration. In the present study, we prospectively elected to monitor a biological marker of gastric aspiration into the lung rather than using techniques that monitor movement into the esophagus, and 24-hour esophageal pH data were not collected in this group of lung transplant patients. It is of interest that our data on patients with chronic cough, with pH studies indicating the existence of GERD, confirm the fact that GERD does not always lead to aspiration. Our decision to use a biomarker approach to document aspiration was calculated, and informed by our interpretation of the current gastroenterology literature, which shows that the precise role of pH and impedance measurements remains to be fully characterized. In particular, pH-based approaches to studying GERD adopt a cutoff point of pH 4.0, which potentially ignores neutral and "mildly" acidic reflux, while not documenting aspiration into the lung. We feel that acid and nonacid reflux may well be physiologically important, especially if accompanied by aspiration in what is an especially vulnerable patient population. Further work on the importance of reflux and aspiration in lung allografts is required and long-term longitudinal follow-up studies in patients who are assessed for gastric aspiration should be performed to determine if this represents a risk factor for chronic allograft dysfunction and BOS. Biomarker approaches have the potential to inform research and possibly clinical management, and, if fundoplication is confirmed as useful prophylaxis in lung allografts, markers of aspiration may contribute to identifying patients who might derive clinical benefit. It is also of potential interest that the macrolide antibiotic azithromycin, which in open trials has been shown to reverse BOS in some patients through unknown mechanisms (36, 37), is potentially a promotility agent (38, 39).
The authors thank Dr. Fiona Black for pathological assessments. Jeffrey P. Pearson and Chris Ward contributed equally to this manuscript.
Supported by the Newcastle-upon-Tyne Hospitals Special Trustees, a European Respiratory Society fellowship (D.M.M.), a Medical Research Council fellowship (I.A.F.), and a GlaxoSmithKline grant-in-aid (R.S.). C.W. is currently supported by a Medical Research Council project grant. Originally Published in Press as DOI: 10.1164/rccm.200610-1485OC on April 5, 2007 Conflict of Interest Statement: None of the authors has a financial relationship with a commercial entity that has an interest in the subject of this manuscript. Received in original form October 17, 2006; accepted in final form April 2, 2007
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TOP
ABSTRACT
AT A GLANCE COMMENTARY
A2 (median, 11.2; range, 5.4 – 51.7 ng/ml; normal median, 1.1; range, 0–2.3 ng/ml; p = 0.004). 
