|
|
||||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the Division of Pulmonary and Critical Care Medicine (Drs. Martinu, Steele, and Palmer), Department of Pathology (Dr. Howell), Department of Thoracic Surgery (Dr. Davis), Duke University Medical Center, Durham, NC.
Correspondence to: Scott M. Palmer, MD, MHS, Duke University Medical Center, Box 3876, Durham, NC 27710; e-mail: palme002{at}mc.duke.edu
Abstract
Rationale: The main hindrance to long-term success of lung transplantation is bronchiolitis obliterans syndrome (BOS), generally thought to be a manifestation of chronic allograft rejection. BOS is associated histologically with epithelial injury, bronchocentric mononuclear inflammation, and fibrosis of small airways known as bronchiolitis obliterans (BO). Few studies have directly compared clinical, radiographic, and histologic findings of BOS and BO, particularly in the era of improved immunosuppression and infection prophylaxis. Patients undergoing pulmonary retransplantation for BOS provide a unique opportunity to investigate these relationships.
Methods: All patients who underwent pulmonary retransplantation for BOS from 1992 to 2004 at Duke University Medical Center were reviewed. Pathology findings in explanted lung allografts were compared with clinical, radiographic, and transbronchial biopsy data.
Results: Over the 12-year study period, 12 patients underwent pulmonary retransplantation for BOS. The median time to BOS was 517 days (intraquartile range, 396 to 819.8 days). BOS scores prior to retransplantation were 2 in 2 patients and 3 in 10 patients. We developed a semiquantitative scoring system for epithelial, inflammatory, and fibrotic changes in affected airways to permit better comparison between BO and BOS. Somewhat surprisingly, only 50% (6 of 12 patients) had severe fibrotic changes, although all had some degree of epithelial injury, fibrosis, or inflammation centered around the bronchi and bronchioles. Furthermore, pathology findings other than BO were present in most explanted allografts and included cholesterol clefts (n = 4), focal invasive aspergillosis (n = 1), interstitial fibrosis (n = 2), and chronic vascular rejection (n = 1).
Conclusions: In this series of patients with advanced BOS undergoing retransplantation, at least some degree of BO was present in all explanted allografts. However, the degree of epithelial changes, fibrosis, and inflammation present among affected bronchi varied considerably. Furthermore, a wide range of pathologic processes of potential clinical significance were evident in half of the patients. We conclude that significant histologic heterogeneity exists among patients undergoing retransplantation for BOS, potentially contributing to the variability of patient responses to treatment.
Key Words: biopsy bronchiolitis obliterans heart-lung transplantation histology homologous transplantation lung transplantation pathology radiography
The first human pulmonary allotransplantation was performed by Hardy and colleagues1 in 1963. During the following 4 decades, lung transplantation evolved into the treatment of choice for end-stage lung diseases in select populations. Survival in pulmonary transplant recipients has been slowly improving, mainly due to careful patient selection, improved surgical techniques, better immunosuppressive regimens, and more aggressive antibiotic prophylaxis and treatment. The twentieth report by the registry of the International Society for Heart and Lung Transplantation (ISHLT) reports survival rates at 3 months, 1 year, 5 years, and 10 years of 83%, 74%, 45%, and 23%, respectively.2 The main hindrance to long-term success of lung transplantation remains bronchiolitis obliterans syndrome (BOS), thought to be the pulmonary-specific manifestation of chronic rejection, which directly or indirectly contributes to most late deaths.2 The presumed underlying histopathologic findings are referred to as bronchiolitis obliterans (BO) and include small airway injury with inflammation, fibrosis, and obliteration. The sensitivity for detection of BO by radiographic studies345 or transbronchial biopsy678 is poor; therefore, BOS is used as the clinical surrogate for histologic BO.9 BOS is defined as FEV1 drop to < 80% of the posttransplant baseline when other causes of graft deterioration have been excluded. The severity of BOS is graded based on the degree of impairment in FEV1, per ISHLT consensus10: grade 0, no BOS (FEV1
80% of baseline FEV1 after transplantation); grade 1, mild BOS (FEV1 66 to 80% of baseline FEV1 after transplantation); grade 2, moderate BOS (FEV1 51 to 65% of baseline FEV1 after transplantation); and grade 3, severe BOS (FEV1
50% of baseline FEV1 after transplantation).
Evidence linking BOS to BO is based on several primary pathology studies conducted in the 1980s and corroborated by later articles from the 1990s.11121314151617181920212223 Only a few of these studies11121516 actually compared clinical findings with pathology. They were all predominantly autopsy reports of lung and heart-lung transplants and included small numbers of patients (1 to 6 patients), except for a few larger reports with 10 patients,22 11 patients,19 and 17 patients.21 None of these studies evaluated the severity or heterogeneity of BO pathology. Two abstracts by Chhajed et al2425 questioned the use of BOS as the main surrogate for BO by showing poor correlation between BOS and BO on autopsies: in the abstract25 from 2004, only 26 of 43 patients with presumed BOS had actual BO on postmortem pathology. Although autopsy has been considered the primary "gold standard" in the histologic diagnosis of BO, autopsy studies are potentially flawed by selection bias (only a fraction of lung transplant patients undergo postmortem evaluation) and by the confounding pathologic processes at the time of death, such as overwhelming infections, multiorgan failure, aspiration, or oxygen toxicity.
Patients who undergo pulmonary retransplantation provide a unique opportunity to study the clinical, radiographic, and histologic features of BOS. Pulmonary retransplantation accounts for approximately 3% of all lung transplant operations,2 and represents the only definitive treatment for BOS; in selected patients with BOS, retransplantation offers a reasonable chance for successful long-term patient survival.2627 Explanted lung allografts have not been extensively evaluated, even though several autopsy studies1112171923 have included, but not focused on, a small number of such explanted lungs. The retransplantation population is unique in that it is relatively homogeneous due to stringent selection criteria. These patients undergo similar standard diagnostic procedures including pulmonary function tests, chest radiographic imaging, and bronchoscopy with transbronchial biopsies prior to retransplantation and are relatively clinically stable to qualify for retransplantation.
The aim of this study was to compare pathologic and clinical features of chronic rejection and describe associated findings in explanted pulmonary allografts from patients undergoing retransplantation. A semiquantitative histologic BO grading system was to be used in order to permit more precise description of the affected airways. At the onset of this study, our hypothesis was that there would be a lack of correlation between the clinical severity of BOS, the histologic severity of BO, and the pre-retransplant radiologic and transbronchial biopsy data.
Materials and Methods
This study is a retrospective observational study done at Duke University Medical Center. After institutional research board approval, all adult patients who underwent lung or heart-lung transplantation at Duke University Medical Center between 1992 and 2002 were identified. A total of 460 such patients were found, 19 of whom were lung or heart-lung retransplant recipients (4%). Of these 19 patients, 7 underwent retransplantation for primary graft failure and 12 patients were identified who underwent pulmonary retransplantation for BOS. Transplantation was performed using standard techniques and immunosuppressive regimens.6 Patients underwent posttransplant biopsies according to standard protocol (at 1, 3, 6, and 12 months after transplant and as clinically indicated). Survival for the overall lung and heart-lung transplant as well as the retransplant populations was calculated. Survivals were calculated using the Kaplan-Meier method and compared with the log-rank test.
Demographic, clinical, pathologic, microbiological, and radiologic data were collected and explanted allograft histology was reviewed by an experienced pulmonary transplant pathologist blinded to the patients clinical status. BOS severity was defined using ISHLT criteria.10 All explanted allografts were analyzed for presence of BO as well as any additional pathology findings. Histology sections from multiple lung lobes were evaluated for each patient. The current nomenclature for pulmonary allograft rejection, the ABC system,28 focuses on acute vessel and airway rejection with description of presence and severity of perivascular mononuclear infiltrates (A), and lymphocytic bronchiolitis (B). The nomenclature for chronic rejection (C) is more restricted: it distinguishes between active and inactive lesions (Ca and Cb), which does not capture the broad spectrum of lesions that can be seen in BO. Therefore, to allow a more precise and versatile description of histologic severity of BO, we graded separately the epithelial, inflammatory, and fibrotic changes of affected bronchi and contrasted these results with clinical BOS. The severity of epithelial (E), fibrotic (F), and inflammatory (I) changes (epithelial-fibrotic-inflammatory [EFI]) in all affected airways was assigned grades of 1 (mild), 2 (moderate), and 3 (severe). This grading was based on consensus of three of the authors (T.M., D.N.H., S.M.P.). The traditional C-nomenclature for chronic airway rejection, which differentiates active (Ca) from inactive (Cb) lesions, roughly equates with inflammatory (I) and fibrotic (F) changes in our grading scheme. Our descriptive system also includes information on the presence of epithelial metaplasia and injury (E). The average of the three (E, F, and I) components was calculated for each patient and referred to as mean EFI. For an additional means of evaluating histologic severity, we also graded BO in each patient by estimating the average percentage of affected bronchioles per slide.
As noted previously, all biopsy findings were graded according to conventional scoring systems for acute rejection (defined per ISHLT criteria).28 Evidence of significant vascular intimal proliferation and fibrosis was considered to represent "D grade" chronic vascular rejection.10 We evaluated as well the presence of other confounding pathologic findings in the explanted allografts.
Results
Demographics and Outcomes Data
Characteristics of patients retransplanted for BOS are illustrated in Table 1
. Five patients were men, and seven patients were women. The reasons for initial transplantation were emphysema (n = 3), cystic fibrosis (n = 4), idiopathic pulmonary fibrosis (n = 4), and aluminum pneumoconiosis (n = 1). At the time of retransplantation, 10 patients had severe clinical BOS (grade 3) and 2 patients had moderate BOS (grade 2). The median time to BOS was 517 days (intraquartile range [IQR] 396 to 819.8 days). Of the 12 patients, 1 underwent single-lung retransplantation and 11 patients underwent bilateral lung retransplantation. The mean age at retransplantation was 40 years. Two patients underwent retransplantation during an earlier era in the life of the Duke University Medical Center transplant program (from 1995 to 1999), and 10 patients underwent retransplantation during the later era (from 2000 to 2003). These same time ranges are used by the ISHLT to evaluate outcomes based on era.2 None of our retransplantations were done prior to 1995.
|
|
|
|
Histologic findings other than BO in patients with BOS were common (Table 2) and included superimposed infectious bronchitis as well as chronic bronchitis (one patient), multifocal organizing pneumonia (one patient), focal areas of emphysema (one patient), focal invasive aspergillosis (one patient), arteriopathy (one patient), and extensive interstitial fibrosis (two patients) (Fig 2, bottom middle, E) as well as focal interstitial fibrosis in one additional patient.
Cholesterol clefts were seen in four patient allografts (Fig 2, bottom right, F), with one of these patients also demonstrating lipoid pneumonia. Table 3 summarizes the occurrence of gastroesophageal reflux disease (GERD) prior to retransplantation and possible features of aspiration on pathology specimens. Of the four patients with cholesterol clefts, two had documented GERD before retransplantation (patients 11 and 12). Of the other two patients with cholesterol clefts, one patient (patient 4) had a negative barium swallow test result before retransplantation. The other patient (patient 2) had normal endoscopy findings 3 years before retransplantation, but severe grade IV esophagitis was noted on endoscopy early (within 1 month) after retransplantation. Of the patients who did not have cholesterol clefts on histologic analysis of their explanted lung allografts, four patients had documented GERD, three patients had negative evaluations for GERD, and one patient had no studies for GERD.
|
The last pre-retransplantation transbronchial biopsies, all done > 1 year after the first transplant, demonstrated acute rejection in five patients: three patients had mild acute rejection (patients 1, 3, and 4), one patient had moderate acute rejection (patient 5), and one patient had severe acute rejection (patient 2).
Radiologic Analysis and Correlates
All patients except one underwent CT scans of their chests prior to retransplantation, which included inspiratory and expiratory high-resolution images for six of these patients. In careful review of these scans, only 3 of 12 patients (sensitivity, 25%) had significant changes suggestive of BO, including heterogeneous ground-glass opacities with intralobular thickening, traction bronchiectasis, areas of fibrosis, and pleural thickening. Only one of these three patients had inspiratory and expiratory CT scan views, which did not show air trapping. There was no apparent correlation between CT scan findings and the severity of BO on histopathologic analysis of explanted allografts.
Discussion
In summary, our report reviews a single-center experience with pulmonary retransplantation for BOS and demonstrates the extent to which BOS correlates with pathologic BO. All patients had evidence of significant BO in their explanted allografts, but a broad range of epithelial, fibrotic, and inflammatory changes were seen and several unexpected pathologic findings were identified.
In order to perform a detailed histopathologic analysis of explanted lung allografts of patients undergoing pulmonary retransplantation for BOS, we developed a semiquantitative histologic grading system of affected airways. Our system was aimed at describing more precisely the nature and the severity of BO in the explanted allografts, allowing for an expansion of the C part of the ABC nomenclature, and accounting for the breadth of histology seen among BO lesions. Our system was based on grading the severity of bronchial fibrosis (F) and inflammation (I), thus dividing the inactive/fibrotic (Cb) and active/inflamed (Ca) lesions into mild, moderate, and severe. Furthermore, our system allowed for the description of epithelial changes (E), which often accompany BO. This system was useful to quantify the observations of the authors; however, to improve the strength of this grading system for future studies, validation would need to be performed with at least one additional pathologist. Comparison of the overall EFI grades (mean EFI) and BOS grades demonstrated a fairly good correlation between the histologic and clinical severity of disease. However, in spite of their advanced clinical BOS status, many patients had only mild or moderate fibrotic changes. One would expect that BO in end-stage BOS patients would be of a very fibrotic kind, demonstrating mostly Cb or inactive lesions of presumed burned-out BO. Our results raise the question whether airflow impairment or obstruction in advanced BOS cases could be due to epithelial and inflammatory changes, instead of obliteration of small airways. This could partially explain the heterogeneity of clinical responses to treatment seen in lung transplant therapeutic trials and leads us to ask whether such epithelial and inflammatory changes may be more reversible and responsive to treatment.
Microscopic features of acute rejection, ranging from mild and focal to advanced, were noted in all explanted allografts. In 7 of our 12 patients, histologic acute rejection was not seen on transbronchial biopsies prior to retransplantation, pointing to the limited diagnostic capability of transbronchial biopsies. Interesting lesions of perivascular lymphocytes adjacent to a bronchus in association with lymphocytic bronchiolitis were observed in every patient in this study, suggesting the possibility of coexistence of acute and chronic rejection. Such lesions might simply reflect the fact that bronchi are usually in close proximity to blood vessels in the lungs. Alternatively, these lesions might provide insight into a plausible mechanism by which acute cellular rejection leads to the development of BO via lymphocyte migration from vessel to airway. Prior reports293031 have described features of acute and chronic rejection on the same pathology specimens; however, they did not specify whether these findings were seen as part of a single vessel-airway lesion.
Additional findings other than BO on explanted allografts are described in this study and suggest the possibility that additional diseases could confound the clinical syndrome of BOS. Several infectious changes were seen, such as pneumonia, bronchitis, and focal invasive aspergillosis. Additional findings of cholesterol clefts and interstitial fibrosis were unexpected and are of unclear origin. Extensive interstitial fibrosis was seen in two patients and may represent an unusual form of alloimmune septal injury. Reports have suggested that humoral rejection might occur with higher frequency than originally expected in BOS,32 and autoantibodies have been linked to septal injury in idiopathic pulmonary fibrosis.33
Several studies3435 from Duke University have focused on the significance of reflux and aspiration as possible risk factors for chronic rejection. One study36 even suggested that early Nissen fundoplication may decrease the incidence of BOS. Cholesterol clefts have been described in prior studies1213 of BO pathology and have often been equated with prior aspiration. In their 1991 article, Abernathy and colleagues19 proposed the differentiation of BO into two categories: pure BO and BO with organizing pneumonia. In two of their four cases with organizing pneumonia, foreign material and giant cells were present, suggesting the possibility of aspiration contributing to the development of early BO in these patients. Intriguingly, although cholesterol clefts were identified in four of our transplant pneumonectomy specimens, no foreign material was identified in any of these specimens by polarization microscopy. Furthermore, the cholesterol clefts were commonly found in upper lung fields (an unusual site for aspiration), were often seen distal to areas of bronchial obstruction, and were generally not associated with clinical evidence of GERD. For these reasons, we hypothesize that the cholesterol clefts seen in our patients may represent breakdown products of cells entrapped by bronchial obstruction rather than stigmata of aspiration. Still, it remains possible that a subset of lungs with BO are primarily injured by aspiration.
Additional data from this study showed that the 1-year survival in the overall population undergoing retransplantation for BOS from 1992 to 2002 at Duke University Medical Center was 75%, which compares favorably with prior registry reports2627 of overall retransplantation 1-year survival of 47%. Also, consistent with previous articles, our report illustrates the poor sensitivity of transbronchial biopsy and chest CT373839 at detecting the presence of histologic BO.
The correlations between whole-lung pathology and clinical, radiologic, or biopsy data have not been studied in detail since the 1980s. In spite of multiple previous autopsy reports describing the pathology of BO, explanted lung allografts have not been given much attention to date in the study of chronic rejection. This study of explanted allografts bypasses the confounding of multiple infections and diseases that lead to death and autopsy, and provides a complete view of lung pathology in BOS patients. However, using this retransplant population, characterized by a good functional status in spite of advanced BOS, may represent a selection bias as well: ours may be a unique population with less rapidly progressive BOS, which would potentially allow for maintenance of a better functional status and lead to eligibility for retransplantation. It is possible that BO in these patients behaves differently compared to patients who become rapidly debilitated and are not eligible for retransplantation. A future study comparing our findings to pathologic appearance of BO in transplanted lung allografts obtained by autopsy could help evaluate differences between these two populations.
This present study provides novel and important insights into the link between BO and clinical BOS in the era of modern pulmonary transplantation and demonstrates the histopathologic variability in lung transplant recipients with severe chronic rejection. Future investigations that focus on the correlation between responses to treatment and various histopathological subsets of BO appear warranted. Furthermore, an expansion of the C part of the current ABC rejection nomenclature, to allow higher grading precision, should be considered, particularly if subtle findings such as airway epithelial injury and inflammation correlate with responsiveness to treatment. Finally, the role of pathologic processes other than BO in the development of clinical BOS needs to be further elucidated.
Footnotes
Abbreviations: BO = bronchiolitis obliterans; BOS = bronchiolitis obliterans syndrome; EFI = epithelial-fibrotic-inflammatory; GERD = gastroesophageal reflux disease; ISHLT = International Society for Heart and Lung Transplantation
This work was performed at Duke University Medical Center.
Received for publication October 12, 2005. Accepted for publication October 17, 2005.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |