1994 84: 3212-3220 Rogers and KW Chan KR Schultz, GJ Green, D Wensley, MA Sargent, JF Magee, JJ Spinelli, S Pritchard, JH Davis, PC transplantation Obstructive lung disease in children after allogeneic bone marrow http://bloodjournal.hematologylibrary.org/site/misc/rights.xhtml#repub_requests Information about reproducing this article in parts or in its entirety may be found online at: http://bloodjournal.hematologylibrary.org/site/misc/rights.xhtml#reprints Information about ordering reprints may be found online at: http://bloodjournal.hematologylibrary.org/site/subscriptions/index.xhtml Information about subscriptions and ASH membership may be found online at: reserved. Copyright 2011 by The American Society of Hematology; all rights 900, Washington DC 20036. weekly by the American Society of Hematology, 2021 L St, NW, Suite Blood (print ISSN 0006-4971, online ISSN 1528-0020), is published For personal use only. by guest on December 8, 2012. bloodjournal.hematologylibrary.orgFrom Obstructive Lung Disease in Children After Allogeneic Bone Marrow Transplantation By Kirk R. Schultz, Gordon J. Green, David Wensley, Michael A. Sargent, J.F. Magee, John J. Spinelli, Sheila Pritchard, Jeffrey H. Davis, Paul C.J. Rogers, Ka Wah Chan, and Gordon L. Phillips Obstructive lung disease (OLD) has been described as a sig- nificant complication after allogeneic bone marrow trans- plantation (BMT). The incidence of OLD in adults appears to be low (-3%). but there is lile data for children. We ana- lyzed 89 consecutive pediatric allogeneic BMTs, 21.5 years post-BMT, performed at British Columbia‘s Children’s Hospi- tal from 1980 to 1992 for evidence of OLD. Diagnosis of OLD was based on clinical findings (nonproductive cough. wheez- ing, and dyspnea with no evidence of infection), pulmonary function tests (RV, < 80% and FEF2S.7S.h c 60% predicted), lung biopsy, and computed tomography scan. Sixty-seven of the 89 children evaluated survived 290 days and were classified as at risk for OLD. Thirteen of 67 (19.4%), devel- oped OLD, 3 of which were transient. The development of OLD was strongly associated with the following high-risk ONE MARROW transplantation (BMT) has become an established life-saving therapy for many children with lethal diseases. These disorders include malignancies, BM dysfunction, inherited metabolic diseases, and immune dis- eases. Complications of BMT include opportunistic infec- tions secondary to immune incompetence, toxicity caused by the cytotoxic agents used to ablate the recipient’s marrow before infusion of the donor marrow, and graft-versus-host disease (GVHD). With improved supportive care, toxicity and infections have become less important as significant side effects. Clinically significant obstructive lung disease (OLD) is one additional complication after BMT. Only rare cases have been examined histologically, and show a pattern character- istic of bronchiolitis obliterans.’ The incidence of OLD in adults appears to be relatively low at -3%?3 Although there are few data regarding this complication in children, the reported mortality in adults who develop OLD secondary to bronchiolitis obliterans after BMT is 65%: Patients develop an obstructive pattern on pulmonary function testing as a late complication (>90 days post-BMT), which usually appears B From the Department of Pediatrics, the Divisions of Hematology/ Oncology/Bone Marrow Transplantation and Intensive Carefiespi- rology, the Department of Radiology, and the Department of Pathol- ogy, University of British Columbia and British Columbia’s Chil- dren’s Hospital, and the Department of Medicine, University of British Columbia, Vancouver General Hospital Bone Marrow Trans- plantation Program, Vancouver, BC, Canada. Submitted December 8, 1993; accepted June 29, 1994. Address reprint requests to Kirk R. Schultz, MD, Department of Pediatrics, Division of Hematology/Oncology/Bone Marrow Trans- plantation, University of British Columbia, British Columbia’s Chil- dren’s Hospital, 4480 Oak St, Vancouver, BC V6H 3V4 Canada. The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked “advehsement” in accordance with 18 U.S.C. section 1734 solely to indicate this fact. 0 1994 by The American Society of Hematology. OOO6-4971/94/8409-0010$3.00/0 3212 groups: chronic graft-versus-host disease (GVHD) (37.1% OLD), increased donor age, acute GVHD, and either mis- matched related or matched unrelated donor transplants. No correlation was found with methotrexate prophylaxis for GVHD, total body irradiation, or cytomegalovirus reactivity in either donor or recipient and the development of OLD. Further analysis of only children with chronic GVHD showed that liver involvement by GVHD before the onset of OLD (57.9%) was the only other significant predictive factor. We observed an overall increased prevalence of OLD in children compared with that previously reported in adults. Further studies are required to confirm whether age is a risk factor for development of OLD after allogeneic BMT. 0 1994 by The American Society of Hematology. within 1.5 years post-BMT.’ Chest roentgenograms are usu- ally normal, or may show slightly hyperinflated lung fields. Associations with OLD after BMT in adults include hypo- gammaglobulinemia and decreased salivary IgA? metho- trexate prophylaxis for GVHD2 that may increase class I1 major histocompatibility complex expression in the lungs: and concurrent viral infections? Others have suggested that esophageal disease with repeated aspiration may con- tribute to small airway di~ease.4.~ The strongest association is that of OLD with chronic GVHD. This leads to the hy- pothesis that OLD secondary to clinical evidence of bron- chiolitis obliterans represents pulmonary involvement from GVHD?8.9 Because little has been previously published describing the complication of OLD in children after allogeneic BMT, we evaluated the occurrence of OLD in the children receiv- ing allogeneic BMTs at British Columbia’s Children’s Hos- pital from 1980 to 1992. MATERIALS AND METHODS Patient population. A retrospective analysis was performed us- ing the British Columbia’s Children’s Hospital BMT program data base. All children receiving allogeneic BMT from 1980 to 1992 aged 0 to 17 were evaluated for the development of OLD. Patients were excluded if there was an acute complication resulting in SW- viva1 less than 90 days or had not had sufficient time to develop OLD with a transplant less than 1.5 years before analysis. These exclusion criteria were based on previous findings in adult patients which demonstrated that OLD is a late complication after BMT associated with chronic GVHD and usually develops within the first 1.5 years post-BMT? Conditioning regimens, transplantation, and supportive care. Conditioning regimens were selected based on the disease and donor type (Table 1). Two children with immunodeficiency syndromes received no conditioning before transplantation. Patients were placed in positive pressure rooms with strict aseptic procedures from days -1 and until an absolute neutrophil count of z 0.5 X lo9 ceUs/L for 2 consecutive days was reached. Broad-spectrum antibiotics were initiated when patients developed fever and after 1986 all patients received acyclovir (1,500 mg/m*/day divided into three doses) if either the donor or recipient were cytomegalovirus (CMV) seroposi- Blood, Vol 84, No 9 (November l), 1994: pp 3212-3220 For personal use only. by guest on December 8, 2012. bloodjournal.hematologylibrary.orgFrom LUNG DISEASE IN CHILDREN AFTER BMT Table 1. Comparison of the Characteristics of Children Unaffected and Affected by OLD Cateaorv Unaffected OLD ($6) 1%) No. of patients Age (mean years 2 SD) Sex ratio M:F Diagnosis ALL AM L AAFA Other malignancies Solid Tumors (NB, Rhabdo) NHL JCMUMDSICMUPCV Nonmalignancies Immunodeficiency (WAS, SCID) Genetic (MLD, Thalassemia, Gaucher) Cell dose (mean 2 SD)* Preparative regimen TB1 (dose Gy) 12.6 + Cy 2 Vcr 12.6 + Ara-C 10 f CP/MNM/D 10 + VP/CP/M 6 + VPIM 4 + cy 7.5 (TLI) + Bu/Cy CY Bu/Cy None 10 + CyNP 54 (80.6) 7.3 2 4.9 1.41 13 (24.1) 11 (20.4) 12 (22.2) 9 (16.7) 4 (7.4) 2 (3.7) 3 (5.6) 12 (22.2) 4 (7.4) a ( 14.8) 54 4.3 2 4.1 16 1 8 2 1 1 2 1 8 12 2 13 (19.4) 7.4 2 4.7 1.6:l 5 (38.9) 2 (15.4) 2 (15.4) 2 (15.4) 1 (7.7) 0 1 (7.7) 2 (15.4) 1 (7.7) 1 (7.7) 13 4.0 2 0.7 4 0 3 1 0 0 1 0 1 3 0 All factors were not significant by the x' test for comparison of sex ratio and diagnosis and the t-test for independent samples for evaluation of age and cell dose. Abbreviations: ALL, acute lymphoblastic leukemia; AML, acute my- eloblastic leukemia; AA, aplastic anemia; FA, Fanconi's anemia; NB, neuroblastoma; Rhabdo, rhabdomyosarcoma; NHL, non-Hodgkin's lymphoma; JCML, juvenile chronic myelogenous leukemia; MDS, my- elodysplastic syndrome; CML, chronic myelogenous leukemia; PCV, polycythemia vera; WAS, Wiskott-Aldrich syndrome; SCID, severe combined immunodeficiency; MLD, metachromatic leukodystrophy; Cy, cyclophosphamide; Vcr, vincristine; VP, VP-16; M, melphalan; VM, VM-26; D, doxorubicin; Bu, busulfan; TLI, total lymphoid irradiation; CP, cisplatin. * Cell dose is expressed in mean number of cells x lO*/kg of the recipient f SD. tive. Prophylactic nystatin mouthwash was given as antifungal pro- phylaxis. Intravenous (IV) Ig 0.5 g/kg was administered weekly after transplantation and monthly after discharge until day 180 if no GVHD was present. Patients with GVHD were treated weekly until day 100 and then monthly until off immunosuppression. GVHD prophylaxis consisted of cyclosporine A 3 mg/kg/d IV divided into two doses and starting on day - 1. Methotrexate prophylaxis when given with cyclosporine A was administered as 15 mg/m' IV on day 1 and 10 mg/m' on days 3, 6, and 11 after transplantation. Patients who received methotrexate without cyclosporine as GVHD prophy- laxis received the same schedule as described above and 10 mg/m2/ dose weekly until day 100 after BMT marrow transplantation. 3213 Criteria for diagnosis of OW. In adults, the diagnosis of ob- structive lung disease has been based on pulmonary function testing, histopathologic evidence of bronchiolitis obliterans, and clinical symptoms. Many of the children who received BMTs were too young (<S years) to perform routine pulmonary function tests. The diagno- sis of OLD in the present study was based on patients meeting all three of the following criteria: (1) No evidence of pulmonary abnormalities before transplantation; (2) clinical features, including cough, wheezing, pneumothorax/mediastinum, or dyspnea; (3) at least one of the following findings, in the absence of infection: (a) Abnormal pulmonary function tests after transplant: - FEV,, 430% predicted; and - FEFz5.,5s <60% predicted; (b) Lung biopsy with evidence of subpleural and/or interstitial fibrosis and bronchiolar abnormalities; (c) computed tomography (CT) findings consistent with bronchiolitis obliterans." Pulmonary function testing. Pulmonary function testing was routinely done 1 year after transplant, if the patient was capable, and was done at other times based on clinical suspicion. Pulmonary function testing was done following the American Thoracic Society guidelines." Forced vital capacity (WC), forced expiratory volume in one second (FEVI), and forced expiratory flow of 25% to 75% of vital capacity (FEF25.75s) and diffusion (DLCO) were performed using Sensormedics Horizon pulmonary function system 5 or Quinton Master lab and Body box system (Jeager, Wurzburg, Ger- many). Mean values for patients were expressed as percent predicted for normals of same age and height based on data from Knudson (flow volume loops) and Polgar (diffusion).".l3 Histopathologic evaluation of lung biopsy samples. Open lung biopsy was performed in five cases. All cases were formalin fixed. Paraffin-embedded sections were stained with hematoxylin-eosin, Gram, Grocott, periodic acid-Schiff (PAS), Masson-trichrome, and Verhoeff van Gieson stains. In addition, samples were cultured for bacterial and viral involvement using standard methods. High-resolution CTevaluation ofpatients. High-resolution com- puted tomography (CT) was performed using l-mm sections at 10- mm intervals through the whole chest. The diagnosis of bronchiolitis obliterans was made if focal or general hypoattenuation, segmental or subsegmental bronchial dilatation, and/or expiratory air trapping in areas of hypoattenuation was present. Initial CT diagnoses were made in a nonblinded fashion. In a separate analysi~,'~ 7 initial and 3 follow-up scans were randomized and compared with 5 scans from normal controls. Three blinded observers correctly identified all 5 controls as normal whereas all scans from the children suspected as having OLD were identified as compatible with bronchiolitis obliterans." Of 30 observer diagnoses, 26 were identified as bronchi- olitis obliterans and 4 were equivocal. Evaluation of potential risk factors. Children affected with OLD and those unaffected were compared on the basis of age, sex, diagno- sis, and nucleated cell dose of marrow infused. Risk factors for the development of OLD, previously described in adults, were evaluated in this study. They included acute and chronic GVHD, donor type, donor age, methotrexate GVHD prophylaxis, total body irradiation (TBI), and (CMV) status. The assessment and grading of GVHD disease was as previously de~cribed.'~.'~ Statistical analysis. Relationships between categorical factors were assessed by the Pearson x' test and comparisons of continuous factors with categorical outcomes were done using the Mann-Whit- ney test. Pulmonary function results were evaluated using the t-test for independent samples. The cumulative incidence of OLD was calculated by the method of Kaplan and Meier. Significance of prog- nostic factors for the development of OLD were assessed on a univar- iate and multivariate basis by the score test from the Cox proportional hazards model. The following covariates were assessed: donor sex and age, BMT type, methotrexate prophylaxis, TBI, and acute and For personal use only. by guest on December 8, 2012. bloodjournal.hematologylibrary.orgFrom 3214 l 0 500 1000 1500 2OOO 2500 3000 3500 Days after BMT Fig 1. Cumulative incidence for the development of OLD. Onset of OLD was measured in a time-dependent manner using a Kaplan- Meier curve and is represented to 3,200 days after BMT. chronic GVHD. Chronic GVHD and liver involvement were used as time-dependent covariates in the Cox model. RESULTS Prevalence of OLD in children after allogeneic BMT at British Columbia's Children's Hospital. Eighty-nine chil- dren received an allogeneic BMT between the years 1980 and 1992 at British Columbia's Children's Hospital. All transplants were performed greater than 1.5 years previous to evaluation by this study. Sixty-seven of the 89 children receiving transplants (75%) survived greater than or equal to 90 days and were classified as at risk of developing OLD (Table 1). Thirteen of 67 (19.4%) developed OLD. The cu- mulative incidence of OLD in children transplanted and sur- viving greater than 90 days after transplant was 26% (95% confidence interval [CI], 14% to 44%) by the method of Kaplan and Meier (Fig 1). SCHULTZ ET AL Characteristics of children with OLD compared with unaf- fected BMT patients. Comparison of the children with OLD to those who were unaffected demonstrated no differ- ence in age distribution (Table 1). Moreover, we saw no difference in the sex ratio, the pre-BMT diagnosis, the cell dose of nucleated marrow cells infused at the time of BMT, and the preparative regimens used between the two groups. Specific characteristics of children with OLD. Diagnosis of OLD was made at a median time of 7.5 (range 3.0 to 55.3) months after BMT (Table 2). Three patients had resolution of OLD and currently have normal pulmonary function tests. All patients presented with cough, 8 of 13 with dyspnea (62%), and 4 of 13 (31%) with wheezing (Table 2). Eight (62%) had a history of bronchitis, 5 (39%) with sinusitis, and 3 (21%) with a pneumothorax or pneumomediastinum. Pulmonary function tests (PFTs) were consistent with severe obstructive disease in 10 patients who were old enough (25 years old) to be tested at time of presentation of OLD. An additional two patients were evaluated by PFT after diagno- sis of OLD at a later age (Table 2). All patients, except for one, had confirmation of the diagnosis of OLD by pulmonary function testing at some time. High-resolution CT confirmed the diagnosis of OLD secondary to bronchiolitis obliterans in 7 of 13 patients." Specific features evaluated were the following: (1) focal or general parenchymal hypoattenuation (present in 7/7), (2) segmental or subsegmental bronchial dilatation (present in 5/7), and/or (3) expiratory air trapping in areas of hypoattenuation (present in 3/3). A representative high-resolution CT (Fig 2) shows diffuse parenchymal hypo- attentuation and bronchial dilatation. Lung biopsies were performed on five patients and showed subpleural and interstitial fibrosis and bronchiolar abnormal- ities in all cases (Fig 3, A and B). All 5 biopsy specimens had histologic findings of variable subpleural and interstitial fibrosis, with 4 having additional interstitial collections of lymphocytes. All had bronchiolar abnormalities including findings of lymphocytic infiltration of the epithelium (4), Table 2. Diagnosis of OLD in Children After Marrow Transplantation Onset of OLD Duration of OLD Resolution Clinical Pulmonary Fibrosis UPN (mo) (mo) of OLD Symptoms Function on Biopsy CT Findings 9 3.0 6 No (died) c. P Abnormal ND ND 10 9.7 87 Yes C. D Abnormal ND ND 55 55.3 9 No C Abnormal ND ND 25 5.4 80 No C, Cl, D, P Abnormal Yes BO 61 7.0 49 No C, Cl, D, W ND Yes ND 62 3.6 21 Yes C Abnormal Yes ND 74 31.8 12 No c, Cl Abnormal* ND BO 75 7.5 36 No C, Cl, D, W Abnormal Yes BO 99 6.8 20 No c, Cl, W Abnormal ND BO 103 10.9 13 No (died) C. W. D Abnormal ND BO 107 3.9 12 Yes C Abnormal ND ND 108 12.0 8 No C, Cl, D, W Abnormal* ND BO 114 5.5 9 No (died) C, Cl, D, W, P Abnormal Yes BO Abbreviations: C, cough; P, pneumothorax and/or pneumomediastinum; D, dyspnea; Cl, chest infection (clinical diagnosis of recurrent bronchi- * Pulmonary function testing was not performed at the time of diagnosis of OLD because of age, but was abnormal at a later time when the tis); W, wheezing; ND, not done; BO, bronchiolitis obliterans. child was older. For personal use only. by guest on December 8, 2012. bloodjournal.hematologylibrary.orgFrom LUNG DISEASE IN CHILDREN AFTER BMT 321 5 Fig 2. High-resolution CT of the chest in a child with OLD. The CT (patient no. 25) shows diffuse parenchymal hypoattenuation and bronchial dilatation. The wedge of higher attenuation anterior to the lateral aspect of the right oblique fissure represents normal lung. regenerative changes of the epithelium (3), subepithelial lymphocytes (l), and luminal fibrosis (I) (Table 3). Type 2 pneumocyte hyperplasia was also seen in the alveoli of 4 cases. No organisms could be identified in the histologic sections examined and in the 4 cases submitted for bacterial and viral culture, no growth of organism was detected. Pulmonary function tests in some children were difficult to obtain because many of the patients were too young (<S years old) to have PFTs done either before or after BMT (Table 4). Of those tested, children appeared to have either transient or permanent obstructive patterns. Three patients developed transient severe obstructive changes (Table 4) in that at the onset of OLD the FEV,, the FEF2S.7SCk and the DLCO were all significantly decreased compared with before transplantation (P < .OS). Each completely resolved after immunosuppressive therapy (Table 4). The only patient to show a response to bronchodilators during evaluation by PFT was one patient with transient OLD. In the patients with permanent changes, PFTs showed a significantly lower FEV, at the time of diagnosis of OLD compared with pre-BMT values (P = .001) (Table 4). FEF2S.7sth was even more se- verely affected (Table 4) in the OLD group at the time of diagnosis compared with those pre-BMT (P = < .001). The DLCO was similarly affected. All three (FEV,, FEF?S.7s,k, and DLCO) continued to be significantly lower at later evalu- ations. Because the obstructive changes were severe, the FVC was also decreased at the time of diagnosis of OLD (Table 4). Restriction of school and physical activities (Karnofsky S 70%) at the time of diagnosis was apparent in 7 of IO with permanent OLD and I of 3 with transient OLD (Fig 4). After treatment with immunosuppressive therapy, Karnofsky scoring of the children with permanent OLD showed an overall improvement in the activity. The three children in the transient group all returned to normal pulmonary function and improved activity performance. One of these three chil- dren subsequently developed a drop in performance as her underlying disease (metachromatic leukodystrophy) pro- gressed, although the marrow graft is functional. Three pa- tients with OLD died; two of pulmonary failure and one of relapse. Associntions between the developntertt of OLD and poten- tial risk factor.7. Donor type strongly correlated with the development of OLD. Patients receiving mismatched related donor (40%) or matched unrelated donor (44.4%) transplants developed OLD compared to 13.2% with matched related donor transplants (P = .OM). Because increased donor age is associated with matched unrelated donors compared to matched sibling donors in children, we evaluated donor age as an independent variable and saw a significant increase in median donor age in the children with OLD compared to unaffected children (P = .036). Further evaluation of the significance of donor age by donor type (HLA typing) showed no significant influence of donor age. On the other hand, the increased median age (12 years) in the OLD group of HLA-identical donors compared with the unaffected chil- dren (10 years) approached significance (P = .0S7). Evalua- tion of other potential risk factors showed that there was no clear association of methotrexate prophylaxis, TBI, or CMV reactivity pre-BMT with the development of OLD after BMT (Table 5). Eleven of 36 (30.6%) patients with acute GVHD had OLD compared to 2 of 31 (6.5%) without. This differ- ence was statistically significant (P = .009). Chronic GVHD (Table 5) was also significantly associated with OLD with 37.1% ( 13/35) developing OLD compared to 0/32 of patients without chronic GVHD (P < .OOl). Because all patients with OLD also had chronic GVHD, further multivariate analysis could not be performed on the entire population. Multivariate analysis was performed for further risk factors in the population of patients with chronic GVHD to evaluate for other contributing factors. Patients with evidence of liver involvement as a part of chronic GVHD had a strong association with OLD, with 57.9% of chronic GVHD patients with liver involvement developing OLD compared to 12.5% in the unaffected patients (P < .001). Evaluation of other potential risk factors in children who developed chronic GVHD showed that there was no significant risk associated with donor sex, donor type, donor age. acute GVHD (=grade 2), methotrexate prophylaxis, cyclosporine prophylaxis, or TB1 after adjustment for liver involvement. Thus, chronic GVHD, particularly of the liver, was the major risk factor associated with OLD in this popula- tion. Comparison ?f children with perntottertt versus trnnsient OLD. Comparison of patients who developed permanent rather than transient OLD showed no significant difference in patient age, onset of acute or chronic GVHD, or in the onset of OLD (Table 6). Patients who developed OLD were treated with many combinations of immunosuppressive ther- apy and Ig preparations (Table 6). Patients with permanent For personal use only. by guest on December 8, 2012. bloodjournal.hematologylibrary.orgFrom 3216 SCHULTZ ET AL . Fig 3. For personal use only. by guest on December 8, 2012. bloodjournal.hematologylibrary.orgFrom LUNG DISEASE IN CHILDREN AFTER BMT 3217 Permanent Transient a higher incidence of OLD than is representative of the adult population. If we eliminate the unrelated donor transplants from our population in this comparison (not included in the Seattle study), 9 of 20 children (5 17 years) with chronic GVHD (45%) in our study developed OLD compared to 33 of 141 (23.4%) adults with chronic GVHD (>l7 years) in the Seattle study. Assuming that these two populations are equivalent except for age, we can see an increase of OLD in children that approaches a significant difference (P = .07). In addition, review of the Johns Hopkins group study evaluating OLD after marrow transplantation shows that all patients receiving allogeneic transplants affected with OLD were under 30 years old (7/291), whereas none of 95 patients older than 30 years were affected with OLD.4 Thus, young age may be a significant risk factor for the development of There are a number of reasons why children might be at Time of evaluation higher risk for the development of OLD. Firstly, children Fig 4. Karnofsky rating of with OLD. status of may be at an increased risk of primary viral infection because each patient was made using the Karnofsky scale at the time of onset they have not had the previous viral exposure equivalent of OLD and with the most recent visit. Patients are divided into those to that in adults. However. we observed no concomitant Onset Most recent Onset Most recent OLD after allogeneic BMT. who have resolution of OLD (transient) and those who have perma- nent changes (permanent). The dotted line represents the level below which patients are restricted in the abili to attend school and main- tain normal physical activities (Karnofsky s 70%). The patient in the transient group with a decrease in activity after onset of OLD cur- rently has no pulmonary symptoms, but has poor performance be- cause of progression of metachromatic leukodystrophy despite suc- cessful marrow engraftment. OLD did receive 3 or more immunosuppressive agents in 6 of 10 patients compared with 1 of 3 in the transient group. However, no significant difference was demonstrable in the therapy used to suggest that the children who were classified with transient OLD were transient because of improved ther- apy. DISCUSSION This study shows a relatively high prevalence of OLD in the pediatric population receiving allogeneic BMTs at British Columbia’s Children’s Hospital. We hypothesize that the incidence of OLD observed in this study which is increased compared with previously published studies in adults, may be partially related to patient age. To evaluate this hypothesis we compared our results with the adult study done in Seattle that used OLD diagnostic criteria similar to ours.’ Because of a strong association of OLD with chronic GVHD, a con- servative comparison between adults and children can be made looking at the prevalence of OLD in all patients who have chronic GVHD. This comparison corrects for the higher prevalence of chronic GVHD in adults, which may lead to infections associated with the onset of OLD. Host presenta- tion of self or viral antigens has been suggested previously” as important in the induction of GVHD and there may be tissue to tissue variation of self-antigen expression.” This leads to the hypothesis that certain tissues in children, such as the lung, may be more susceptible to involvement by GVHD. It is possible that there is a difference in antigen presentation of self antigen in the lungs of children compared with adults by dendritic cells,’’ but this has not been investi- gated. Gastro-esophageal reflux may be more prominent in children, and has been associated with an increased incidence of OLD in children.’ However, the most plausible explana- tion relates to the size of a child’s airways compared with that of an adult. Because the pediatric airway is already smaller in size, once an inflammatory response secondary to GVHD has ensued the resulting airway obstruction is much more clinically significant in a child compared with that seen in an adult. This may lead to a higher incidence of OLD. Whatever the reason, further investigation is definitely war- ranted to address these questions. OLD has been previously associated with chronic GVHD. We observed a very strong correlation between OLD and chronic GVHD. One interesting observation was a very strong association of liver enzyme elevation interpreted as chronic liver GVHD with OLD in our population. It could be argued that the alanine transferase (ALT) elevations are secondary to pulmonary damage, but we would have ex- pected an even greater elevation of the aspartate aminotrans- ferase (AST). This leads us to conclude that we were observ- ing ALT elevations secondary to liver involvement, not lung. < Fig 3. Pathology of a lung biopsy specimens in children with OLD. (A) An open-lung biopsy (unique patient number [UP] 114) showing spindled fibroblasts almost totally plugging airspace (hematoxylin-eosin, original magnification x 400). fBJ An open-lung biopsy (UPN 62) showing a bronchiole lumen containing inflammatory debris, bronchiolar epithelium exhibiting reactive change, and a predominantly lympho- cytic infiltrate extending into surrounding interstitium (hematoxylin and eosin, original magnification x 100). For personal use only. by guest on December 8, 2012. bloodjournal.hematologylibrary.orgFrom 3218 SCHUCTZ ET AL Table 3. Summary of Histopathologic Findings From Five Open-Lung Biopsy Specimens UPN Pleura Interstitium 25 Fibrosis Fibrosis Lymphocytes 61 Fibrosis Fibrosis Lymphocytes 62 Fibrosis Fibrosis 75 Fibrosis 114 Fibrosis Fibrosis Lymphocytes Fibrosis Lymphocytes Bronchioles Epithelial lymphocytes Epithelial lymphocytes Epithelial regeneration Epithelial lymphocytes Epithelial regeneration Subepithelial lymphocytes Epithelial lymphocytes Epithelial regeneration Epithelial loss Luminal fibrosis Alveoli Type 2 pneurnocyte hyperplasia Macrophages Macrophages Type 2 pneurnocyte hyperplasia Macrophages, PMN Type 2 pneumocyte hyperplasia Type 2 pneurnocyte hyperplasia Macrophages Moreover, we observed that liver enzyme elevations were present either at the time of diagnosis of OLD or preceded the diagnosis in 11 of 13 patients. These abnormalities cor- rected in most patients after increased immunosuppression was initiated for treatment of chronic GVHD. Whether this association of liver GVHD with OLD is unique for children is uncertain, but the association is strong enough in our population that we believe that it may he a marker for the development of OLD. We observed no other associations with other factors pre- viously identified in adults. Previously, methotrexate GVHD prophylaxis has been identified as a risk factor for develop- ment of OLD in adults.' We observed no similar association. The number of children not receiving methotrexate were few in number decreasing the statistical power of our analysis, possibly excluding our ability to identify this as a risk factor. Mechanisms involved in the development of airspace fi- brosis leading to OLD in humans are poorly understood. One group evaluated the cell populations isolated by bron- choalveolar lavage after development of bronchiolitis ohlit- Table 4. Pulmonary Function Tests of Patients Affected With OLD FEV, FEF25-,5% FVC DLCO UPN Pre-BMT Onset Current Pre-BMT Onset Current Pre-BMT Onset Current Pre-BMT Onset Current Bronchodilators Response to Permanent 9 25 55 61 74 75 99 103 108 114 Mean 95% cl Transient 10 62 107 Mean 95% Cl Unaffected 87 ND ND ND ND ND 111 131 ND 96 106 (87, 124) 74 49 40 85 78 ND ND ND 49 38 37 42 57 29 36 ND 27 41 43 51 46 (36, 67) (35, 57) - 76 58 ND 24 14 ND 49 52 ND ND ND ND ND 91 ND 18 9 123 17 17 118 10 13 ND ND 23 112 8 9 107 26 29 (86, 128) (11, 41) (9, 49) - 90 81 ND 53 67 ND 96 87 ND ND ND ND ND 43 ND 75 77 102 60 82 129 72 52 ND ND 26 90 63 74 103 71 64 (85, 121) (61, 81) (49, 79) - 87 65 90 68 50 107 102 65 86 88 46 95 76 31 105 92 53 90 133 79 126 130 54 130 130 86 103 (74, 132) (44, 82) (81, 125) (54, 128) (30, 59) (98. 130) (85. 131) (49, 87) (83, 103) 103 63 103 91 45 114 108 68 93 ND ND ND ND ND ND 95 98 ND 92 95 (92,98) ND 96 ND 96 ND - ND 91 77 82 ND ND ND ND ND 73 81 102 69 41 ND ND 19 51 62 73 (34, 90) (54, 90) ND 112 54 93 54 61 54 89 (60, 118) No No No ND No No No No ND No No No Yes controls Mean 99 - 93 95 - 112 97 - 90 89 - 92 95%CcI (91, 107) - 186, 100) (79, 111) - (96, 130) (89, 105) - (83, 97) (93, 951 - (84, 100) FEV,, FEFZ5.759/., FVC, and DLCO are expressed as percent of that expected for age and height. The values are shown as baseline evaluations before BMT (pre-BMT), at the time of diagnosis of OLD (onset), and as the most recent value (current). Unaffected patients are shown as the most recent values (current). UPN 9 died soon after diagnosis of OLD and therefore no current PFTs are available. Abbreviation: ND, not done, because of young age (<5 years). For personal use only. by guest on December 8, 2012. bloodjournal.hematologylibrary.orgFrom LUNG DISEASE IN CHILDREN AFTER BMT 3219 Table 5. Potential Risk Factors for the Development of OLD Category Unaffected With OLD No. No. Patients Patients (%l PValue" Donor type HLA identical 1 Ag mismatch Matched unrelated Evaluation for donor age, median years (range) HLA identical 1 Ag mismatch Matched unrelated All donors Methotrexate prophylaxis Yes No TB1 Yes No CMV DonorIRecipient +I+ +/- -l+ GVHD -I- Acute GVHD (zgrade 2) Yes No Chronic GVHD Yes No Liver involvement in patients with chronic GVHD Yes No 46 7 (13.2) 3 2 (40.0) 5 4 (44.4) 10 (1, 27) 12 (6, 23) 23 (5, 35) 4 (2, 6) 42 (31, 47) 39 (34, 43) 11 (1, 47) 16 (2, 43) 51 12 (19.0) 3 1 (25.0) 32 10 (23.8) 22 3 (12.0) 16 3 ( 15.8) 5 5 (50.0) 5 0 (0) 24 4 ( 14.3) 25 11 (30.6) 29 2 (6.5) 22 13 (37.1) 32 0 (0) 8 11 (57.9) 14 2 (12.5) .033 ,057 .l4 .88 ,036 .82 .2 1 .55t ,009 <.001 <.001 Abbreviations: f, recipients or donors who had serology that was positive for the presence of CMV reactive IgG; -, recipients or donors who had serology that was negative for the presence of CMV reactive IgG. * Statistical analysis was performed using the score test from the Cox proportional hazards model. t Analysis of all groups with reactivity with CMV in either donor and/or recipient (+l+, +I-, -l+) compared to those with both donor and recipient CMV nonreactive (-1-1. erans in both BMT and non-BMT patients." They noted that most BMT patients had a lymphocyte predominance in the lavage samples compared with a preponderance of neutro- phils seen in the non-BMT patients. This suggests a different pathologic process of airspace fibrosis associated with GVHD in BMT patients. Bronchial associated lymphocytic tissue (BALT) and epithelium can express class I1 major histocompatibility complex antigens and may be the target for T-cell-mediated injury leading to fibrosis. Histology shows mural fibrosis with mononuclear inflammatory cells with some areas completely obliterated. There appears to be a moderate interstitial infiltration of lymphocytes and plasma cells adjacent to the involved bronchioles and normal paren- chyma in other areas. In addition, dendritic cells have been important in bronchiolitis obliterans in patients receiving heart-lung transplants." Why it develops in some patients with chronic GVHD and not in others is poorly understood. Diagnosis of OLD in children is particularly difficult. Many children are too young to perform pulmonary function tests in a manner acceptable for an accurate evaluation. Lung biopsy is invasive and we have noted excessive morbidity in children receiving biopsies. Less invasive measures are required. High-resolution CT has proven to be highly sensi- tive in this seriesI4 and has allowed us to diagnose some children not evaluable by other means. Noninvasive diagnos- tic methods such as this need to be pursued in a more aggres- sive manner to diagnose OLD in children early and accu- rately. Various therapies were used in treatment of the children with OLD. Because chronic GVHD is strongly associated with OLD, immunosuppressive therapy was used in all pa- tients resulting in an improvement in performance in 10 of 13 patients. Patients with more severe disease (permanent) received more intensive immune suppression with no obvi- ous advantage of one immunosuppressive agent or approach over another. Most probably this represents the difference in severity of OLD rather than in any efficacy of a particular therapeutic approach. Thus, any immunosuppressive therapy Table 6. Comparison of Children With Permanent Versus Transient OLD Category Permanent Transient P Value No. of patients Age (yrs) Onset of acute GVHD (days after BMT) Onset of chronic GVHD (days after BMT) Onset of OLD (days after BMT) Treatment of OLD Aza Budesonide CsNPred Predl CsNPredIAzaflhal CsNPredTThalI Budesonide CsNPredI Budesonide CsNPredlThalI AzaIPredl cyclophosphamide Pentaglobulin Cyclophosphamide 10 4.8*(0.5, 17.0) 13(5, 57) 90(90, 108) 217(90, 1,659) 2 1 1 1 0 3 8.5(5.7, 13.7) .69 13.5(12, 15) .91 108(90, 11 1) .22 117(108, 291) .29 0 1 Abbreviations: Pred, Prednisone; Aza, Azathioprine; CsA, cyclo- sporine; Thal, Thalidomide; Budesonide, aerosolized steroid; Pentag- lobulin; IV Ig with increased IgA content. * Median (range). For personal use only. by guest on December 8, 2012. bloodjournal.hematologylibrary.orgFrom 3220 SCHULTZ ET AL appears to result in some therapeutic improvement, although significant morbidity may still be present. This study suggests that there appears to be a high preva- lence of OLD in children compared to adults and is strongly associated with chronic GVHD, particularly of the liver. Prospective studies evaluating the onset of OLD and possible causative factors are necessary to improve understanding of this significant complication and possibly allow for early intervention. ACKNOWLEDGMENT We thank Drs Ronald Anderson, John Wu, Christopher Fryer, and Yigal Kaikov for the excellent clinical care provided to these pa- tients. Additional thanks to the nursing staff of 3B, the clinical pharmacists and the dietitians at British Columbia's Children's Hos- pital for outstanding supportive care. Lastly, we thank Nita Takeuchi for clinical coordination and Dr Michael Seear and Julia A. Schultz for critical review of this manuscript. REFERENCES 1. Roca 3, Granena A, Rodriguez Rosin R, Alverez P, Agusti- Vidal A, Rozman C: Fatal airways disease in an adult with chronic graft versus host disease. Thorax 37:77, 1982 2. Clark JG, Crawford SW, Madtes DK, Sullivan KM: Obstruc- tive lung disease after allogeneic marrow transplantation: Clinical presentation and course. Ann Intern Med 11 1:368, 1989 3. Ralph DD, Springmeyer SC, Sullivan KM, Hackman RC, Storb R, Thomas ED: Rapidly progressive air-flow obstruction in marrow transplant recipients. Am Rev Respir Dis 129541,1984 4. Holland HK, Wingard JR, Beschorner WE, Sara1 R, Santo GW: Bronchiolitis obliterans in bone marrow transplantation and its relationship to chronic graft-vs-host disease and low serum IgG. Blood 72:621, 1988 5. Epler GR: Bronchiolitis obliterans and airways obstruction as- sociated with graft-versus-host disease. Clin Chest Med 9:551, 1988 6. Chien J, Chan CK, Chamberlain D, Patterson B, Fyles G, Minden M, Meharchand J, Messner H: Cytomegalovirus pneumonia in allogeneic bone marrow transplantation: An immunopathologic process? Chest 948:1034, 1990 7. Schultz KR, Israel DM, Wensley D, Abella E, Green GJ, Has- sal1 E, Ravindranath Y, Davis JH: Association of gdstro-esophageal reflux with obstructive lung disease (OLD) in children after alloge- neic bone marrow transplantation. Med Pediatr Oncol 2 I :S6 I. 1993 8. Clark JG, Schwartz DA, Flournoy N, Sullivan KM, CrdWfOrd SW, Thomas ED: Risk factors for airflow obstruction in recipients of bone marrow transplants. Ann Intern Med 107:648, 1987 9. Wyatt SE, Nunn P, Hows JM, Jal Yin, Hayes MC, Catovsky D. Gordon-Smith EC, Hughes JMB, Goldman JM Galton DAG: Airway obstruction associated with graft-versus-host disease after bone marrow transplantation. Thorax 39:887, 1984 IO. Graham J, Muller NL, Miller RR, Shepherd JD: Intrathoracic complications following allogeneic bone marrow transplantation. Radiology 18 I : 153, I99 I I I. American Thoracic Society: ATS statement-Snowbird work- shop on standardization of spirometry. Am Rev Respir Dis 1 19:83 I. 1979 12. Polgar G, Promadhat V: Pulmonary Function Testing in Chil- dren: Techniques and Standards. Philadelphia, PA, Saunders, 1971 13. Knudson RJ, Lebowitz MD, Holberg CJ, Burrows B: Changes in the normal maximal expiratory flow-volume curve with growth and aging. Am Rev Respir Dis 127:725, 1983 14. Sargent MA, Cairns RA, Murdoch MJ, Wensley D, Schultz KR: High resolution CT of obstructive lung disease following pediat- ric bone marrow transplantation. Society for Pediatric Radiology, Colorado Springs, CO April 29, 1994 15. Thomas ED, Storb R, Clift RA: Bone marrow transplantation. N Engl J Med 292332 and 895, 1975 16. Sullivan KM: Acute and chronic graft versus host disease in man. Int J Cell Cloning 4:42, 1986 (Suppl 1) 17. O'Reilly RJ: Commentary: Antigen-presenting cells: A key to tolerance? Marrow Transplant Rev 2:60, 1993 18. Greim P, Wallny H-J, Falk K, Rotzscke 0, Arnold B, Schon- rich G, Hammerling G, Rammensee H-G: Uneven tissue distribution of minor histocompatibility proteins versus peptides is caused by MHC expression. Cell 65633, 1991 19. Yousem SA, Ray L, Paradis IL, Dauber JA, Griftith BP: Potential role of dendritic cells in bronchiolitis obliterans in heart- lung transplantation. Ann Thorac Surg 49:424, I990 20. St John RC, Gadek JE, Tutscka PJ, Kapoor N. Dorinsky PM: Analysis of airflow obstruction by bronchoalveolar lavage following bone marrow transplantation. Chest 98:600, 1990 For personal use only. by guest on December 8, 2012. bloodjournal.hematologylibrary.orgFrom . Sargent, JF Magee, JJ Spinelli, S Pritchard, JH Davis, PC transplantation Obstructive lung disease in children after allogeneic bone marrow http://bloodjournal.hematologylibrary.org/site/misc/rights.xhtml#repub_requests Information. 2012. bloodjournal.hematologylibrary.orgFrom Obstructive Lung Disease in Children After Allogeneic Bone Marrow Transplantation By Kirk R. Schultz, Gordon