Bone Marrow Transplantation (2008) 41, 971–975 & 2008 Nature Publishing Group All rights reserved 0268-3369/08 $30.00
Bronchiolitis obliterans following pediatric allogeneic hematopoietic stem cell transplantation CN Duncan, MR Buonanno, EV Barry, K Myers, D Peritz and L Lehmann Department of Pediatrics, Dana-Farber Cancer Institute, Jimmy Fund Clinic, Boston, MA, USA
Bronchiolitis obliterans (BrOb) is a well-recognized complication of allogeneic hematopoietic stem cell transplantation (HSCT). It is associated with substantial morbidity and mortality in adult patients. However, the incidence and morbidity of this complication have not been well described in the pediatric population. We report our experience of BrOb in 216 pediatric allogeneic HSCT patients between 1 January 2001 and 31 December 2005. In total 18 of 216 patients developed BrOb during this time. The diagnosis of BrOb was based on pulmonary function abnormalities, radiographic ﬁndings or lung biopsy. In total 14 of 18 patients with BrOb received stem cells from unrelated donors. In total 17 of 18 patients received bone marrow as a stem cell source, and 1 received peripheral blood stem cells. All pediatric patients in this report had a known risk factor for BrOb, most commonly chronic GVHD (l8 of 18 patients). Additionally, 7 of 18 patients had either toxic lung injury or virally mediated pulmonary disease before the diagnosis of BrOb. With a median of 45.1 months of follow-up, the outcomes were 5 of 18 patients died of lung disease, 2 died of other causes, 3 had progressive lung disease, 6 achieved partial resolution of disease and 2 had stable disease. BrOb, while uncommon, is associated with considerable morbidity and mortality in pediatric HSCT. Bone Marrow Transplantation (2008) 41, 971–975; doi:10.1038/bmt.2008.19; published online 25 February 2008 Keywords: stem cell transplant; bronchiolitis obliterans; pediatric
Introduction Allogeneic hematopoietic stem cell transplant (HSCT) can provide a curative option for many pediatric patients with both malignant and nonmalignant conditions. As survival rates after HSCT improve, it is increasingly important to
diagnose and effectively treat associated long-term complications. Bronchiolitis obliterans (BrOb) is a noninfectious, late-onset pulmonary complication of HSCT.1,2 While there are rare reports of BrOb occurring after autologous HSCT, it almost exclusively occurs following allogeneic transplantation.3,4 The reported incidence of BrOb in adult allogeneic HSCT patients ranges from 1 to 10%.1,2,5–7 Few studies address the incidence of BrOb following pediatric HSCT and estimates range from 2 to 6%.8,9 BrOb typically presents with symptoms of cough, dyspnea with or without exertion and/or wheezing. Patients may also be asymptomatic, with lung dysfunction detected only on routine post-HSCT surveillance. Pathologically, BrOb results from obstruction and/or obliteration of the small airways. It is characterized by luminal occlusion of the terminal and respiratory bronchioles by inﬂammatory and ﬁbrous tissue.10 In severe cases, the airway develops circumferential scarring with complete obliteration of the lumen.10,11 BrOb is thought to be an immune-mediated process, and in pediatric patients it is most commonly seen following lung transplantation and HSCT in association with chronic GVHD.10 In both settings, immunosuppression is the main approach to treatment.6 Adult HSCT patients with BrOb are at signiﬁcant risk of mortality, with rates reported between 13 and 100%.1,2,6,12 While few published studies exist in pediatric HSCT patients, mortality rates between 11 and 67% have been reported in small series.8,13 There are multiple proposed risk factors for post-HSCT BrOb. These include acute and chronic GVHD, mismatched HLA donor, older donor, abnormal pre-HSCT pulmonary function, GVHD prophylaxis with MTX, myeloablative conditioning regimens, conditioning with BU, prior history of interstitial pneumonitis, female donor to male recipient, long duration of time from diagnosis to transplant, the use of peripheral blood stem cells as a donor source, respiratory viral infection within the ﬁrst 100 days following transplant and low immunoglobulin levels following HSCT.1,2,5–7,12,14,15
Materials and methods Correspondence: Dr CN Duncan, Department of Pediatrics, DanaFarber Cancer Institute, 44 Binney Street, Jimmy Fund Clinic, Boston, MA 02115, USA. E-mail: [email protected]
Received 14 May 2007; revised 31 December 2007; accepted 7 January 2008; published online 25 February 2008
Patients We performed a retrospective review of 216 consecutive pediatric allogeneic transplant patients at the Dana-Farber Cancer Institute (DFCI) and Children’s Hospital, Boston.
Bronchiolitis obliterans following pediatric HSCT CN Duncan et al
These patients underwent 220 transplants at our institutions between 1 January 2001 and 31 December 2005. We identiﬁed all patients diagnosed with BrOb. In addition to baseline demographics, we collected information regarding underlying disease, transplant characteristics, potential risk factors for BrOb, diagnosis and treatment of BrOb, and outcome. For patients having more than one transplant during the study period, age at transplant, donor source, stem cell source and degree of HLA matching were recorded for each transplant. All other data were recorded once per patient. Data for each patient were recorded from prior to transplant through death or most recent clinic visit. To meet criteria for a diagnosis of BrOb, patients were required to have either a pathologic diagnosis following lung biopsy or an irreversible obstructive pattern on pulmonary function tests (PFTs) on more than two occasions in association with characteristic ﬁndings on computed tomography (CT) scans or chest radiograph (CXR). Patients without a pathologic diagnosis were classiﬁed as bronchiolitis obliterans syndrome (BOS). Hyperinﬂation and ﬂattening of the diaphragm on CXR were considered characteristic of BrOb. Characteristic CT ﬁndings included air trapping, bronchiectasis, mosaic perfusion, bronchial wall thickening and bronchial dilatation. Patients with only radiographic ﬁndings did not meet criteria for a diagnosis of BrOb unless the diagnosis was subsequently conﬁrmed pathologically on postmortem examination. PFTs are routinely performed on all patients who are developmentally able to complete the studies at our institution prior to transplant and at 1 year following HSCT. Patients with pulmonary ﬁndings undergo testing at the onset of symptoms. We reviewed the results of all available PFTs. Speciﬁc pulmonary function values reviewed included forced expiratory volume in 1 s (FEV1) and forced vital capacity (FVC). A decline in FEV1 greater than or equal to 20% of baseline or FEV1/FVC o70% of the predicted value was considered an obstructive pattern. Demographic and baseline data collected included age, sex and pretransplant diagnosis. The following information regarding the donor and transplant was collected: relationship of recipient to donor, degree of donor HLA-matching and source of stem cells. Additional data collected on those with BrOb included the length of time following HSCT to diagnosis of BrOb, method of diagnosis (radiographic, pathologic and/or with PFT) and treatment of BrOb. We examined potential risk factors for BrOb in all patients including the presence of acute or chronic GVHD, the diagnosis of a viral pulmonary process within the ﬁrst 100 days following HSCT and a severe drug or transfusion reaction involving pulmonary symptoms within the ﬁrst 100 days following transplant. Outcome was recorded for all patients. The clinical outcome for patients with BrOb was classiﬁed as resolved, partially improved, stable, progressive disease, death due to pulmonary disease or death from a nonpulmonary cause. This classiﬁcation was based on clinical status as described in the attending physician’s most recent note, radiographic ﬁndings and PFT results. Patients without BrOb were classiﬁed as alive or dead. The cause of death for each non-BrOb patient was classiﬁed as relapse mortality, multisystem organ failure, Bone Marrow Transplantation
infectious, respiratory failure and GVHD. Information was documented from the pretransplant evaluation until the last known follow-up visit for those still living and until last date of available information for those patients who died. The Institutional Review Board at the DFCI approved this study.
Statistical methods Differences in characteristics between groups (those with BrOb and those without BrOb) were compared using Fisher’s exact tests for categorical data. All tests conducted were two-sided at the signiﬁcance level of 0.05. There were no corrections for multiple comparisons.
Results In total 216 patients underwent a total of 220 allogeneic HSCT transplants during the study period. A total of 10 patients without BrOb and 2 with BrOb had one or more HSCTs prior to the study period. One patient with BrOb and three patients without BrOb had two transplants during the study period. All patients who developed BrOb did so after their ﬁnal transplant. All living patients diagnosed with BrOb were seen at least once in clinic in the year preceding completion of this manuscript. Demographic data and baseline transplant characteristics of all patients are presented in Table 1.
Transplant characteristics Donor type, degree of donor matching and stem cell source are presented in Table 1. A total of 78.9% of patients with BrOb had unrelated donor HSCT as compared with 52.2% of patients without BrOb. A total of 73.7 and 84.1% of patients with BrOb and without BrOb, respectively, received stem cells from six of six HLA-matched donors. The remaining patients received stem cells from donors matched at less than six of six HLA loci. All patients had high-resolution typing of class II antigens and lowresolution molecular typing of class I antigens. Bronchiolitis obliterans diagnosis A total of 18 (8.3%). patients were diagnosed with BrOb/ BOS. A pathologic diagnosis of BrOb was made in 12 patients. Of these, 11 diagnostic specimens were obtained prior to treatment, and one remaining specimen was from autopsy. Five patients were diagnosed by PFT results and chest CT ﬁndings. The remaining patient was diagnosed from PFT and an abnormal CXR. The mean time from HSCT to BrOb/BOS diagnosis was 385.8 days (median 328 days; range, 48–927 days). Table 2 summarizes the diagnostic information, treatment and outcome of the patients with BrOb/BOS. Of those with BrOb/BOS, pretransplant PFTs were available for 10 patients. Five patients were developmentally unable to complete PFTs prior to HSCT and the values from three patients are missing though they were noted to be normal in the transplant admission note. The mean pretransplant FEV1 for this group was 99.2% predicted (median 89.5%) and the mean FEV1/FVC was
Bronchiolitis obliterans following pediatric HSCT CN Duncan et al
973 Table 1
Patient characteristics Patients with BrOb/BOS
Patients without BrOb/BOS
18 (8.3%) 19
198 (91.7%) 201
11(61.1%):7 (38.9%) 13 (72.2%):5 (27.8%) 9.6 years (10.7 years median)
139 (70.2%):59 (29.8%) 121 (61.1%):78 (39.4%) 9.4 years (9.8 years median)
8 6 4
59 58 81
Number of patients Number of transplants Alive:dead Male:female Mean age at transplant Pretransplant diagnosis ALL AML/MDS Other Donor Related Unrelated
4 (21%) 15 (79%)
96 (47.7%) 105 (52.2%)
HLA matching 6/6 o6/6
14 (73.3%) 5 (26.3%)
169 (84.1%) 32 (15.9%)
Stem cell source BM PBSC UCB UCB/BM
18 (94.7%) 1 (5.3%) 0 0
175 7 16 3
(87.1%) (3.5%) (8%) (1.5%)
Abbreviations: BOS ¼ bronchiolitis obliterans syndrome; BrOb ¼ bronchiolitis obliterans; UCB ¼ umbilical cord blood. UCB and BM were used for same transplant. For patients having multiple HSCTs, age at transplant, donor, degree of HLA matching and stem cell source were recorded per transplant. Pretransplant diagnosis, survival and sex were recorded per patient.
89.9% (median 86.5%). Sixteen patients were able to complete PFTs at diagnosis of BrOb/BOS. At diagnosis, the mean FEV1 for this group was 56.5% predicted (median 48%) and the mean FEV1/FVC was 73.4% predicted (median 62%). Of the patients without BrOb/BOS, pretransplant PFTs were available for 99 of 198 (50%) patients and the mean FEV1 was 98% predicted. In total 39 of 198 (19.7%) patients were unable to complete PFTs due to young age or developmental delay. From the non-BrOb group, 58 of 198 (29.3%) had PFT results available at 1 year (±2 months). Results for 83 of 198 (42%) patients were unavailable due to patient age/developmental capacity or death prior to 1 year following transplant. None of these patients died due to obstructive lung disease. PFT results for 57 of 198 (28.8%) patients were unavailable due to incomplete medical records or because they were performed at an outside institution. Of the patients without BrOb/BOS, the mean 1-year post transplant FEV1 was 95% predicted (median 98%).
Risk factors for bronchiolitis obliterans/bronchiolitis obliterans syndrome All patients diagnosed with BrOb/BOS had one or more of the following potential risk factors for the disease: acute or chronic GVHD, second or third transplant, less than six of six HLA-matched donor, viral pneumonitis or
pneumonia, severe reaction to a drug or transfusion with respiratory involvement or PBSC as the stem cell source. In total 11 of 18 patients had multiple risk factors. Both chronic and acute GVHD were more common in patients with BrOb/BOS than in patients without BrOb/BOS. Acute GVHD was present in 72.2 and 31.8% of BrOb/BOS and non-BrOb patients, respectively (P ¼ 0.001). In total 100% of patients with BrOb/BOS had chronic GVHD versus 31.3% in the non-BrOb group (Po0.001). All 18 patients diagnosed with BrOb/BOS were receiving immunosuppressive medications for GVHD prophylaxis or treatment at the time of diagnosis. In total 3 of 18 (16.7%) of patients with BrOb/BOS had a viral illness or presumed viral illness in the 100 days following HSCT as compared with 6.6% in non-BrOb patients (P ¼ 0.14). All three BrOb/BOS patients with viral or presumed viral illnesses had ﬁndings on CXR and required pulmonary support deﬁned as continuous oxygen by nasal cannula or facemask or greater support for at least 24 h. These included one patient with adenovirus and one with varicella pneumonitis, adenovirus and parainﬂuenza. One patient had symptoms and ﬁndings on CXR consistent with viral pneumonitis or pneumonia, but did not have a speciﬁc virus identiﬁed. A total of 13 non-BrOb patients had a viral illness diagnosed within the ﬁrst 100 days following HSCT. These included varicella pneumonitis (one patient), respiratory syncytial virus (RSV, four patients), parainﬂuenza (ﬁve patients), adenovirus (two patients) and inﬂuenza (two patients). One patient had both RSV and parainﬂuenza. In total 5 of 18 (27.8%) patients with BrOb/BOS had either an anaphylactic/anaphylactoid reaction to a drug, toxic epidermal necrolysis (TEN) or Stevens–Johnson syndrome (SJS) during the ﬁrst 100 days following HSCT. These included an anaphylactoid reaction to inhaled pentamidine (two patients), SJS or TEN (two patients) and one patient with multiple drug reactions that involved respiratory distress. This compares with four (2%) patients in the non-BrOb group (Po0.001). No patients without BrOb/BOS had SJS or TEN.
Treatment of bronchiolitis obliterans/bronchiolitis obliterans syndrome In total 17 of 18 patients received BrOb/BOS-directed therapy. This consisted of the addition of new immunosuppressive agents or the increase in dose of an existing immunosuppressive agent. All 17 patients treated had corticosteroids added or their existing dose increased. Previously initiated calcineurin inhibitors were continued in 15 patients. Mycophenolate mofetil was started in three patients. Azithromycin was added as an anti-inﬂammatory agent in eight patients. Pravastatin, also used as an antiinﬂammatory agent, was initiated in four patients, two as part of a study at our institution. Other agents were used in two instances (pentostatin and rapamycin). The patient who did not receive BrOb/BOS-directed treatment was diagnosed at autopsy following death due to cardiac arrest. Outcome The mean duration of follow-up for living patients was 771.5 days (median 919 days). In total 7 of 18 (38.9%) Bone Marrow Transplantation
Bronchiolitis obliterans following pediatric HSCT CN Duncan et al
Bone Marrow Transplantation
Abbreviations: Az ¼ azithromycin; BOS ¼ bronchiolitis obliterans syndrome; BrOb ¼ bronchiolitis obliterans; FEV1 ¼ forced expiratory volume in 1 s; FVC ¼ forced vital capacity; HSCT ¼ hematopoietic stem cell transplantation; MMF ¼ mycophenolate mofetil; NA ¼ not applicable; Pr ¼ pravastatin; St ¼ corticosteroids. (–) ¼ PFTs performed, but results unavailable. NA ¼ PFTs not performed due to age or developmental ability.
Partial Death Death Partial Death Partial Progressed Death Partial Progressed Partial Stable Death Death Partial Progressed Stable Death St St, other None St, Az, Pr St, Az St St, Pr St, other St, other St St, Az, MMF St, Az, MMF, Pr St St, Az St, Az, Pr St, Az, MMF St, Az St (+) (+) NA (+) (+) (+) (+) (+) (+) (+) (+) (+) NA (+) (+) (+) (+) (+) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
218 293 861 138 191 140 367 755 656 614 328 166 140 234 709 927 348 385
108 110 118 88 76 103 96 – NA NA NA – – 93 105 NA NA 95
72 110 90 90 82 91 93 – NA NA NA – – 89 87 NA NA 95
92 39 75 59 51 72 31 52 44 NA NA 55 21 36 73 75 69 60
81 73 83 78 78 92 56 87 58 NA NA 66 37 58 81 95 84 67
(+) NA (+) (+) (+) (+) (+) NA (+) (+) (+) NA (+) NA (+) (+) NA NA
Treatment Pre-FEV1 Time from HSCT to diagnosis (days) Patient
Summary of patients diagnosed with BrOb/BOS
patients with BrOb/BOS died. Death in ﬁve of seven patients was either partially or completely attributed to respiratory failure. The remaining two patients died of cardiac arrest and of hepatic failure. The mean duration from time of BrOb/BOS diagnosis to death was 312.5 days (median 101 days). Of the surviving patients, three had progressive lung disease, six had a partial response to therapy and two had stable disease. In total 59 of 198 (29.8%) patients without BrOb/BOS died, with a nonrelapse mortality of 14.1% in this group. Nonrelapse causes of death in the non-BrOb group included: multisystem organ failure (10 patients), infectious complications (6 died), respiratory failure (4 patients), acute GVHD (4 patients), unknown causes (2 patients), cardiac complications (1 patients) and fatal neurological event (1 patient). The difference between nonrelapse mortality in non-BrOb patients and in BrOb/BOS patients was statistically signiﬁcant (P ¼ 0.01).
Discussion This retrospective review reports on the incidence and clinical outcome of BrOb/BOS in pediatric allogeneic HSCT patients over a 5-year period at a single center. During this time, 8.3% of patients were diagnosed with BrOb/BOS similar to the incidence reported in studies of adult patients, and higher than the 4.5% in recent pediatric series.8 We investigated the role of GVHD, degree of donor matching, toxic lung injury from drug or transfusion, viral pulmonary illness and source of stem cells as potential risk factors. In total 100% of patients developing BrOb/BOS had previously diagnosed chronic GVHD and 44.4% had at least one additional risk factor. Given this concordance, clinicians should have a high level of suspicion for BrOb/BOS in these at-risk patients and the utility of frequent PFT or radiologic evaluation for early identiﬁcation of affected patients should be addressed in future studies. Despite aggressive immunosuppressive therapy, none of the patients in our cohort had complete resolution of disease. In addition, despite a long course (median time to death ¼ 126 days in patients with BrOb/BOS who died), the eventual outcome for these patients is unsatisfactory. Consistent with previously reports, our patients with BrOb/ BOS experienced a 40% nonrelapse mortality, signiﬁcantly greater than those without BrOb/BOS.2,5,16–23 Thus, clinical trials of new or novel therapies are essential, including agents with anti-inﬂammatory properties such as azithromycin or pravastatin that have more tolerable toxicity proﬁles and potentially additive therapeutic beneﬁt when combined with current treatment modalities. This was a retrospective study of a rare complication of HSCT that signiﬁcantly impacts on outcome. As such, it is limited by a relatively small sample size and potential sampling bias due to the retrospective nature of the study. We attempted to address these issues by including all allogeneic HSCT patients over a 5-year time period, locating data from multiple sources, adhering to a clear deﬁnition of BrOb/BOS and having complete longitudinal follow-up of affected patients. A multicenter, prospective
Bronchiolitis obliterans following pediatric HSCT CN Duncan et al
study of the incidence, risk factors and outcomes of BrOb/ BOS is needed to allow prospective identiﬁcation of patients at greatest risk and early therapeutic intervention to ultimately improve survival in this group of patients.
References 1 Dudek AZ, Mahaseth H, DeFor TE, Weisdorf DJ. Bronchiolitis obliterans in chronic graft-versus-host disease: analysis of risk factors and treatment outcomes. Biol Blood Marrow Transplant 2003; 9: 657–666. 2 Clark JG, Crawford SW, Madtes DK, Sullivan KM. Obstructive lung disease after allogeneic marrow transplantation. Clinical presentation and course. Ann Intern Med 1989; 111: 368–376. 3 Frankovich J, Donaldson SS, Lee Y, Wong RM, Amylon M, Verneris MR. High-dose therapy and autologous hematopoietic cell transplantation in children with primary refractory and relapsed Hodgkin’s disease: atopy predicts idiopathic diffuse lung injury syndromes. Biol Blood Marrow Transplant 2001; 7: 49–57. 4 Paz HL, Crilley P, Patchefsky A, Schiffman RL, Brodsky I. Bronchiolitis obliterans after autologous bone marrow transplantation. Chest 1992; 101: 775–778. 5 Yoshihara S, Tateishi U, Ando T, Kunitoh H, Suyama H, Onishi Y et al. Lower incidence of bronchiolitis obliterans in allogeneic hematopoietic stem cell transplantation with reduced-intensity conditioning compared with myeloablative conditioning. Bone Marrow Transplant 2005; 35: 1195–1200. 6 Holland HK, Wingard JR, Beschorner WE, Saral R, Santos GW. Bronchiolitis obliterans in bone marrow transplantation and its relationship to chronic graft-v-host disease and low serum IgG. Blood 1988; 72: 621–627. 7 Sakaida E, Nakaseko C, Harima A, Yokota A, Cho R, Saito Y et al. Late-onset noninfectious pulmonary complications after allogeneic stem cell transplantation are signiﬁcantly associated with chronic graft-versus-host disease and with the graftversus-leukemia effect. Blood 2003; 102: 4236–4242. 8 Griese M, Rampf U, Hofmann D, Fuhrer M, Reinhardt D, Bender-Gotze C. Pulmonary complications after bone marrow transplantation in children: twenty-four years of experience in a single pediatric center. Pediatr Pulmonol 2000; 30: 393–401. 9 Wieringa J, van Kralingen KW, Sont JK, Bresters D. Pulmonary function impairment in children following hematopoietic stem cell transplantation. Pediatr Blood Cancer 2005; 45: 318–323.
10 Kurland G, Michelson P. Bronchiolitis obliterans in children. Pediatr Pulmonol 2005; 39: 193–208. 11 Mauad T, Dolhnikoff M. Histology of childhood bronchiolitis obliterans. Pediatr Pulmonol 2002; 33: 466–474. 12 Chien JW, Martin PJ, Gooley TA, Flowers ME, Heckbert SR, Nichols WG et al. Airﬂow obstruction after myeloablative allogeneic hematopoietic stem cell transplantation. Am J Respir Crit Care Med 2003; 168: 208–214. 13 Ratjen F, Rjabko O, Kremens B. High-dose corticosteroid therapy for bronchiolitis obliterans after bone marrow transplantation in children. Bone Marrow Transplant 2005; 36: 135–138. 14 Santo Tomas LH, Loberiza Jr FR, Klein JP, Layde PM, Lipchik RJ, Rizzo JD et al. Risk factors for bronchiolitis obliterans in allogeneic hematopoietic stem-cell transplantation for leukemia. Chest 2005; 128: 153–161. 15 Schultz KR, Green GJ, Wensley D, Sargent MA, Magee JF, Spinelli JJ et al. Obstructive lung disease in children after allogeneic bone marrow transplantation. Blood 1994; 84: 3212–3220. 16 Rovelli A, Pezzini C, Silvestri D, Tana F, Galli MA, Uderzo C. Cardiac and respiratory function after bone marrow transplantation in children with leukaemia. Bone Marrow Transplant 1995; 16: 571–576. 17 Quigley PM, Yeager AM, Loughlin GM. The effects of bone marrow transplantation on pulmonary function in children. Pediatr Pulmonol 1994; 18: 361–367. 18 Kaplan EB, Wodell RA, Wilmott RW, Leifer B, Lesser ML, August CS. Late effects of bone marrow transplantation on pulmonary function in children. Bone Marrow Transplant 1994; 14: 613–621. 19 Kaplan EB, Wodell RA, Wilmott RW, Leifer B, Lesser ML, August CS. Chronic graft-versus-host disease and pulmonary function. Pediatr Pulmonol 1992; 14: 141–148. 20 Nenadov Beck M, Meresse V, Hartmann O, Gaultier C. Longterm pulmonary sequelae after autologous bone marrow transplantation in children without total body irradiation. Bone Marrow Transplant 1995; 16: 771–775. 21 Frisk P, Arvidson J, Bratteby LE, Hedenstrom H, Lonnerholm G. Pulmonary function after autologous bone marrow transplantation in children: a long-term prospective study. Bone Marrow Transplant 2004; 33: 645–650. 22 Arvidson J, Bratteby LE, Carlson K, Hagberg H, Kreuger A, Simonsson B et al. Pulmonary function after autologous bone marrow transplantation in children. Bone Marrow Transplant 1994; 14: 117–123. 23 Leneveu H, Bremont F, Rubie H, Peyroulet MC, Broue A, Suc A et al. Respiratory function in children undergoing bone marrow transplantation. Pediatr Pulmonol 1999; 28: 31–38.
Bone Marrow Transplantation