Downloaded from www.ajronline.org by 120.39.180.118 on 10/18/15 from IP address 120.39.180.118. Copyright ARRS. For personal use only; all rights reserved
461
Diffuse Pulmonary Alveolar Hemorrhage After Bone Marrow Transplantation: Radiographic Findings in 39 Patients
J. Witte1 Jud W. Gurney1 Richard A. Robbins2 James Linder3 Stephen I. Rennard2 Mark Ameson4 William P. Vaughan4 Elizabeth C. Reed4 Karel A. Dicke4
Diffuse
Robert
alveolar
transplantation.
transplantation the findings
diffuse
hemorrhage We
patients
clinical
hemorrhage
transplantation,
and
the
the
complication
radiographic
with a diagnosis
with the patients’
alveolar
is a life-threatening
investigated
after
abnormalities
of diffuse
alveolar
that
hemorrhage
course. The initial radiographic an average of 1 1 days abnormalities
preceded
the
marrow 39
in
and correlated
abnormalities after after bone marrow
developed
radiographic
bone occurred
clinical
diagnosis
by
an average of 3 days. Twenty-seven patients initially had bilateral radiographic abnormalities; 10 initially had unilateral abnormalities (seven in the right lung, three in the left lung). Two patients had normal chest radiographs throughout their clinical course. All 37 patients with radiographic abnormalities had abnormalities involving the central portion of the lung, primarily the middle and lower lung zones. The initial radiographic pattern was interstitial in 27 and alveolar in 10. In 24 patients, radiographic abnormalities were initially judged to be mild; three were severe from the onset. Radiographic abnormalities
rapidly
worsened
in most
patients
over
6 days.
In 30
patients,
diffuse
bilateral radiographic abnormalities involving all lung zones developed. Eleven patients persisted in having only interstitial radiographic abnormalities; 26 had a confluent alveolar pattern. At the height of radiographic abnormalities, 27 cases were judged to be severe, and only one case was judged to be mild. The mortality rate in patients with diffuse alveolar hemorrhage was 77%. The radiographic abnormalities of diffuse alveolar hemorrhage are nonspecific and usually precede the clinical diagnosis. The clinical course after hemorrhage is short, often resulting in death. AJR
157:461-464,
Pulmonary Received February vision April 10,1991.
19, 1991;
accepted
after re-
Presented at the annual meeting of the American Roentgen Ray Society, Washington, DC, May 1990. Department of Radiology, University of Nebraska Medical Center, 600 S. 42nd St., Omaha, NE 68198-1045. Address reprint requests to J. W. Gurney. 2Pulmonary and Critical Care Section, Department Medical
of Internal Center,
3Department braska Medical
Medicine, Omaha,
university
of Nebraska
NE 68198-1045.
of Pathology, Center, Omaha,
September
complications
are common
after
bone
marrow
transplantation
and
include opportunistic infection, cardiac and noncardiac pulmonary edema, drug and radiation toxicity, and metastatic spread of tumor [1, 2]. Diffuse alveolar hemorrhage (DAH) is another newly described complication occurring in transplantation patients [3]. The radiographic abnormalities that occur with DAH in bone marrow transplantation patients have not previously been described. It was our purpose to review the radiographic abnormalities that occur with DAH and correlate these with the patient’s
Materials
clinical
course.
and Methods
Patients
University of NeNE 681 95-1 045.
The
hospital
records
and
chest
Hematology and Oncology Section, Department of Internal Medicine, University of Nebraska Medical Center, Omaha, NE 68195-1045.
a discharge diagnosis of DAH marrow transplantations were
0361 -803x/91/1 573-0461 Roentgen Ray Society
with Hodgkin
C American
1991
whom two
DAH eventually lymphoma,
with ovarian
radiographs
were reviewed performed for
developed,
including
39 bone marrow
of
15
patients
six with breast carcinoma,
carcinoma,
one
with
transplant
from a total of 288 patients. a variety of medical conditions
aplastic
with
non-Hodgkin
three with melanoma, anemia,
and
one
recipients
Autologous
with
bone
in the patients lymphoma, two
with acute
in
nine
with sarcoma, myelogenous
Downloaded from www.ajronline.org by 120.39.180.118 on 10/18/15 from IP address 120.39.180.118. Copyright ARRS. For personal use only; all rights reserved
462
WITTE
leukemia. The average age of the patients with DAH was 41 years (range, 1 6-68 years). There were 21 women and 1 8 men. Autopsies were performed in 21 (70%) of the 30 patients who died. DAH was the primary cause of death or significantly contributed to the cause of death in 1 9 (90%) of the 21 patients. At autopsy, the lungs were found to have scattered areas of hemorrhage. Microscopic examination of the lungs revealed intraalveolar hemorrhage. Reactive epithelial
Diagnosis
cells
were
often
present,
but
no pathogens
were
identified.
of DAH
DAH was diagnosed on the basis of the findings from bronchoalveolar lavage when the lavage fluid became progressively bloodier with each aliquot aspirated from separate subsegments of the lung. DAH was diagnosed only when other conditions known to result in hemorrhage were excluded [3, 41.
ET
AL.
AJR:157,
The syndrome
progressed
had the most
severe
We reviewed the chest radiographs of all patients in whom DAH was diagnosed, beginning with the pretreatment film and all films after transplantation. In most patients, portable films were obtained daily. The films were reviewed by two radiologists unaware of the time of bronchoscopy or the date of the diagnosis of DAH. The radiographs were reviewed independently by each radiologist and the findings were agreed on by consensus. Serial radiographs were analyzed for the following: (1) the initial pattern of pulmonary involvement (i.e., interstitial or alveolar), the distribution within the chest, and the pattern’s severity; (2) the time between bone marrow transplantation and initial radiographic abnormality and the time between the initial radiographic abnormality and bronchoscopic diagnosis of DAH; (3) the time between the initial radiographic abnormality and the radiograph showing the most severe radiographic abnormality; and (4) other radiographic abnormalities that developed, such as pleural
effusions or cardiomegaly. An interstitial pattern was defined as one in which reticular or small nodular opacities predominated, or in which the vessel margins became indistinct. An alveolar pattern was defined as one in which acinar opacification with air bronchograms predominated. The distilbution of radiographic abnormalities was categorized into two regions: (1)central or peripheral and (2) upper, middle, and lower zones. This distribution was determined by reference to the midpoint of the interlobar pulmonary artery. A central region was within a 4-cm radius from the artery and a peripheral region was from this 4-cm edge to the edge of the lung. Mirror-image regions were used for the left lung. The lower zone was a horizontal region from 3 cm below the midpoint of the interlobar pulmonary artery to the base of the lung; the middle zone was a horizontal region from 3 cm below to 3 cm above the interlobar pulmonary artery; the upper zone was a horizontal region from 3 cm above the interlobar pulmonary artery to the lung apex. Mirror-image boundaries were used in the left lung. Severity was graded as mild if the normal pulmonary arteries in the lung were easily identified but their borders were indistinct, moderate if the arteries were partially obscured but still visible, and severe if the arteries were completely obscured.
The initial radiographic abnormalities of 10.6 ± 8.5 days after bone marrow
identified
(range,
0-27
(32%) of these 37 patients, the radiographic extended to the peripheral lung. There were’no initial radiographic
middle
abnormalities
lung,
limited
abnormalities patients with
to the peripheral
lung.
(59%) had involvement of the upper (1 4%), this region was the only one patients (78%) had involvement of the
and 31 patients
(84%)
had involvement
of the
lower lung. A total of four patients (1 1%) had initial bilateral radiographic abnormalities involving all lung zones. From their initial radiographic presentation, most patients developed diffuse radiographic abnormalities in the lungs (Figs.
1 and
2). Thirty
radiographic
patients
abnormalities
developed
involving
diffuse
all lung zones,
tients had persistent unilateral involvement and four patients had bilateral involvement
lower lung zones. the left lung.
No patients
The initial radiographic
of the right lung, of middle and
had unilateral
pattern
bilateral
three pa-
involvement
was interstitial
of
in 27 patients
(73%) and alveolar in 10 patients (27%). At peak, 1 1 patients (30%) persisted in having only interstitial radiographic abnormalities and 26 patients (70%) had radiographic abnormalities that progressed to an alveolar pattern. Initially, 24 patients (65%) had radiographic abnormalities that were mild, 10(27%) were moderate, and three (8%) were severe. Only one patient (3%) had peak radiographic abnormalities that were mild, nine (24%) had moderate abnormalities, and 27 (73%) had severe abnormalities.
Other radiographic patients
findings
included
(1 4%) and mild cardiomegaly
pleural effusions in seven
patients
in five (19%).
None of the patients with pleural effusions had cardiomegaly. Of the 37 patients whose radiographs were abnormal, 23 (62%) had no radiographic evidence of resolution, eight (22%) had partial resolution, and six (1 6%) resolved completely. The clinical outcome of patients with DAH was poor. Thirty died (77%)
and nine recovered.
Discussion DAH is a serious
estingly,
transplantation (range, 0-31 days) and preceded the diagnosis by bronchoalveolar lavage by an average of 3.4 ± 4.6 days (range, 0-24 days).
days).
bilateral radiographic abnormalities, seven had abnormalities initially limited to the right lung, and three had abnormalities initially limited to the left lung. The regional involvement was more common in the central and lower lung (Figs. 1 and 2). All 37 patients with abnormal radiographs had initial involvement of the central lung. In 12
to contribute drug toxicity, have
in patients
undergoing
The clinical signs and symptoms
patients
hemoptysis
The cause
DAH
complication
transplantation.
nonspecific-most were seen an average
an average
abnormalities
Of the 39 patients, two had normal findings on chest radiographs. Of the remaining 37 patients, 27 had initial
marrow Results
abnormalities
of 5.9 ± 6.5 days after the initial radiographic were
1991
and the chest radiographs
radiographic
Twenty-two patients lung. In five patients involved. Twenty-nine
Radiographs
rapidly,
September
have hypoxia
is uncommon
of DAH is unknown.
and dyspnea.
bone
are Inter-
[3]. Several
factors
are thought
to its pathogenesis, including radiation toxicity, and WBC influx into the lung [3]. Patients with a higher
prevalence
of chest
irradiation,
either
Downloaded from www.ajronline.org by 120.39.180.118 on 10/18/15 from IP address 120.39.180.118. Copyright ARRS. For personal use only; all rights reserved
AJR:157, September
BONE
1991
MARROW
TRANSPLANTS
AND
ALVEOLAR
HEMORRHAGE
Fig. 1.-Chest radiographs in a 37-year-old woman with breast carcinoma. A, Severe interstitial prominence is present in central lung and also involves middle and lower lungs 11 days B, 5 days later, there has been a progression to alveolar pattern and extension to all regions of lung.
A
B
after
463
bone
marrow
transplantation.
C
Fig. 2.-Chest radiographs in a 45-year-old man with melanoma. A, Baseline radiograph coned to left base shows no abnormalities. B, 11 days after bone marrow transplantation, vessels are now indistinct centrally. C, 8 days later, a severe alveolar pattern is present extending to lung periphery.
whole body, chest, marrow transplantation
administration,
or
mediastinal
irradiation
before
bone
[1-3]. Treatment consists of platelet and recently, high-dose steroids have shown
promise in reducing the high mortality this syndrome (Armitage JA, unpublished
rate associated observations).
with
We found the radiographic abnormalities in DAH to be nonspecific. The findings associated with hemorrhage in the bone
marrow
transplantation
population
do not
differ
from
descriptions of hemorrhage that occurs in other conditions [5, 6]. Most patients initially exhibited a mild interstitial or alveolar pattern in the central and lower lung zones. This is indistinguishable from pulmonary edema or opportunistic in-
fection, other transplantation.
common complications after bone marrow In addition, cardiomegaly and pleural effu-
sions were occasionally seen in patients findings are usually suggestive of pulmonary
with DAH; these edema. Although
464
DAH
WITTE
was
normality,
Downloaded from www.ajronline.org by 120.39.180.118 on 10/18/15 from IP address 120.39.180.118. Copyright ARRS. For personal use only; all rights reserved
infection, istically,
usually
a bilateral
process,
unilateral abwith opportunistic
an initial
which could be easily confused
was sometimes seen (1 0 cases, 27%). CharacterDAH was fulminant, with rapid progression during 6
days to a diffuse, severe alveolar pattern. This mirrored the clinical course, in which 77% died. Although clinical signs and symptoms and the radiographic time course may help in the evaluation of the transplant recipient with diffuse pulmonary abnormalities, ultimately bronchoscopy or open lung biopsy
is needed for differentiation. The reported prevalence of hemorrhage after bone marrow transplantation varies. Pagani et al. [7] found one case of pulmonary hemorrhage out of 35 children after bone marrow transplantation.
Cordonnier
7% in 130 patients Crawford
et al. [1] found
after bone marrow
et al. [8] found
an 8% prevalence.
a prevalence
transplantation,
of
and
In our population
of patients, 14% of the patients undergoing transplantation suffered DAH. It has been speculated that DAH may be related to idiopathic interstitial pneumonitis [3, 8-1 0]. Both have a similar prevalence, time of onset after bone marrow transplantation (within the first 7 weeks after bone marrow transplantation), and a high mortality rate. Histologically, nonspecific inflammatory changes are seen in idiopathic interstitial pneumonitis; however, hemorrhage has not been specifically mentioned. Whether the two disorders represent the same entity is unknown
[9, 101.
Radiographic abnormalities of DAH usually occur within the first 2 weeks after transplantation and rapidly progress to severely involve both lungs. The findings, however, are nonspecific
and are identical
and infection.
to the findings
in pulmonary
edema
ET AL.
AJR:157.
September1991
ACKNOWLEDGMENTS We thank Mary Wilke for manuscript preparation. We also thank the UNMC Bone Marrow Transplant Pulmonary Study Group for their assistance in this study.
REFERENCES 1 . Cordonnier C, Bemaudin J-F, Bierling P, Huet V. Vemant J-P. Pulmonary complications occurring after bone marrow transplantation. Cancer 1986;58: 1047-1054 2. Krowka MJ, Rosenow EC, Hoagland HC. Pulmonary complications of bone marrow transplantation. Chest 1985;87:237-246 3. Robbins RA, Under J, Stahl MG, et al. Diffuse alveolar hemorrhage in autologous bone marrow transplant recipients. Am J Med 1989;87: 511-518 4. Under J, Robbins A, Rennard S. Cytologic criteria for diffuse alveolar hemorrhage. Acta Cytol 1988;32:763 5. Albelda SM, Gefter WB, Epstein DM, Miller WT. Diffuse pulmonary hemorrhage: a review and classification. Radiology 1985;154:289-297 6. Palmer PES, Finley TN, Drew WL, Golde DW. Radiographic aspects of occult pulmonary haernorrhage. Clin Radio! 1978;29: 139-1 43 7. Pagani JJ, Kangarloo H, Gyepes MT, Feig SA, Falk PM. Radiographic manifestations of bone marrow transplantation in children. AJR 1979;132:883-890 8. Crawford SW, Hackman AC. Clark JG. Open lung biopsy: diagnosis of diffuse pulmonary infiltrates after bone marrow transplantation. Chest 1988;94:949-953 9. Cardozo BL, Hagenbeek A. Interstitial pneumonitis following bone marrow transplantation: pathogenesis and therapeutic considerations. Eur J Cancer Clin Oncol 1985;21 :43-51 10. Meyers JD, Floumoy N, Thomas ED. Nonbactenal pneumonia after allogeneic marrow transplantation: a review of ten year’s experience. Rev Infect Dis 1982;4:1 119-1132