carried out in other cases too, represented a rapid and safe
method to recognize the neoplastic involvement of the
pericardium, in spite of the scanty effusion. Finally, in the
third case, a transbronchial approach was the only way to
obtain a large amount of pericardial fluid after the failure
of the conventional percutaneous approach.
ACKNOWLEDGMENT: We thank Miss Elisa Ceron for the
graphic arrangement and Mrs. Susan Roe for the amendment of
the text.
References
1 Tsang TS, Freeman WK, Sinak LJ, et al. Echocardiographically guided pericardiocentesis: evolution and state-of-the-art
technique. Mayo Clin Proc 1998; 73:647– 652
2 Tomkowski W, Szturmowicz M, Fijalkowska A, et al. New
approaches to the management and treatment of malignant
pericardial effusion. Support Care Cancer 1997; 5:64 – 66
3 Bastian A, Meissner A, Lins M, et al. Pericardiocentesis:
differential aspects of a common procedure. Intensive Care
Med 2000; 26:572–576
4 Tsang TS, El-Najdawi EK, Seward JB, et al. Percutaneous
echocardiographically guided pericardiocentesis in pediatric
patients: evaluation of safety and efficacy. J Am Soc Echocardiogr 1998; 11:1072–1077
5 Drummond JB, Seward JB, Tsang TS, et al. Outpatient
two-dimensional echocardiography-guided pericardiocentesis. J Am Soc Echocardiogr 1998; 11:433– 435
6 Gibbs CR, Watson RD, Singh SP, et al. Management of
pericardial effusion by drainage: a survey of 10 years’ experience in a city centre general hospital serving a multiracial
population. Postgrad Med J 2000; 76:809 – 813
7 Uemura S, Kagoshima T, Hashimoto T, et al. Acute left
ventricular failure with pulmonary edema following pericardiocentesis for cardiac tamponade: a case report. Jpn Circ J
1995; 59:55–59
8 Anguera I, Pare C, Perez-Villa F. Severe right ventricular
dysfunction following pericardiocentesis for cardiac tamponade. Int J Cardiol 1997; 59:212–214
9 Hamaya Y, Dohi S, Ueda N, et al. Severe circulatory collapse
immediately after pericardiocentesis in a patient with chronic
cardiac tamponade. Anesth Analg 1993; 77:1278 –1281
10 Bender F. Hemoperitoneum after pericardiocentesis in a
CAPD patient [letter]. Perit Dial Int 1996; 16:330
11 Taggart SC, Roberts TE, Marshall DA. Chylopericardium
complicating pericardiocentesis for acute idiopathic pericardial effusion. J Thorac Cardiovasc Surg 1994; 108:388 –389
12 Cotoi S, Moldovan D, Carasca E, et al. Sinus node dysfunction occurring immediately after pericardiocentesis. Physiologie 1987; 24:63– 68
13 Calabrese P, Iliceto S, Rizzon P. Pericardiocentesis induced
intrapericardial thrombus: visualization of thrombus and
spontaneous internal lysis by two-dimensional echocardiography. J Clin Ultrasound 1985; 13:49 –51
14 Quecedo E, Febrer I, Martinez-Escribano JA, et al. Tumoral
seeding after pericardiocentesis in a patient with a pulmonary
adenocarcinoma. J Am Acad Dermatol 1994; 31:496 – 497
15 Fisher JD, Kim SG, Ferrik KJ, et al. Internal transcardiac
pericardiocentesis for acute tamponade. Am J Cardiol 2000;
86:1388 –1389
16 De Divitiis M, Dialetto G, Covino FE, et al. An unusual
procedure for the treatment of simultaneous pericardial and
pleural effusions. G Ital Cardiol 1999; 29:796 –798
17 Verrier RL, Waxman S, Lovett EG, et al. Transatrial access to
the normal pericardial space: a novel approach for diagnostic
sampling, pericardiocentesis, and therapeutic interventions.
Circulation 1998; 98:2331–2333
18 Wang KP. How I do it: transbronchial needle aspiration.
J Bronchol 1994; 1:63– 68
19 Duvernoy O, Magnusson A. CT-guided pericardiocentesis.
Acta Radiol 1996; 37:775–778
20 Witte MC, Opal SM, Gilbert JG, et al. Incidence of fever and
bacteremia following transbronchial needle aspiration. Chest
1986; 89:85– 87
21 Ceron L, Cecchetto A, Manzato M, et al. L’agoaspirazione
transbronchiale (T. B. N. A.) nella diagnosi della patologia
ilo-mediastinica e nella stadiazione del tumore polmonare: 6
anni di esperienza. In: Ferrante G, Loizzi M, Deodato G, et
al, eds. Endoscopia toracica: attualità e prospettive. Napoli,
Italy: G. De Nicola, 2001; 183–189
Pulmonary Embolism in
Idiopathic Pulmonary Fibrosis
Transplant Recipients*
Steven D. Nathan, MD, FCCP; Scott D. Barnett, PhD;
Bruce A. Urban, MD; Cynthia Nowalk, RN;
Brian R. Moran, RN, BSN; and Nelson Burton, MD, FCCP
The objectives of the study were the assessment of
the incidence of pulmonary embolism (PE) in lung
transplant recipients. We performed a retrospective
review of the medical records in a tertiary center
lung transplant program. A total of 72 lung transplants were performed. There were seven symptomatic PE events diagnosed among six patients (group
1). All PE events were in the subgroup of patients
with idiopathic pulmonary fibrosis (IPF) [6 of 23
patients (27%) vs 0 among all other patients (0%);
p < 0.001]. All patients were out of the hospital, not
receiving oxygen therapy, and were ambulatory at
the time of the event. The median time to occurrence of the PE was 175 days posttransplant (range,
26 to 541 days). All patients who developed PEs were
men. The group of IPF patients with no PEs was
evenly split between genders (group 2; p < 0.009).
PE patients required a longer posttransplant hospitalization (mean [ⴞ SD], 18.5 ⴞ 3.9 vs 13.5 ⴞ 4 days,
respectively; p < 0.018). Aside from this, there was
no apparent difference in patient functional status
between the two groups. PEs appear to be relatively
common in IPF lung transplant recipients. This
should be considered in the differential diagnosis of
*From the Inova Transplant Center (Drs. Nathan and Burton,
Ms. Nowalk, and Mr. Moran), the Inova Heart Institute (Dr.
Barnett), and the Department of Radiology (Dr. Urban), Inova
Fairfax Hospital, Falls Church, VA.
Manuscript received May 22, 2002; revision accepted October 9,
2002.
Reproduction of this article is prohibited without written permission from the American College of Chest Physicians (e-mail:
[email protected]).
Correspondence to: Steven D. Nathan, MD, FCCP, Inova Transplant Center, 3300 Gallows Rd, Falls Church, VA 22042; e-mail:
[email protected]
1758
Downloaded From: http://publications.chestnet.org/pdfaccess.ashx?url=/data/journals/chest/21993/ on 06/18/2017
Selected Reports
any such patient who presents with dyspnea or
hypoxia posttransplant. Patients do not appear to
have been predisposed to their embolic events
through lack of activity or prolonged hospital stays.
(CHEST 2003; 123:1758 –1763)
group of patients with PEs. The time to presentation and the
clinical manifestations of the PEs also were assessed. Continuous
and categoric data were assessed via t tests and ␹2 tests, where
appropriate. Tests were two-tailed, and a p value of ⱕ 0.05 was
considered to be statistically significant. All statistical analyses
were conducted using statistical software (SAS, version 8.01; SAS
Institute; Cary, NC).
Key words: lung; lung diseases; obstructive; pulmonary fibrosis;
respiratory function tests; thromboembolism; transplantation
Abbreviations:
IPF ⫽ idiopathic
pulmonary
fibrosis;
LSLTx ⫽ left single-lung transplant; PE ⫽ pulmonary embolus;
PFT ⫽ pulmonary function test; RSLTx ⫽ right single-lung
transplant
here are many causes of allograft dysfunction in the
T lung
transplant recipient. Most of these are of paren-
chymal or airway origin. The most common complications
seen are related to infection, ischemia/reperfusion, or
immunologic injury. Posttransplant pulmonary vascular
complications are rare, although complications from vascular anastomotic problems can occur.1,2 Pulmonary embolisms (PEs) have been reported to occur after transplant, but there have been no reports of a predilection for
any particular patient group.3
Some patients with idiopathic pulmonary fibrosis (IPF)
will succumb to PEs as the terminating event of their
disease.4 This has been attributed mostly to inactivity in
the later stages of the disease. However, to our knowledge,
there have been no reports of a predisposition among IPF
patients for the development of thromboembolic complications. We report a high incidence of PEs in IPF patients
who received single-lung transplants.
Materials and Methods
We reviewed the records of all patients who had undergone
transplantion at our facility between September 1996 and January
2002. The breakdown of primary disease for which the patients
received transplants is shown in Table 1. Patients who developed
PEs were compared and contrasted to those with similar primary
diseases who did not. Specifically in this regard, we compared the
pretransplant pulmonary function test (PFT) results from IPF
patients who had developed PEs (group 1) to IPF patients who
did not develop PEs (group 2). Comparisons also were made
between the length of the posttransplant hospital stays and the
results of the posttransplant PFTs between the two groups. Our
program did not obtain routine perfusion scans posttransplant,
and therefore there were no baseline scans for comparison in the
Results
There were a total of 72 patients who received lung
transplants during the time period. All except two patients
were successfully discharged from the hospital posttransplant, and only one patient died within the first month for
a 1-month survival rate of 98.6%. There were seven
episodes of PEs among six patients. All episodes occurred
in the subpopulation of IPF transplant recipients (6 of
23 patients) for an incidence of 27%. None of the
recipients with other primary diseases (0 of 49 patients;
0%) developed clinical evidence or were diagnosed with
PEs (p ⬍ 0.001). Of the six patients, all (left-sided lung
transplant [LSLTx], three patients; right-sided lung transplant [RSLTx], three patients) developed PEs in their
transplanted lungs, while the one patient who had two
embolic episodes had his first to the left native lung. Of
note, this patient was not receiving coumadin therapy at
the time of his second event. In terms of diagnostic
procedures, one of the patients had undergone a ventilation-perfusion scan, three of the patients had undergone
pulmonary angiograms, and five of the patients had undergone spiral CT scans (Fig 1, 2). The clinical characteristics of those IPF patients who developed PEs posttransplant (group 1) and those who did not (group 2) are
contrasted in Table 2. There were no differences in their
pretransplant or first posttransplant spirometric indexes.
Similarly, there was no difference in the maximal values
that the groups attained posttransplant. Based on the
FEV1 value obtained in closest proximity to the subsequent PE event (mean, 15 days prior), there was no
patient who qualified as having bronchiolitis obliterans
syndrome at the time of the PE. Only three of the patients
underwent PFTs within 3 days of their embolic event. In
one of these patients, there was a 9% decrement in the
FVC and FEV1 on the day of the PE. Spirometry was
stable in the second patient, and in the third patient the
set of PFTs that he underwent on the day prior to the PE
represented his best posttransplant spirometry effort.
The clinical presentations of the six patients are shown in
Table 3.
Discussion
Table 1—Summary of Transplant Recipients by
Primary Disease
Transplants
Primary Disease
No.
%
COPD
IPF
Cystic fibrosis
Connective tissue disorder
Pulmonary hypertension
Sarcoid
Lymphagioleiomyomatosis
34
23
4
3
3
3
2
47.2
31.9
5.6
4.2
4.2
4.2
2.8
The major causes of death in the early post-lung
transplant period include primary graft failure, infection,
and, rarely, acute rejection. Although there has been one
prior report3 of thromboembolic complications post-lung
transplant and after heart-lung transplant, there is generally a paucity of data on this potentially lethal complication
in lung transplant recipients. Kroshus et al3 reported a
12.1% incidence of PE. However, there was no mention of
any predilection for one patient group. In our series, we
have shown a strong propensity for PEs in the IPF
subpopulation of transplant recipients. Although the difference in the occurrence rate between our COPD and
IPF patients appears to be quite striking, with the small
www.chestjournal.org
Downloaded From: http://publications.chestnet.org/pdfaccess.ashx?url=/data/journals/chest/21993/ on 06/18/2017
CHEST / 123 / 5 / MAY, 2003
1759
Figure 1. Spiral CT scan of a 56-year-old man with PE. The soft-tissue window reveals large, discrete,
filling defects in the main right pulmonary artery of the transplanted lung.
numbers in our study we cannot exclude the possibility
that we have erroneously concluded that the occurrence
rates in COPD and IPF patients differ (ie, type I error).
Similarly, the incidence of this complication exclusively in
male recipients is noteworthy but could also represent a
type 1 error. With patients having complicated conditions
such as lung transplant recipients, we also cannot exclude the
possibility that other patients may have had PEs that went
undiagnosed. However, it is noteworthy that in four of the
episodes, these were not subtle events, with three of the
patients presenting with oxygen saturations in the 60% range
(one of whom had a respiratory arrest) and one patient
presenting with hemodynamic compromise and right heart
failure.
Why patients with IPF might be more predisposed to
thromboembolism is uncertain and raises a number of
issues. First, are these patients sicker and hence less
mobile prior to undergoing transplantation? This appears
to be unlikely since, compared to patients in the COPD
population, IPF patients generally have been sick for less
time when listed for transplantation. Also, a prerequisite
for acceptance into our transplant program is that patients
have to be actively engaged in pulmonary rehabilitation.
The median time to occurrence of the embolic events was
175 days, which would suggest that patients’ pretransplant
functional status had a minimal impact on their predisposition for the embolic events.
There appeared to be a slightly longer hospital length of
stay posttransplant, which might imply that these patients
had a slower recovery. However, a median in-hospital stay
of 19 days is unremarkable for patients with transplanted
lungs. Indeed, most of the embolic events were far removed
from the patient’s initial hospitalization, with more than half
of the events occurring ⬎ 4 months after transplantation.
The functional status of the patient at the time of the
event also does not appear to have played a significant
predisposing role. All patients were oxygen-independent,
ambulatory without limitation, and outpatients at the time
of the events. As a further objective measure of their
functional status, we found no difference between their
posttransplant PFT results compared to the IPF patients
who had no thromboembolic complications. The PFT
results that were available prior to the PE events also
attest to the patients’ otherwise stable pulmonary status at
the time of their thromboembolic complication, with none
of the patients qualifying as having bronchiolitis obliterans
syndrome.
It is interesting to speculate that there may be something inherent in the disease itself that predisposes patients to thromboembolism. Indeed, approximately 3% of
patients with IPF will succumb to a thromboembolic
event.4 This possibly represents an underestimation as
some patients with IPF who decompensate may incor-
1760
Downloaded From: http://publications.chestnet.org/pdfaccess.ashx?url=/data/journals/chest/21993/ on 06/18/2017
Selected Reports
Figure 2. Spiral CT scan of a 58-year-old man with PE. The soft-tissue window reveals large, discrete,
filling defects in the left lower lobe pulmonary artery of the transplanted lung.
rectly have this attributed to progression of their underlying disease. Although there has been at least one case
report5 of an association with antiphospholipid syndrome,
to our knowledge, there has been no systematic research into
whether IPF patients have a thromboembolic predisposition.
Until the advent and popularity of spiral CT scanning,
Table 2—Clinical Features of the Two Groups of IPF Patients*
Parameter
Group 1 (n ⫽ 6)
Group 2 (n ⫽ 17)
p Value
Gender
Male
Female
Age, yr
Pretransplant FVC, % predicted
Pretransplant FEV1, % predicted
First posttransplant FVC, % predicted
First posttransplant FEV1,† % predicted
Maximum posttransplant FVC,‡ % predicted
Maximum posttransplant FEV1, % predicted
Mean time to first PFT, d
Posttransplant LOS, d
Time to PE,§ d
6
0
58.7 ⫾ 4.3
55.5 ⫾ 18.0
60.2 ⫾ 20.9
58.8 ⫾ 12.4
56.5 ⫾ 11.0
70.0 ⫾ 12.6
76.7 ⫾ 12.5
15.2 ⫾ 6.9
18.5 ⫾ 3.9
175.0 ⫾ 193.9
8
9
54.7 ⫾ 7.0
49.3 ⫾ 13.0
54.4 ⫾ 13.8
52.9 ⫾ 15.6
52.0 ⫾ 14.2
78.7 ⫾ 18.4
80.2 ⫾ 18.4
12.6 ⫾ 5.1
13.5 ⫾ 4.0
NA
0.009
0.220
0.381
0.460
0.398
0.500
0.306
0.673
0.356
0.018
NA
*Values given as mean ⫾ SD, unless otherwise indicated. NA ⫽ not applicable; LOS ⫽ length of stay.
†First posttransplant FVC and FEV1 data are from first set of PFT results obtained posttransplant.
‡Max posttransplant FVC and FEV1 data are taken from posttransplant PFT results with highest FEV1.
§Includes both PE episodes in patient 4.
www.chestjournal.org
Downloaded From: http://publications.chestnet.org/pdfaccess.ashx?url=/data/journals/chest/21993/ on 06/18/2017
CHEST / 123 / 5 / MAY, 2003
1761
1762
Downloaded From: http://publications.chestnet.org/pdfaccess.ashx?url=/data/journals/chest/21993/ on 06/18/2017
Selected Reports
3
RSLTx/R
6
30
101
541
26
339
132
56
Time
Posttransplant,
d
SOB; large right effusion; room air saturation, 82%;
p ⫽ 106/min; BP, 106/73 mm Hg; no PFTs
SOB; oxygen saturation in the 60% range; BP, 100/60
mm Hg; p ⫽ 90/min
No PFTs
SOB; oxygen saturation in mid-80% range; p ⫽ 80/
min; BP, 113/70 mm Hg; 9% decrease in FVC and
FEV1
SOB; saturation, 94% on room air; p ⫽ 140/min; BP,
100/50 mm Hg; right heart failure; highest PFT set
on the day prior
Respiratory arrest with oxygen saturations in the 60%
range; p ⫽ 113/min; BP, 155/90 mm Hg
Y
SOB; room air saturation in 60% range; p ⫽ 100/min;
BP, 170/100 mm Hg; home spirometry stable; no
PFTs done
SOB; room air saturation, 88%; P ⫽ 87/min; BP, 120/
70 mm Hg; PFTs 3 d prior stable
Y
N
Y
Y
N
N
DVT
Clinical Presentation
V̇/Q̇ scan; pulmonary
angiogram
Spiral CT scan
Spiral CT scan
Spiral CT scan
Pulmonary angiogram
Spiral CT scan
Pulmonary angiogram
Spiral CT scan
Diagnosis
TPA, IV heparin
TPA, heparin,
coumadin, IVC
filter
IV heparin, coumadin
IV heparin, coumadin
TPA, heparin,
coumadin, IVC
filter
Mechanical
thrombectomy
TPA, IVC filter
IV heparin, coumadin,
IVC filter
Treatment
Recovered
Recovered
Recovered
Recovered
Recovered
Recovered
Outcome
Died 16 h after presentation
*L ⫽ left; R ⫽ right; SOB ⫽ shortness of breath; V̇/Q̇ ⫽ ventilation-perfusion; Y ⫽ yes; N ⫽ no; IVC ⫽ inferior vena cava; TPA ⫽ tissue plasminogen activator; p ⫽ pulse.
RSLTx/R
5
RSLTx/R
LSLTx/L
2
Episode 2
LSLTx/L
1
RSLTx/L
LSLTx/L
Patient No.
4
Episode 1
Transplant/
PE side
Table 3—Clinical Presentation of PEs in IPF Patients*
the only definitive way to diagnose PEs in this patient
population was via pulmonary angiography, since ventilation-perfusion scanning is known to be inaccurate in the
IPF patient.6 Contrast-enhanced spiral CT scanning has
been used increasingly to diagnose PE, and, although
there are data to attest to its accuracy, the sensitivity and
specificity have not as yet been fully determined.7 For the
five patients in our cohort in whom PE was diagnosed by
this modality, the detected PEs were large in size and
central in location, and as such were diagnosed with
certainty. Spiral CT scanning also helped to quantify the
clot burden, which supported the clinical decision to use
thrombolytic therapy in two of the patients.
The clinical presentation of PE in this group of patients
is well worth noting. In most the cases, the initial index of
suspicion for PE was high, based on their clinical presentation. Consequently, only one patient underwent PFTs
on presentation as part of his workup. Although this
patient’s PFT results were reduced, there is no apparent
reason to believe that PE should be added to the list of
causes of reduced spirometry in lung transplant recipients.
Indeed, when there is hypoxemia that is out of proportion
to any spirometric or radiographic changes, then the index
of suspicion for PE should be raised. It is not surprising
that 86% of the embolic events (six of seven events) were
on the side of the transplant, since it is well-recognized
that the majority of the blood flow goes to the allograft
posttransplant. All of the patients presented with shortness
of breath, and none of them had associated chest pain.
This is interesting from the standpoint that the patients
still had their native parietal pleura. In terms of the clinical
consequences of their PEs, the lack of a bronchial circulation theoretically placed them at higher risk for lung
infarction. Chest imaging, including CT scans, in four of
the patients at the time of their presentation did not show
any evidence of this. One of the patients had a respiratory
arrest requiring intubation but was stable hemodynamically throughout the event. One of the six patients presented with associated right heart failure and was the only
patient to succumb to the event. The other patients
appeared to tolerate their embolic events quite well from
a hemodynamic standpoint, despite evidence of large clot
burden in five of the remaining six episodes. This raises
the notion of their right ventricles being “preconditioned”
by virtue of their underlying IPF and associated pretransplant pulmonary hypertension.
In summary, thromboembolic events appear to be relatively common in lung transplant recipients with IPF. This
should be considered in the differential diagnosis of any such
patient who presents with shortness of breath and/or hypoxia
posttransplant. With their abnormal lung parenchyma, spiral
CT scans can be very useful diagnostic tools. The question of
whether IPF patients in general have a predilection for
thromboembolic events may warrant further study.
References
1 Griffith BP, Magee MJ, Gonzalez IF, et al. Anastomotic
pitfalls in lung transplantation. J Thorac Cardiovasc Surg
1994; 107:743–754
2 Leibowitz DW, Smith CR, Michler RE, et al. Incidence of
3
4
5
6
7
pulmonary vein complications after lung transplantation: a
prospective transesophageal echocardiographic study. J Am
Coll Cardiol 1994; 24:671– 675
Kroshus TJ, Kshettry VR, Hertz MI, et al. Deep venous
thrombosis and pulmonary embolus after lung transplantation. J Thorac Cardiovasc Surg 1995; 110:540 –544
Panos RJ, Mortenson RL, Niccoli SA, et al. Clinical deterioration in patients with idiopathic pulmonary fibrosis: causes
and assessment. Am J Med 1990; 88:396 – 404
Kelion AD, Cockcroft JR, Ritter JM. Antiphospholipid syndrome in a patient with rapidly progressive fibrosing alveolitis. Postgrad Med J 1995; 71:233–235
Pochis WT, Krasnow AZ, Collier BD, et al. Idiopathic
pulmonary fibrosis: a rare cause of scintigraphic ventilationperfusion mismatch. Clin Nucl Med 1990; 15:321–323
Blachere H, Latrabe V, Montaudon M, et al. PEssm revealed
on helical CT angiography: comparison with ventilationperfusion radionuclide lung scanning. AJR Am J Roentgenol
2000; 174:1041–1047
Chronic Eosinophilic
Pneumonia*
A Case Report and National Survey
Catherine Wubbel, MD; Deborah Fulmer, MD; and
James Sherman, MD
Few reports of chronic eosinophilic pneumonia
(CEP) in the pediatric population can be found in the
literature. Our patient, a 16-year-old male subject
presenting with signs and symptoms of CEP,
prompted a survey of pediatric pulmonary training
centers in the United States to determine the prevalence of eosinophilic pneumonia. The survey
showed a low prevalence of acute eosinophilic pneumonia and CEP in the pediatric population, with an
overall male/female ratio of 1.6:1.
(CHEST 2003; 123:1763–1766)
Key words: adolescence; BAL; chronic eosinophilic pneumonia;
transbronchial lung biopsy
Abbreviations: ACR ⫽ American College of Rheumatology;
AEP ⫽ acute eosinophilic pneumonia; CEP ⫽ chronic eosinophilic pneumonia; CSS ⫽ Churg-Strauss syndrome
pneumonia is a rare cause of lung disease
E inosinophilic
children and adolescents. The relatively nonspe1
cific nature of the clinical presentation of this disease
*From the Department of Pediatrics (Drs. Wubbel and Sherman), University of Florida College of Medicine, Gainesville, FL;
and Department of Pediatrics (Dr. Fulmer), Memorial Health
University Medical Center, Mercer University, Savannah, GA.
Manuscript received May 28, 2002; revision accepted September
27, 2002.
Reproduction of this article is prohibited without written permission from the American College of Chest Physicians (e-mail:
[email protected]).
Correspondence to: Catherine Wubbel, MD, Department of
Pediatrics, University of Florida College of Medicine, 1600 SW
Archer Rd, Gainesville, FL 32610; e-mail: [email protected]
www.chestjournal.org
Downloaded From: http://publications.chestnet.org/pdfaccess.ashx?url=/data/journals/chest/21993/ on 06/18/2017
CHEST / 123 / 5 / MAY, 2003
1763