Spontaneous Pneumomediastinum: A Comparative Study and Review of the

Literature
Manuel Caceres, Syed Z. Ali, Rebecca Braud, Darryl Weiman and H. Edward Garrett,
Jr
Ann Thorac Surg 2008;86:962-966
DOI: 10.1016/j.athoracsur.2008.04.067

The online version of this article, along with updated information and services, is
located on the World Wide Web at:
http://ats.ctsnetjournals.org/cgi/content/full/86/3/962

The Annals of Thoracic Surgery is the official journal of The Society of Thoracic Surgeons and the
Southern Thoracic Surgical Association. Copyright 2008 by The Society of Thoracic Surgeons.
Print ISSN: 0003-4975; eISSN: 1552-6259.

Downloaded from ats.ctsnetjournals.org by on April 16, 2012

Spontaneous Pneumomediastinum: A Comparative

Study and Review of the Literature
Manuel Caceres, MD, Syed Z. Ali, MD, Rebecca Braud, BS, Darryl Weiman, MD,
and H. Edward Garrett, Jr, MD
Department of Cardiothoracic Surgery, Appalachian Regional Healthcare, Hazard, Kentucky; Department of Cardiothoracic
Surgery, University of Tennessee HSC, Memphis, Tennessee; Louisiana State University, Baton Rouge, Louisiana;
and Department of Cardiothoracic Surgery, Baptist Memorial Hospital, Memphis, Tennessee
GENERAL THORACIC

Background. Spontaneous pneumomediastinum (SPM)

is an unusual occurrence with few cases reported. It is
seen after intrathoracic pressure changes leading to alveolar rupture and dissection of air along the tracheobronchial tree. This study was undertaken to provide a
thorough clinical and radiologic analysis of this patient
population.
Methods. A retrospective comparative analysis was
performed on patients with SPM over 12 years. Patient
demographics, clinical presentation, and radiographic
and diagnostic studies were recorded. A clinical and
radiologic comparison was performed with secondary
pneumomediastinum.
Results. Seventy-four patients were identified with a
diagnosis of pneumomediastinum. A total of 28 patients
with SPM were identified. The major initial complaints
were chest pain (54%), shortness of breath (39%), and
subcutaneous emphysema (32%). The main triggering
events were emesis (36%) and asthma flare-ups (21%). No

apparent triggering event was noted in 21% of patients.

Chest radiograph was diagnostic in 69%; computed tomography was required in 31%. Esophagram, esophagoscopy, and bronchoscopy were performed on an individual basis and were invariably negative. When compared
with secondary pneumomediastinum, SPM is more
likely to be discovered by chest radiography, has a lower
incidence of pneumothorax and pleural effusion, requires a shorter hospital stay, and has no associated
mortality.
Conclusions. Spontaneous pneumomediastinum is a
benign condition that often presents with chest pain or
dyspnea. It can develop without a triggering event and
with no findings on chest radiography. Treatment is
expectant and recurrence is low. Secondary causes must
be ruled out to avoid an unfavorable outcome.

pulmonary parenchyma leads to alveolar rupture with

further leakage of air throughout the interstitium and
bronchovascular tissue sheath following a centripetal
pattern toward the mediastinum. In clinical practice,
SPM often develops as a result of various precipitating
events triggering a strong Valsalva maneuver, creating
the pressure differential needed for its genesis. In many
instances, however, it is difficult to determine the precise
source of the mediastinal air, and the diagnosis of SPM
remains one of exclusion.
Secondary pneumomediastinum comprises the bulk of
experience with pneumomediastinum. In the majority of
cases, it is initiated by blunt or penetrating trauma,
recent interventions in the esophageal or tracheobronchial tree, rupture of a hollow viscus, tissue dissection
originating from a spontaneous pneumothorax, or pulmonary or mediastinal infection by gas-forming organisms [3]. Spontaneous pneumomediastinum is often difficult to differentiate from more subtle causes of
secondary pneumomediastinum, such as contained
esophageal perforation, minor tears in the central tracheobronchial tree, and smoldering pulmonary and mediastinal infections. It is essential to confirm the absence
of a secondary pathologic event responsible for the
presence of mediastinal air.

neumomediastinum is defined as the presence of free

air in the mediastinum. Its presence is commonly
viewed as an ominous finding, with potentially catastrophic complications. Pneumomediastinum is further
divided into two groups of patients: spontaneous pneumomediastinum (SPM), without any obvious primary
source, and secondary pneumomediastinum, with a specific responsible pathologic event, such as trauma, intrathoracic infections, or violation of the aerodigestive
track.
Spontaneous pneumomediastinum was originally described by Louis Hamman in 1939 [1]; thus, the crepitus
heard with the heartbeat on chest auscultation is known
as the Hamman sign. Spontaneous pneumomediastinum
is generally described as a benign condition, presenting
in young adults exposed to a sudden pressure change
within the intrathoracic cavity. The pathogenesis of spontaneous pneumomediastinum was first postulated by
Macklin [2]. According to this report, a sudden increase
in intrathoracic pressure results in increased intraalveolar pressure. The pressure differential created within the
Accepted for publication April 21, 2008.
Address correspondence to Dr Garrett, 6029 Walnut Grove Rd, Suite 401,
Memphis, TN 38120; e-mail: egarrettmd@cvsclinic.com.

(Ann Thorac Surg 2008;86:962 6)

2008 by The Society of Thoracic Surgeons

2008 by The Society of Thoracic Surgeons

Published by Elsevier Inc

Downloaded from ats.ctsnetjournals.org by on April 16, 2012

0003-4975/08/$34.00
doi:10.1016/j.athoracsur.2008.04.067

Ann Thorac Surg

2008;86:962 6

CACERES ET AL
SPONTANEOUS PNEUMOMEDIASTINUM

Clinical Findings

Chest pain
Dyspnea
Cough
Odynophagia
Subcutaneous emphysema
Neck swelling
Pneumothorax

15
11
9
1
9
4
2

54%
39%
32%
4%
32%
14%
7%

these patients were excluded from the group of patients

evaluated with secondary pneumomediastinum.
The variables assessed in both cohorts of pneumomediastinum patients were age, sex, chest pain, respiratory
distress, presence of pneumothorax or pleural effusion
during hospitalization, sensitivity of chest roentgenogram to diagnose pneumomediastinum, presence of subcutaneous emphysema on physical examination, presence of subcutaneous emphysema on computed
tomography (CT) scan of the chest, hospital stay, and
mortality.

Statistical Analysis
The purpose of this study is to report our experience in
the management of 28 patients with SPM. Their clinical
presentation, diagnostic evaluation, radiologic findings,
and outcome are analyzed. A comparison with secondary
pneumomediastinum with regard to radiologic findings
and major outcome variables is conducted. An extensive
review of the literature is provided.

Categorical variables are expressed as percentages and

continuous variables are expressed as mean SD. Univariate analysis using a 2 test or Fishers exact test when
indicated was performed for categorical variables,
whereas Students t test was used when comparing
continuous variables. A p value of less than 0.05 was
considered statistically significant.

Results
Material and Methods
A retrospective comparative study of the cases of spontaneous pneumomediastinum diagnosed over a 12-year
period is presented. Patients with SPM and a cohort of
patients presenting with secondary pneumomediastinum during the same time frame were compared. Approval to conduct this retrospective study was granted by
the Baptist Memorial Hospital and University of Tennessee (UT) Institutional Review Boards, and individual
patient consent was waived. Data collection was performed following strict guidelines to protect patient
information.
Between January 1995 and June 2006, all patients with
an ICD code of mediastinal emphysema- pneumomediastinum were identified at Baptist Memorial Hospital
and the Regional Medical Center, both teaching institutions of the UT Health Science Center. All charts were
reviewed for demographic data, clinical presentation,
radiologic studies, diagnostic interventions, and outcome. Spontaneous pneumomediastinum was defined as
the radiologic confirmation of air within the tissue planes
of the mediastinum without any obvious underlying
etiology. If a pneumothorax was present, the patient was
included only in the absence of pulmonary pathology
explaining the initial event (eg, blebs, bullae, pulmonary
emphysema). A triggering factor was defined as the most
likely immediate event precipitating the SPM.
Spontaneous pneumomediastinum was excluded and
the pneumomediastinum considered secondary in all
cases of trauma-related admissions, recent aerodigestive
tract interventions, recent thoracic, cardiac, abdominal,
or cervical surgical interventions, cervical soft tissue or
intrathoracic infection, esophageal perforation, human
immunodeficiency virus disease with suspected Pneumocystis carinii pneumonia or recent endotracheal intubation. Pneumomediastinum was not routinely identified
and coded in patients with penetrating trauma; therefore,

A total of 74 patients were identified with mediastinal

emphysema or pneumomediastinum. According to the
inclusion and exclusion criteria previously defined, SPM
was identified in 28 patients. The average age was 27 17
years with a range from 3 to 71. Fifty-seven percent (16 of
28) were male.
The most frequently reported symptom was chest pain
in 54% (15 of 28), followed by dyspnea in 39% (11 of 28)
and cough in 32% (9 of 28). Relevant findings on physical
examination included subcutaneous emphysema in 32%
(9 of 28) and neck swelling in 14% (4 of 28). Pneumothorax was present in 7% of patients (2 of 28) upon admission. The pneumothorax in each patient was small and
evident only on the CT scan of the chest (Table 1). No
pathologic lung abnormality (eg, bleb, cavity, bullae) was
identified as the etiology of the pneumothorax. Laboratory analysis included complete cell count, electrolytes,
and arterial blood gases. The white blood cell count was
elevated in 39% of patients (11 of 28). The remainder of
the laboratory work performed was otherwise
unremarkable.
Radiologic images upon admission included chest
roentgengram in 93% of patients (26 of 28) and chest CT
scan in 71% (20 of 28). The chest roentgengram revealed
mediastinal air in 69% (18 of 26) and subcutaneous air in

Table 2. Predisposing Conditions in the Cohort of

Spontaneous Pneumomediastinum Patients
Predisposing Condition

Smoker
Asthma
Idiopathic pulmonary fibrosis
Chronic obstructive pulmonary disease
Inhaled drugs

8
4
2
1
0

29%
14%
7%
4%
0%

Downloaded from ats.ctsnetjournals.org by on April 16, 2012

GENERAL THORACIC

Table 1. Spontaneous Pneumomediastinum, Clinical

Findings on Presentation

963

964

CACERES ET AL
SPONTANEOUS PNEUMOMEDIASTINUM

Ann Thorac Surg

2008;86:962 6

SPM; this group of patients was considered to have

secondary pneumomediastinum. This second cohort included patients in whom pneumomediastinum developed as a result of blunt thoracic trauma in 39% (18 of 46),
barotrauma in 33% (15 of 46), P carinii pneumonia in
human immunodeficiency virus patients in 6% (3 of 46),
pneumothorax with pulmonary cavitary lesions in 6%
(3 of 46), esophageal perforation in 4% (2 of 46), meconium aspiration in the newborn in 4% (2 patients),
surgical intervention (tracheostomy in 2% [1 patient] and
thyroidectomy in 2% [1 patient]), and amiodaroneinduced lung injury in 2% (1 patient).
This group of 46 patients with secondary pneumomediastinum was compared with the original cohort of
patients with SPM. Patients with secondary pneumomediastinum were of older age (39 versus 27 years, p 0.05),
were less likely to be diagnosed with chest roentgengram
(47% versus 69%, p 0.05), more likely to have an
associated pneumothorax (56% versus 14%, p 0.001),
more likely to have a chest tube placed (46% versus 7%,
p 0.001), more likely to have an associated pleural
effusion (12% versus 0%, p 0.001), had a longer hospital
stay (19 versus 3 days, p 0.001), and were more likely to
die (39% versus 0%, p 0.001). In addition, patients with
secondary pneumomediastinum were found to have a
higher presence of clinical and radiologic subcutaneous
emphysema compared with SPM patients (45% versus
32%, p 0.05, and 64% versus 40%, p 0.05, respectively;
Table 4).

Table 3. Triggering Event in the Origin of Spontaneous

Pneumomediastinum

GENERAL THORACIC

Triggering Event

Emesis
Asthma
Cough
Physical activity
Choking
Defecation
Unknown
Inhaled drugs

10
6
2
1
1
1
6
0

36%
21%
7%
4%
4%
4%
21%
0%

35% (9 of 26). The CT scan of the chest revealed mediastinal air in 100% (20 of 20) and subcutaneous air in 40%
(8 of 20) of the patients assessed. Additional evaluation
was performed on an individual basis, including contrast
esophagram in 57% (16 of 28), esophagoscopy in 11% (3 of
28), and fiberoptic bronchoscopy in 4% (1 of 28). These
interventions were invariably negative.
Medical history predisposing to the development of
SPM included smoking in 29% patients (8 of 28), asthma
in 14% (4 of 28), idiopathic pulmonary fibrosis in 7% (2 of
28), and chronic obstructive pulmonary disease in 4%
(1 of 28). Inhalational drugs, although an established
precipitating event for SPM, were not found in any of our
patients (Table 2).
Among the suspected triggering factors in the origin of
SPM, emesis was noted to be the predominant initiating
event in 36% of patients (10 of 28). Asthma exacerbation
was seen in 21% (6 of 28) and cough in 7% (2 of 28).
Additional triggering events noted were physical activity,
defecation, and choking episodes. There was no apparent
triggering factor for mediastinal air in 21% of patients
(6 of 28; Table 3).
In the majority of cases, patients were admitted to the
hospital, placed on oxygen, and treated expectantly. In
many instances, oral intake was briefly withheld under the
suspicion of esophageal perforation. Average hospital stay
was 3 days (range, 0 to 11). There were no in-hospital
deaths. Follow-up (1 to 10 years) was obtained in 79% of
patients (22 of 28), and no recurrences were reported.
Among the patients with pneumomediastinum, 46
cases did not fulfill the criteria previously defined for

Comment
Pneumomediastinum is viewed by the medical community as an ominous sign with potentially devastating
complications. The differential diagnosis for chest pain,
dyspnea, or subcutaneous emphysema focuses on cardiac or pulmonary sources, including acute coronary
events, pericarditis, pulmonary embolism, pneumonia,
and pneumothorax. Esophageal perforation, spasm, and
reflux disease are also included in the differential diagnosis. Once these etiologic factors are excluded, a high
degree of suspicion is required to consider SPM as the
underlying condition.
Spontaneous pneumomediastinum is an unusual occurrence, with few reports in the literature [4 15]. Ac-

Table 4. Spontaneous Pneumomediastinum Compared With Secondary Pneumomediastinum

Male
Age, years
Diagnostic chest radiograph
Subcutaneous air on computed tomography
Associated pneumothorax
Chest tube placement
Associated pleural effusion
Hospital stay, days
Mortality

Spontaneous Pneumomediastinum

Secondary Pneumomediastinum

p Value

57%
27
69%
40%
14%
7%
0%
3
0%

68%
39
47%
64%
56%
46%
12%
19
39%

0.05
0.05
0.05
0.001
0.001
0.001
0.001
0.001

Downloaded from ats.ctsnetjournals.org by on April 16, 2012

CACERES ET AL
SPONTANEOUS PNEUMOMEDIASTINUM

cording to our data, chest pain, shortness of breath, and

subcutaneous emphysema are the most common presenting complaints. These findings are consistent with
prior reports in the literature [4, 6 11].
In our study, SPM presented with similar frequency
among men and women. This finding has also been
reported by Mondello and coworkers [5] in their review
of 18 cases and by Weissberg and Weissberg [7] in their
review of 22 cases. In other studies, however, this condition has presented mainly in men [4, 6, 10]. Our review
suggests that this condition presents predominantly in
healthy young adults, consistent with other published
reports [4 6].
According to previously published series, the presence
of a predisposing condition ranges widely, from 20% to
80% [4, 6, 16, 17]. In our study, only 39% of cases
presented with any of these risk factors. This may be
explained by the variability among evaluating physicians
in seeking and documenting predisposing conditions.
Most series describe the presence of a triggering event
before the development of SPM. Numerous immediate
preceding events have been noted, including forceful
emesis [4], intense coughing [7], inhalational drugs [8],
physical activity [18], intense screaming [11], spirometry
[19], childbirth [20], bronchospasm [11], and playing of
wind instruments [1]. There are variable data in the
literature regarding triggering events responsible for the
origin of the SPM; however, a compilation of reported
triggering events lists the most frequent ones as emesis
[4], intense physical activity [18], coughing [3, 5], and
intense screaming [7].
There is agreement among published reports that in a
significant number of patients, there is no evidence of a
triggering event responsible for SPM. Among the reports
listed in the bibliography, the presence of a precipitating
event ranges from 39% to 100%. After combining all the
series reported and adjusting for the sample size in each
study, 32% of the patients had SPM without having a
precipitating event identified. In our study, there was no
evidence of a precipitating event in 21% of the patients.
The diagnosis of SPM is revealed by radiographic
examination. A review of prior studies of SPM focuses
primarily on the clinical presentation without discussion
of the radiologic findings and no comparison with the
overall population of patients with pneumomediastinum.
The largest study was reported by Macia and associates
[4] and included 41 patients over a period of 16 years,
followed by a study by Campilo-Soto and colleagues [6]
who reported 36 patients. In both publications, the diagnosis was made by chest radiograph alone, possibly
overlooking patients identifiable only by CT. In our
series, only 69% of the cases were discovered by chest
radiograph, with the remaining 31% found on chest CT
scan. There is only one report in the literature describing
the correlation between chest radiograph and CT scan for
the diagnosis of SPM. Kaneki and coworkers [10] described 33 patients with SPM, but 30% of the cases of
SPM were missed with a plain chest x-ray film and were
diagnosed only by CT scan of the chest (Fig 1). Computed

965

GENERAL THORACIC

Ann Thorac Surg

2008;86:962 6

Fig 1. Spontaneous pneumomediastinum with extensive mediastinal

and subcutaneous emphysema, discovered by computed tomography
scan of the chest.

tomography has become the gold standard for diagnosing pneumomediastinum.

The lack of sensitivity of the plain chest x-ray film to
detect secondary pneumomediastinum is explained by
the severity of the pathology. In general, SPM is a benign
condition with preserved lung fields and mediastinal
structures. In cases of secondary pneumomediastinum,
as seen in blunt thoracic trauma and intrathoracic septic
processes, there is additional pathology involved that
may obscure otherwise distinct radiographic findings of
mediastinal air. In our series, in patients with secondary
pneumomediastinum, plain chest x-ray film was found to
have a higher incidence of associated pneumothorax and
pleural effusion and a lower sensitivity for the diagnosis.
In addition, subcutaneous air was more frequently seen
by CT scan in secondary pneumomediastinum as compared with SPM (64% versus 40%, respectively; p 0.05).
This comparative analysis has not been previously
reported.
Recurrence of spontaneous pneumomediastinum has
been rarely reported. The first case was described by
Yellin and colleagues [17] in 1983. Recurrence has been
reported in 5 additional patients [4, 9, 11, 14, 18]. No
recurrence was identified in our patients with 79% follow-up during a 1- to 10-year period.
In conclusion, SPM is a benign condition presenting
primarily in young adults, with an uneventful recovery
and unlikely recurrence. The clinical presentation in the
majority of cases involves chest pain, shortness of breath,
or subcutaneous emphysema. A triggering event with a
sudden increase in intrathoracic pressure is frequently
seen, and emesis is the most frequent inciting event.
Diagnosis requires a high level of suspicion as more than
30% of patients present without any precipitating factor,
and more than 30% of cases can be missed by a plain

Downloaded from ats.ctsnetjournals.org by on April 16, 2012

966

CACERES ET AL
SPONTANEOUS PNEUMOMEDIASTINUM

chest radiograph. Other sources of mediastinal air need

to be excluded, because secondary pneumomediastinum
may have an unfavorable outcome if not diagnosed
promptly.

The authors thank Pansy Adams, library technician, KCTCS

Library, for her helpful assistance in preparation of this
manuscript.

References
GENERAL THORACIC

1. Ito S, Takada Y, Tanaka A, Ozeki N, Yazaki Y. A case of

spontaneous pneumomediastinum in a trombonist. Kokyu
To Junkan 1989;37:1359-6.
2. Mecklin CC. Transport of air along sheaths of pulmonic
blood vessels from alveoli to mediastinum. Arch Int Med
1979;64:91326.
3. Newcomb AE, Clarke CP. Spontaneous pneumomediastinum: a benign curiosity or a significant problem? Chest
2005;128:3298 302.
4. Macia I, Moya J, Ramos R, et al. Spontaneous pneumomediastinum: 41 cases. Eur J Cardiothorac Surg 2007;31:1110 4.
5. Mondello B, Pavia R, Ruggeri P, Barone M, Barresi P,
Monaco M. Spontaneous pneumomediastinum: experience
in 18 adult patients. Lung 2007;185:9 14.
6. Campillo-Soto A, Coll-Salinas A, Soria-Aledo V, et al. [Spontaneous pneumomediastinum: descriptive study of our experience with 36 cases.] Arch Bronconeumol 2005;41:528 31.
7. Weissberg D, Weissberg D. Spontaneous mediastinal emphysema. Eur J Cardiothorac Surg 2004;26:885 8.
8. Koullias GJ, Korkolis DP, Wang XJ, Hammond GL. Current
assessment and management of spontaneous pneumomediastinum: experience in 24 adult patients. Eur J Cardiothorac
Surg 2004;25:8525.

Ann Thorac Surg

2008;86:962 6

9. Gerazounis M, Athanassiadi K, Kalantzi N, Moustardas M.

Spontaneous pneumomediastinum: a rare benign entity.
J Thorac Cardiovasc Surg 2003;126:774 6.
10. Kaneki T, Kubo K, Kawashima A, Koizumi T, Sekiguchi M,
Sone S. Spontaneous pneumomediastinum in 33 patients:
yield of chest computed tomography for the diagnosis of the
mild type. Respiration 2000;67:408 11.
11. Abolnik I, Lossos IS, Breuer R. Spontaneous pneumomediastinum. A report of 25 cases. Chest 1991;100:935.
12. Miura H, Taira O, Hiraguri S, Ohtani K, Kato H. Clinical
features of medical pneumomediastinum. Ann Thorac Cardiovasc Surg 2003;9:188 91.
13. Halperin AK, Deichmann RE. Spontaneous pneumomediastinum: a report of 10 cases and review of the literature. NC
Med J 1985;46:213.
14. Shindo Y, Kurumada T, Ohta U, et al. A recurrent case of
spontaneous mediastinal emphysema. J Exp Clin Med 1995;
20:17.
15. Jougon JB, Ballester M, Delcambre F, MacBride T, Dromer
CE, Velly JF. Assessment of spontaneous pneumomediastinum: experience with 12 patients. Ann Thorac Surg 2003;75:
1711 4.
16. Panacek EA, Singer AJ, Sherman BW, Prescott A, Rutherford
WF. Spontaneous pneumomediastinum: clinical and natural
history. Ann Emerg Med 1992;21:12227.
17. Yellin A, Gapany-Gapanavicius M, Lieberman Y. Spontaneous pneumomediastinum: is it a rare cause of chest pain?
Thorax 1983;38:3835.
18. Mihos P, Potaris K, Gakidis I, Mazaris E, Sarras E, Kontos Z.
Sports-related spontaneous pneumomediastinum. Ann Thorac Surg 2004;78:983 6.
19. Nemet D, Suchard JR, DiBernardo LM, Mukai DS, Cooper
DM. Pneumomediastinum and subcutaneous emphysema
after pulmonary function tests in a young healthy woman.
Eur J Emerg Med 2004;11:1057.
20. Praski A, Kupis W, Florczak M, Szpetnar M. Mediastinal and
subcutaneous emphysema as a complication of labor.
Ginekol Pol 2007;78:329 32.

Downloaded from ats.ctsnetjournals.org by on April 16, 2012

Spontaneous Pneumomediastinum: A Comparative Study and Review of the

Literature
Manuel Caceres, Syed Z. Ali, Rebecca Braud, Darryl Weiman and H. Edward Garrett,
Jr
Ann Thorac Surg 2008;86:962-966
DOI: 10.1016/j.athoracsur.2008.04.067
Updated Information
& Services

including high-resolution figures, can be found at:

http://ats.ctsnetjournals.org/cgi/content/full/86/3/962

References

This article cites 19 articles, 7 of which you can access for free at:
http://ats.ctsnetjournals.org/cgi/content/full/86/3/962#BIBL

Citations

This article has been cited by 12 HighWire-hosted articles:

http://ats.ctsnetjournals.org/cgi/content/full/86/3/962#otherarticles

Subspecialty Collections

This article, along with others on similar topics, appears in the

following collection(s):
Lung - other
http://ats.ctsnetjournals.org/cgi/collection/lung_other

Permissions & Licensing

Requests about reproducing this article in parts (figures, tables) or

in its entirety should be submitted to:
http://www.us.elsevierhealth.com/Licensing/permissions.jsp or
email: healthpermissions@elsevier.com.

Reprints

For information about ordering reprints, please email:

reprints@elsevier.com

Downloaded from ats.ctsnetjournals.org by on April 16, 2012