RELATIONSHIP BETWEEN RADIOGRAPHIC EVIDENCE OF

TRACHEOBRONCHIAL LYMPH NODE ENLARGEMENT AND DEFINITIVE

OR PRESUMPTIVE DIAGNOSIS

BRIAN G. JONES, RACHEL E. POLLARD

Tracheobronchial lymphadenomegaly is commonly associated with lymphosarcoma and disseminated mycotic

infection. Available data also suggest other pathologic causes of enlarged tracheobronchial lymph nodes. Our
objective was to establish the distribution and prevalence of diseases that cause radiographically evident tra-
cheobronchial lymphadenomegaly in a large population of dogs. Patients were divided into groups based on
the methods of diagnoses with 25 having confirmed diagnoses and 85 with presumptive diagnoses. Of the 110
dogs in the study, 92 (84%) had neoplasia and 18 (16%) had infectious diseases. Infections were attributed to
Coccidioides (12, 67%), Aspergillus (3, 17%), and 1 each (6%) to Nocardia, Penicillium, and Mycobacteriosis.
Tumors were characterized as lymphoma (66, 60%) or nonlymphoid (26, 23.6%). Nonlymphomas in Group
1 included histiocytic sarcoma complex (16%), carcinoma (12%), adenocarcinoma (8%), osteosarcoma (8%),
chemodectoma (4%), ganglioneuroblastoma (4%), and neuroendocrine (4%). The number of dogs with tracheo-
bronchial lymphadenomegaly scores 1, 2, 3, 4, or 5 (with 5 being the greatest) was 8 (7%), 15 (14%), 30 (27%),
15 (14%), and 44 (38%), respectively. The results suggest that in addition to diagnoses of lymphoma and fungal
infections, other neoplasms and in particular histiocytic sarcoma and metastatic adenocarcinoma, should be
considered when tracheobronchial lymphadenomegaly is identified radiographically in dogs. When comparing
the degree of tracheobronchial lymphadenopathy by disease category, there was no significant affiliation (P =
0.33). 
C 2012 Veterinary Radiology & Ultrasound.

Key words: dog, hilar, lymphadenomegaly, radiography, tracheobronchial.

Introduction metastatic neoplasia.2–5 Lymphomatoid graunlomatosis,

eosinophilic bronchopneumopathy, Lagenidium infection

I N THE DOG, there are three consistent tracheobronchial

lymph nodes surrounding the tracheal bifurcation. The
right and left nodes are located laterally and a larger cen-
tral V-shaped node conforms to the origin of the primary
bronchi.1 The tracheobronchial lymph nodes receive ef-
ferent lymphatics from adjacent pulmonary lymph nodes,
lungs, bronchi, thoracic portions of the aorta, esopha-
gus, trachea, heart, mediastinum, and diaphragm.1 Ra-
diographically, tracheobronchial lymphadenomegaly has
been associated with ventral deviation of the trachea
and compression of the carina on lateral projections. On
dorsoventral and ventrodorsal projections, tracheo-
bronchial lymphadenomegaly causes lateral displacement
of the caudal pulmonary bronchi. Differential diagnoses for
tracheobronchial lymphadenomegaly include fungal gran-
ulomas, bacterial granulomas, multicentric neoplasia or
From the School of Veterinary Medicine, University of California,
Davis (Jones) and the Department of Surgical and Radiological Sciences,
School of Veterinary Medicine, University of California, Davis, CA 95616
(Pollard).
Address correspondence and reprint requests to Brian G. Jones, at
the above address. E-mail: jones.4084@osu.edu
Received August 18, 2011; accepted for publication December 15,
2011.
doi: 10.1111/j.1740-8261.2011.01921.x
and Mycobacterium avium infection have also been associ-
ated with tracheobronchial lymphadenomegaly in dogs.6–9
Accurate antemortem tissue sampling for diagnosis is dif-
ficult due to the anatomic location of the nodes.
We are unaware of information relating the distribution
of disease types and tracheobronchial lymph node enlarge-
ment. Our purpose was to establish the distribution and
prevalence of diseases associated with radiographically evi-
dent tracheobronchial lymphadenomegaly in a large popu-
lation of dogs. We hypothesized that dogs with lymphosar-
coma and systemic fungal infection would comprise the
majority of patients. Further, we speculated that dogs
with lymphosarcoma and fungal infections would have
more profound tracheobronchial lymph node enlargement,
whereas dogs with other causes of tracheobronchial lymph
node enlargement would have less pronounced lymph node
enlargement.
Methods
Electronic medical record archives from January 2000 to
November 2009 were reviewed for patients with evidence
Vet Radiol Ultrasound, Vol. 00, No. 0, 2012, pp 1–6.

1
2 JONES AND POLLARD 2012

TABLE 1. Subjective Grading Scheme Applied to Thoracic Radiographs

for the Assessment of Tracheobronchial Lymph Node Enlargement (In
order for a higher grade to be assigned, all criteria for the lower grades
must also have been met

Grade Criteria
1 Increased opacity seen in the perihilar region on the lateral projection.
2 Ventral deviation of the carina on the lateral projection.
3 Increased opacity between and widening of the bronchi on the
DV projection.
4 An obvious mass in the perihilar region on all projections.
5 Compression of the mainstem bronchi on all projections.

of tracheobronchial or hilar lymphadenopathy, based on

a 2- or 3-view thoracic radiographic study. One hundred
and ten dogs with a definitive or presumptive diagnosis as
to the cause of hilar lymphadenomegaly and two or three
view thoracic radiographic studies with a report indicating FIG. 1. A right lateral thoracic radiograph of a dog with grade 1 tra-
cheobronchial lymph node enlargement. There is increased opacity in the
that there was an evidence of hilar lymphadenomegaly were perihilar region (arrows) without a discrete mass. The final diagnosis was
identified. Dogs were subsequently placed into one of two B-cell lymphosarcoma.
groups, depending on whether a definitive or presumptive
diagnosis was known.
Dogs in Group 1 had a definitive histologic or cyto-
logic diagnosis based on necropsy (23/110; 21%) or biopsy
(2/110; 2%) of the tracheobronchial lymph nodes. Dogs
in Group 2 had a presumptive diagnosis as to the cause of
their tracheobronchial lymph node enlargement. Dogs were
included in Group 2 based on one of the following meth-
ods of diagnosis: peripheral lymph node or parenchymal
organ biopsy diagnosis of lymphosarcoma (60/110; 55%)
or histiocytic sarcoma complex (3/110; 3%), positive fun-
gal titer for Coccidioides immitus (11/110; 10%), thoracic
mass (lung or mediastinal) biopsy diagnosis of neoplasia
(10/110; 9%), or fungal infection (1/110; 1%).
There were 6 of 110 (6%) intact females, 48 of 110 (44%)
neutered females, 15 of 110 (14%) intact males, and 41 of
110 (37%) neutered males. There were 13 of 110 (12%) dogs
≤10 kg, 32 of 110 (29%) dogs were between 10.1 and 25 kg,
and 63 of 110 (57%) dogs were >25 kg. The weight was not
FIG. 2. A right lateral thoracic radiograph of a dog with grade 2 tracheo-
recorded for 2 of 110 (2%) dogs. A variety of breeds were bronchial lymph node enlargement. In addition to increased opacity in the
represented. perihilar region (arrows) there is ventral displacement of the carina (arrow
heads). The final diagnosis was systemic Aspergillus terreus.
Thoracic radiographs were reviewed by a board certified
radiologist (REP), who was unaware of the clinical diagno-
sis. If patients had multiple radiographic studies, only the
phadenomegaly scores and disease category. A P-value of
first study with radiographic evidence of tracheobronchial
<0.05 was considered significant.
lymphadenomegaly was reviewed. The degree of tracheo-
bronchial lymphadenomegaly was scored based on an esca-
lating grading scheme (Table 1 and Figs. 1–5). The presence
Results
of pleural effusion, pulmonary nodules or masses, and me-
diastinal masses, including cranial mediastinal and sternal In Group 1, 19/25 (76%) were diagnosed with neopla-
lymph node enlargement, was also recorded. sia, 5/25 (20%) with fungal infections, and 1/25 (4%) with
All data were summarized and variables compared for Mycobacteriosis (lymph node grade 4). Of the 25 dogs, 3
their association with degree of tracheobronchial lym- (25%) were diagnosed with lymphosarcoma (lymph node
phadenomegaly. A Fischer’s exact t-test was performed grades 1, 5 and 5) and 16 (64%) with nonlymphoid tumors
to evaluate the correlation between tracheobronchial lym- (mean lymph node grade 3.8; range 2–5). Of the dogs with
VOL. 00, NO. 0 RELATIONSHIP BETWEEN RADIOGRAPHIC EVIDENCE AND DEFINITIVE DIAGNOSIS
FIG. 3. (A) A right lateral thoracic radiograph of a dog with grade 3
tracheobronchial lymph node enlargement. There is increased opacity in the
perihilar region and ventral deviation of the carina. (B) In a dorsoventral
radiograph there is widening of the mainstem bronchi (double arrow). The
final diagnosis was B-cell lymphosarcoma.
lymphoma, 3 of 3 (100%) had a mediastinal mass, 0 of 3 had
pulmonary masses, and 1 of 3 (33%) had pleural effusion.
Nonlymphoid neoplasms in Group 1 dogs included 4
of 25 (16%; mean lymph node grade 5) histiocytic sarco-
mas complex, 3 of 25 (12%; lymph node grades 1, 2, and 5)
metastatic pulmonary carcinomas, 2 of 25 (8%; lymph node
grades 3 and 3) metatstatic pancreatic adenocarcinomas, 2
of 25 (8%; mean lymph node grades 2 and 4) metastatic
3
FIG. 4. (A) A right lateral thoracic radiograph of a dog with grade 4
tracheobronchial lymph node enlargement. There is an obvious mass visible
in the perihilar region (arrows). (B) Dorsoventral radiograph. There is a mass
between the mainstem bronchi (double arrow). The final diagnosis was B-cell
lymphosarcoma.
osteosarcomas, and 1 of 25 (4%) each of metastatic anal
sac carcinoma (lymph node grade 5), hepatocellular carci-
noma (lymph node grade 5), neuroendocrine tumor (lymph
node grade 4), chemodectoma (lymph node grade 5), and
ganglioneuroblastoma (lymph node grade 5). Of the dogs
with histiocytic sarcoma, 3 of 4 (80%) had at least one pul-
monary mass, 2 of 4 (50%) had a mediastinal mass, and 0 of
4 had pleural effusion. Of the dogs with metastatic tumors,
8 of 16 (50%) had at least one pulmonary mass, 2 of 16
(13%) had a mediastinal mass, and 1 of 16 (6%) had pleural
effusion.
4 JONES AND POLLARD
FIG. 5. (A) A right lateral thoracic radiograph of a dog with grade 5
tracheobronchial lymph node enlargement. There is an obvious mass in the
perihilar region in addition to multiple pulmonary nodules. (B) Dorsoventral
radiograph. There is a mass between the mainstem bronchi with compression
of the left caudal lobar bronchus (arrows). The final diagnosis was metastatic
hemangiosarcoma.
Fungal infections in Group 1 dogs included 3 of 25 (12%)
Aspergillus spp. (lymph node grades 2, 4 and 5) and 1 each
(4%) of Coccidioides immitus (lymph node grade 5) and
Nocardia asiatica (lymph node grade 5). Of the dogs with
fungal infections, 0 of 5 had at least one pulmonary mass,
3 of 5 (60%) had a mediastinal mass, and 1 of 5 (20%) had
pleural effusion.
Of the 85 dogs in Group 2, 73 (86%) were diagnosed
with neoplasia and 12 (14%) with fungal infections. Of the
2012
TABLE 2. Frequency of Tracheobronchial Lymph Node Score by Disease
Category, Including Dogs from Groups 1 and 2 (n = 110)
Lymph Node Score Disease
Lymphoma Nonlymphoid Fungal Mycobacteriosis
Neoplasia Disease
1 7 1
2 9 4 2
3 22 5 3
4 7 3 4 1
5 20 14 8
85 dogs, 63/85 (74%) had a presumptive diagnosis of lym-
phosarcoma (mean lymph node grade 3.3; range 1–5) and
10/85 (12%) with nonlymphoid tumors (mean lymph node
grade 4.0; range 2–5). Of the dogs with lymphoma, 36 of 63
(57%) had a mediastinal mass, 3 of 63 (5%) had pulmonary
masses, and 12 of 63 (19%) had pleural effusion.
Nonlymphoid neoplasms in Group 2 dogs included 7/85
(8%) histiocytic sarcoma complex (mean lymph node grade
4.4; range 3–5), 2/85 (2%) metastatic carcinomas of un-
known origin (lymph node grades 2 and 3), and 1/85 (<1%)
metastatic pulmonary carcinoma (lymph node grade 4). Of
the dogs with histiocytic sarcoma complex, 5 of 7 (71%)
had at least one pulmonary mass, 2 of 7 (29%) had a medi-
astinal mass, and 1of 7 (14%) had pleural effusion. Of the
dogs with metastatic tumors, 3 of 3 (100%) had at least one
pulmonary mass, 0 of 3 had a mediastinal mass, and 1 of 3
(33%) had pleural effusion.
Fungal infections in Group 2 dogs included 11 of 85
(13%) C. immitus (mean lymph node grade 4.0; range 2–5)
and 1 of 85 (<1%) Penicillium katagense (lymph node grade
4). Of the dogs with fungal infections, 2 of 12 (17%) had at
least one pulmonary mass, 1 of 12 (8%) had a mediastinal
mass, and 0 of 12 had pleural effusion.
Considering both Group 1 and Group 2 dogs, the distri-
bution of tracheobronchial lymph node scores 1, 2, 3, 4, or
5 was 8 of 110 (7%), 15 of 110 (14%), 30 of 110 (27%), 15
of 110 (14%), and 42 of 110 (38%), respectively (Table 2).
Given the subtle abnormalities affiliated with a score of “1,”
the presence or absence of a heart murmur was retrieved
from the medical records for dogs in this category. Only
1 of 8 dogs that received a score of “1” had a heart mur-
mur. When comparing the prevalence of tracheobronchial
lymph node score by disease type, there was no significant
affiliation (P = 0.33).
Discussion
The majority of dogs in this retrospective study had
definitive or presumptive diagnoses of lymphoma (66/110)
or mycotic disease (17/110). Of the 17 dogs with fungal in-
fections, 12 (71%) were attributed to C. immitus. Although
these findings are consistent with our hypothesis that lym-
phoma and fungal disease would constitute the majority of
VOL. 00, NO. 0 RELATIONSHIP BETWEEN RADIOGRAPHIC EVIDENCE AND DEFINITIVE DIAGNOSIS
dogs with tracheobronchial lymphadenomegaly, it should
be noted that the number of dogs with nonlymphoid origin
neoplastic causes of tracheobronchial lymph node enlarge-
ment was not inconsequential (24%). This suggests that
nonlymphoid neoplasms should not be excluded when con-
sidering differentials for radiographically evident tracheo-
bronchial lymph node enlargement. Moreover, the size of
the lymph nodes does not appear to be predictive of the
type of underlying disease. Because the majority of dogs in
this series received tracheobronchial lymph node scores of
3 or higher, this may be due to the advanced stage of disease
with which many veterinary patients present.
In people and more recently in dogs, computed tomog-
raphy has been shown more accurate than radiographs
when assessing tracheobronchial and hilar lymph node
metastasis.10–12 It is plausible that the increased sensitivity
of more advanced imaging techniques such as computed
tomography might show an association between degree of
hilar lymphadenopathy and disease prevalence. However,
mild lymphadenomegaly can be challenging to identify even
with the benefit of computed tomography.
Nonlymphoid origin neoplasms included histiocytic sar-
coma complex, pulmonary carcinoma, pancreatic ade-
nocarcinoma, hepatocellular carcinoma, osteosarcoma,
anal sac carcinoma, chemodectoma, ganglioneuroblas-
toma, and neuroendocrine neoplasia. Most of the dogs
with these tumor types had other thoracic disease such as
pulmonary masses, pleural effusion, or mediastinal masses.
These findings are consistent with a previous report, where
there was frequent involvement of the lungs and thoracic
lymph nodes in association with malignant histiocytosis in
the Bernese Mountain Dog.13
In one patient, mycobateria was identified on ante-
mortem tracheobronchial lymph node aspirate, which was
later confirmed on postmortem. Mycobacteria enter the
body through respiratory or alimentary routes, multiply in
phagocytic cells and spread via lymphatic or hematogenous
dissemination.14 Canine mycobacteriosis has most fre-
quently been associated with Mycobacterium tuberculosis
and Mycobacterium bovis, although disseminated M. avium
has been reported.7 Although mycobacteriosis may repre-
sent a rare cause of tracheobronchial lymphadenomegaly, it
should be considered when systemic infection is suspected
or neoplastic and fungal disease has been sufficiently ruled
out.
One major limitation inherent in this retrospective study
stems from the inaccessible nature of the tracheobronchial
REFERENCES
1. Evans HE. Miller’s anatomy of the dog, 3rd ed. Philadelphia: W.B.
Saunders, 1993.
5
lymph nodes, making routine needle aspiration difficult.
The majority (85/110) of the dogs had a presumptive
diagnoses of hilar lymphadenopathy. In creating the in-
clusion criteria for Group 2, we attempted to incorpo-
rate diseases where a diagnosis, although presumptive, was
highly likely to be the cause of tracheobronchial lymph
node enlargement. Considering that 60% (66/110) of the
dogs were diagnosed with lymphoma, it is not surprising
that peripheral lymph node cytology was the most com-
mon method of diagnosis. However, the exclusion of these
dogs from the study would falsely skew the results to imply
that neoplasms other than lymphoma were the most likely
cause of tracheobronchial lymph node enlargement seen
radiographically.
It should also be noted that our population only in-
cluded dogs from California and regions known to be en-
demic for Coccidioides.15 Therefore, fungal infections due
to C. immitus may have been overrepresented in comparison
to other geographic areas. Moreover, the incidence of tra-
cheobronchial lymph node enlargement due to other fungal
organisms would likely be substantially different in areas
where Blastomycosis and Histoplasmosis are endemic.
One final consideration involves the inclusion of dogs re-
ceiving a tracheobronchial lymph node enlargement score
of “1.” Given the complex anatomic features of the perihi-
lar region, increased opacity in that area without tracheal
deviation may be attributable to tracheobronchial lymph
node enlargement, left atrial enlargement, or less likely a
pulmonary lesion located on midline. Only one dog that
received a score of “1” had a heart murmur which makes
the presence of left atrial enlargement unlikely in the ma-
jority of dogs in this category. However, it is important to
note that other causes of increased perihilar opacity could
mimic mild tracheobronchial lymph node enlargement.
In conclusion, based on our results, we suggest that
in addition to diagnoses of lymphoma and fungal infec-
tions, other nonlymphoid origin neoplasms, and in partic-
ular malignant histiocytosis complex and metastatic ade-
nocarcinoma, should be considered when tracheobronchial
lymphadenomegaly is identified radiographically. Further,
we failed to detect a significant relationship between de-
gree of tracheobronchial lymphadenopathy and disease
category.
ACKNOWLEDGMENT
The authors sincerely thank Stephanie Venn-Watson of Epitracker (San
Diego, CA) for her statistical contributions.
2. Suter PF. Thoracic radiography. A text atlas of thoracic diseases of
the dog and cat. Wettswil: Peter F. Suter, 1984;277–284.
6 JONES AND POLLARD
3. King LJ, Padley SPG, Wotherspoon AC, Nicholson AG. Pulmonary
MALT lymphoma: imaging findings in 24 cases. Eur Radiol 2000;20:1932–
1938.
4. Lee KS, Kim Y, Prmack SL. Imaging of pulmonary lymphomas. Am
J Roetgenol 1997;168:339–345.
5. Geyer NE, et al. Radiographic appearance of confirmed pulmonary
lymphoma in cats and dogs. Vet Radiol Ultrasound 2010;51:386–390.
6. Berry CR, et al. Pulmonary lymphomatoid granulomatosis in seven
dogs (1976–1987). J Vet Intern Med 1990;4:157–166.
7. Shackelford CC, Reed WM. Disseminated Mycobacterium avium in-
fection in a dog. J Vet Diagn Invest 1989;1:273–275.
8. Walter PA, et al. Radiographic appearance of pulmonary metastases
from transitional cell carcinoma of the bladder and urethra of the dog. J Am
Vet Med Assoc 1984;185:411–418.
9. Grooters AM, et al. Clinicopathologic findings associated with Lageni-
dium sp. infection in 6 dogs: initial description of an emerging oomycosis. J
Vet Intern Med 2003;17:637–646.
2012
10. Ballenger EA, Adams WM, Dubielzig RR, et al. Computed tomog-
raphy characteristics of canine tracheobronchial lymph node metastasis. Vet
Radiol Ultrasound 2010;51:397–403.
11. Nambu A, Kato S, Saito A, Araki T. Appearance of mediastinal and
pulmonary hilar lymph nodes on thin-section CT: comparison with 5mm
slice thickness CT. Clin Imaging 2007;31:375–378.
12. Mancuso AA, Harnsberger HR, Muraki AS, Stevens MH. Computed
tomography of cervical and retropharyngeal lymph nodes: normal anatomy,
variants of normal and applications in staging head and neck cancer. Radi-
ology 1983;148:709–714.
13. Moore PF, Rosin A. Malignant histiocytosis of Bernese mountain
dogs. Vet Pathol 1986;23:1–10.
14. Liu SK, Weitzmann I, Johnson GG. Canine tuberculosis. J Am Vet
Med Assoc 1980;177:164–167.
15. Shubitz LF, Butkiewicz CD, Dial SM, et al. Incidence of Coccidioides
infection among dogs residing in a region in which the organism is endemic.
J Am Vet Med Assoc 2005;226:1846–1850.