Vol. 120 No.

6 December 2015

Cone beam computed tomography (CBCT) sialographydan

adjunct to salivary gland ultrasonography in the evaluation
of recurrent salivary gland swelling
Tobias Kroll, MD,a Andreas May, DDS,b Claus Wittekindt, MD, PhD,a Christopher Kähling, MD,b
Shachi Jenny Sharma, MD,a Hans-Peter Howaldt, MD, DDS, PhD,b Jens Peter Klussmann, MD, PhD,a and
Philipp Streckbein, MD, DDSb

Objective. Cone beam computed tomography (CBCT) sialography could help improve the visualization of the ductal system
of salivary glands. The aim of this retrospective investigation was to monitor the use of CBCT sialography for the diagnosis of
pathologies within the intraglandular ductal system when ultrasonography was inconclusive.
Study Design. Fourteen consecutive patients suffering from recurrent swelling of a major salivary gland were evaluated. In 12
patients (8 female; 4 male; average age 46 years), a radiopaque contrast agent could be injected into the ductal system,
followed by a routine CBCT. Four blinded examiners evaluated the acquired data sets retrospectively.
Results. CBCT revealed seven stenosis, two salivary stones, one complete duct atresia, one intraglandular duct ectasia, and
one regular duct system. Three of the detected pathologies were strictly intraglandular.
Conclusions. CBCT sialography shows promise as a supplementary noninvasive diagnostic tool for the visualization of the
intraglandular ductal system of the major human salivary glands. Controlled studies to further validate this method should be
undertaken. (Oral Surg Oral Med Oral Pathol Oral Radiol 2015;120:771-775)

Various radiologic and nonradiologic methods are
available for the imaging diagnostics of salivary gland
parenchyma and salivary ducts. Currently, because of
its known advantages, ultrasonography is the key
nonradiologic modality used for imaging the paren-
chyma of major salivary glands.1 However, salivary
ducts can only be demonstrated when they are filled.
This occurs either when they are obstructed through a
salivary stone or a stenosis, or when a salivation-
inducing substance, such as ascorbic acid, has been
orally administered.2 Injection of an ultrasound contrast
agent into the ductal system can improve visualization
of the gland parenchyma and the ductal system during
examination.3,4
A further, established, nonradiologic imaging
method for the visualization of salivary ducts is mini-
mally invasive sialendoscopy, which, since 2006, is
increasingly being performed as a routine procedure in
Otorhinolaryngology in Germany.5
Radiologic imaging modalities available to visualize
salivary gland parenchyma and salivary gland ducts are
conventional sialography, computed tomography (CT),
and magnetic resonance imaging (MRI) sialography.6
With the introduction and increasing use of cone beam
computed tomography (CBCT) in otorhinolaryngology
and cranio-maxillo-facial surgery, a new radiologic
imaging method appears to have become available for
the visualization of salivary glands, particularly sali-
vary gland ducts, after injection of a radiopaque
contrast agent into the ductal system. This approach,
called CBCT sialography, is, however, still in an early
stage of development.7,8
The aim of this retrospective review is to demonstrate
that CBCT sialography is a suitable imaging modality
for the diagnosis of salivary duct pathologies, in
particular, when sonography has failed to demonstrate
pathology in patients suffering from recurrent swelling
of a salivary gland.
MATERIAL AND METHODS
Indication for CBCT sialography remained uncertain
because of recurrent gland swellings after the initial
clinical examination. Contraindications were detected
due to recurrent gland swellings, such as stones, duct
lesions, or intraglandular lesions. All of the patients
(n ¼ 92) who presented from February 2013 to
February 2014 at our institution with recurrent swelling

a
Department of Otorhinolaryngology, University Hospital Gießen,
Justus Liebig University, Klinikstraße 33, Giessen, 35392, Germany.
b
Statement of Clinical Relevance
Department of Cranio-Maxillo-Facial Surgery, University Hospital
Gießen, Justus Liebig University, Klinikstraße 33, Giessen, 35392,
Cone beam computed tomography sialography al-
Germany.
Received for publication May 13, 2014; returned for revision Feb 14, lows for visualization of ductal anomalies even in
2015; accepted for publication Sep 2, 2015. peripheral locations. It shows promise as a supple-
Ó 2015 Elsevier Inc. All rights reserved. mentary noninvasive diagnostic tool for the visuali-
2212-4403/$ - see front matter zation of the intraglandular ductal system.
http://dx.doi.org/10.1016/j.oooo.2015.09.005

771
ORAL AND MAXILLOFACIAL RADIOLOGY
772 Kroll et al.
of a parotid or submandibular gland were scanned for
this retrospective evaluation. After a comprehensive
anamnesis and clinical examination, diagnostic ultra-
sonography of the swollen salivary gland was per-
formed routinely in all cases. In 14 patients, the cause
of the recurrent salivary gland swelling could not be
established in the initial examination or sialendoscopy.
CBCT sialography was indicated for further imaging of
the salivary duct system. Each patient signed an
informed consent form before routine CBCT
sialography.
In preparation for the CBCT sialography, the efferent
duct of the swollen gland was dilated to 22 gauges with
salivary duct probes (Marchal Salivary Duct Probes,
Karl Storz, Tuttlingen, Germany). Then, depending on
the gland, the caruncle or papilla was cannulated with
an indwelling venous cannula (Vasofix Braunüle,
Braun, Melsungen, Germany), through which a radi-
opaque contrast agent (Ultravist-370, Bayer Vital,
Leverkusen, Germany) was injected into the ductal
system of the swollen gland until the patient reported
fullness of the gland or the contrast agent flowed past
the probe into the oral cavity. Before scanning, patients
were positioned such that they did not have to be
repositioned after injection of the contrast agent. A
radiologic image was then acquired with a CBCT
scanner (ProMax 3-D Max, Planmeca, Helsinki,
Finland). The following operational parameters were
used to acquire the scan: For the parotid gland, a field of
view with a width of 230 mm, a height of 160 mm, and
a resolution of 0.4 mm was chosen. The resultant tube
voltage was 96 kV, with a tube current of 12 mA. The
data acquisition time for this scan was 9 seconds. For
the submandibular gland, a field of view with a width of
130 mm, a height of 55 mm, and a resolution of 0.2 mm
was chosen. The tube voltage for this scan was 96 kV,
with a tube current of 12 mA, and a data acquisition
time of 12 seconds.
The acquired CBCT data sets were visualized as
multiplanar and three-dimensional reconstructions (3-D
recons) and analyzed by four independent blinded ex-
aminers qualified to perform radiologic evaluations.
Data analysis was performed with DICOM viewer
software (OsiriX MD, v2.8.5 64 bit, Pixmeo SARL,
Bern, Switzerland). For every patient’s images, visual-
ization of the ductal system was individually optimized,
followed by an evaluation for certain pathologies, such
as stones, duct stenosis, duct ectasia, or duct atresia.
RESULTS
Fourteen consecutive patients with indication for sia-
lography were included in the retrospective review. Due
to impenetrable salivary ducts, CBCT could not be
performed in two cases. Of the remaining 12 patients,
eight (66.7%) were female, and four (33.3%) were
OOOO
December 2015
male. The average patient age was 46 years. The parotid
gland was examined in eight (66.7%) and the sub-
mandibular gland in four (33.3%) cases. Five (41.7%)
patients underwent sialendoscopy before CBCT sia-
lography. In these cases, the indication for CBCT sia-
lography was unsolved pathology within the
sialendoscopy.
Evaluation of the acquired CBCT images (Table I)
revealed seven (58.3%) salivary duct stenoses, of
which three (42.9%) were found close behind the
papilla, even though the papilla had showed no signs
of obstruction in prior clinical examinations or
ultrasonography; three (42.9%) were in the main
efferent duct, and one (14.3%) was in an
intraglandular duct (Figure 1). In addition, in two
(16.7%) submandibular glands, salivary stones that
had remained undetected during ultrasonography
could be demonstrated as the cause of the recurrent
swelling of the gland. One stone was found in an
intraglandular duct (50%), and the other was found in
the main efferent duct (50%). Complete atresia of the
Stensen duct was found in one (8.3%) parotid gland,
and intraglandular duct ectasia was found in one
(8.3%) other parotid gland (Figure 2). For one (8.3%)
gland, no correlating pathology could be found with
CBCT sialography (Figure 3).
On the basis of the results of CBCT sialography,
three (25.0%) patients were indicated for a subsequent
sialendoscopy, or revision sialendoscopy. One (8.3%)
patient was treated with a combined approach. On the
basis of an individual concept, a wait-and-watch strat-
egy was chosen for eight (66.7%) patients.
DISCUSSION
Visualizing the ductal system of the major salivary
glands has always been a challenge for clinicians,
although various imaging modalities are available.
Currently, the four most frequently used imaging mo-
dalities in everyday clinical routine are ultrasonogra-
phy, sialography, CT, and MRI. Compared with the
first two methods, the two latter methods play a sub-
ordinate role due to cost, possible radiation exposure,
and availability issues.9 As an additional imaging
method, minimally invasive sialendoscopy is being
increasingly used.5
Currently, ultrasonography is the method of choice
for imaging salivary glands because it can be performed
by the attending physician, does not incur a radiation
burden, and provides an adequate imaging accuracy for
diagnosis in many cases.1 However, to be adequately
visible during ultrasonography, the salivary ductal
system has to be filled. Salivary ducts can either be
intentionally filled by orally administering ascorbic
acid to induce flow of saliva or may already be filled
as the result of obstructive pathologies, such as
OOOO ORIGINAL ARTICLE
Volume 120, Number 6 Kroll et al. 773

Table I. Patient list with patient history, diagnosis, and procedure after cone beam computed tomography (CBCT)
sialography
Duration Diagnosis Procedure
of symptoms after CBCT after CBCT
Gland Side Symptoms (anamnesis) Comorbidities sialography sialography
1 Parotid gland Right Recurrent swellings 36 months None Duct stenoses Combined approach
with acute parotits (sialendoscopy and
surgery)
2 Parotid gland Left Recurrent swellings 15 months Metabolic Duct atresia Sialendoscopy
syndrome
3 Parotid gland Right Recurrent swellings 18 months Arterial Duct stenoses Wait-and-watch
hypertension at papilla
4 Parotid gland Right Recurrent swellings 3 months None Duct stenoses Sialendoscopy
at papilla
5 Parotid gland Left Recurrent swellings 1 month Adiposity Intraglandular Wait-and-watch
with acute parotits duct ectasia
6 Parotid gland Right Recurrent swellings 4 months Arterial Regular salivary Wait-and-watch
hypertension duct
7 Parotid gland Left Recurrent swellings 60 months None Intraglandular Wait-and-watch
duct stenoses
8 Submandibular left Recurrent swellings 30 months None Sialolithiasis Sialendoscopy
gland
9 Submandibular Left Recurrent swellings 5 months None Intraglandular Wait-and-watch
gland sialolithiasis
10 Submandibular Left Recurrent swellings 6 months None Duct stenoses Wait-and-watch
gland
11 Submandibular Right Recurrent swellings 18 months None Duct stenoses Wait-and-watch
gland
12 Parotid gland Right Recurrent swellings 2 months Arterial Duct stenoses Wait-and-watch
hypertension at papilla

salivary stones or duct stenosis.2 The retrograde
administration of a contrast agent, similar to the
procedure for sialography, can aid in the visualization
of the ductal system during ultrasonography.3,4 A lim-
itation of ultrasonography in demonstrating the cause of
recurrent salivary gland swelling is that pathologies
other than obstruction by stones can be difficult to di-
agnose with this method.10
Sialendoscopy has been established as an additional
procedure for the diagnosis and therapy of salivary
gland disease. Koch et al. preserved over 97% of sali-
vary glands with stenoses of the Wharton ducts or
Stensen ducts by using this method along with a com-
bined approach.11,12 However, the method can only be
used for extraglandular duct pathologies or those
located close to the hilum.

Fig. 1. A, Three-dimensional reconstruction of the efferent duct system of the parotid gland. B, Axial section in a planar
reconstruction. The images show a stenosis located behind the junction to the tertiary division branch of the duct system. The
stenosis is indicated by a red arrow.
ORAL AND MAXILLOFACIAL RADIOLOGY OOOO
774 Kroll et al. December 2015

Fig. 2. Intraglandular ectasia of the efferent parotid gland duct system. Ectasia can be seen to begin at the junction point between
the primary and secondary duct branches and to continue into the secondary branch. A, Three-dimensional reconstruction of the
efferent duct system. B, Axial section in a planar reconstruction. In both images, the red arrows indicate the ectatic portion of the
duct.

Although sialography, which was initially described
in 1902, still holds a place in salivary gland diagnostics,
this method is increasingly being replaced by modern
methods, such as CBCT.13-15 CBCT has become an
established diagnostic tool in cranio-maxillo-facial
surgery and is rapidly gaining entry into otorhinolar-
yngology. This imaging method is being used for the
diagnostics of the bony structures of the paranasal si-
nuses and petrous bone.
Early studies have shown that CBCT is also a suit-
able modality for the visualization of salivary ducts that
have been enhanced with a radiopaque contrast agent.7-10
In a study that directly compared conventional two-
dimensional sialography with three-dimensional sia-
lography, Jadu et al. postulated that CBCT sialography
outperformed conventional sialography for the visuali-
zation of salivary gland parenchyma and salivary
stones.8 Our findings support the results reported by
Varoquaux et al. that the salivary gland duct system
can be visualized up to the sixth branch with CBCT
sialography after a contrast agent has been injected.10
This high imaging accuracy is due to the isotropic
voxel resolution afforded by the CBCT unit.8 In a
first comparative study on radiation dose, Jadu et al.
found that by choosing appropriate parameters, the
effective radiation dose from CBCT sialography
equaled that of conventional sialography.9
In the present retrospective review, the three-
dimensional visualization of the salivary duct system
through CBCT allowed us to diagnose a total of three
previously undetected intraglandular pathologies. Our
findings thus support Varoquaux et al.’s suggestion that
the underlying causes for recurrent swelling of a sali-
vary gland, not due to obstruction through a stone, can

Fig. 3. No correlating pathology could be found, and the reasons for recurrent swellings remain uncertain. A normal parotid gland
duct system is shown as a three-dimensional reconstruction (A), and as a planar reconstruction (B).
OOOO
Volume 120, Number 6
elude ultrasonographic diagnosis. It remains to be
clarified if ultrasonography can accurately image
intraglandular duct pathologies after contrast medium
has been injected into the salivary duct system.
Comparative studies need to be undertaken to shed light
on this question.
Although it must be stressed that CBCT sialography
cannot be a substitute for sialendoscopy, the targeted
use of CBCT sialography could, however, reduce the
indication for sialendoscopy for pathologies of the
extraglandular duct system or the area around the hi-
lum. In these cases, CBCT sialography could be helpful
in accurately identifying patients who could profit from
further gland-preserving interventions.
CONCLUSIONS
In this retrospective review, CBCT sialography was
helpful in diagnosing unexplained cases of swelling of
one of the major salivary glands with intraglandular
pathologies when ultrasound was inconclusive. We
could thus demonstrate that CBCT sialography shows
promise as a supplementary noninvasive diagnostic tool
for the visualization of the intraglandular duct system.
Our encouraging findings suggest that performing a
CBCT sialography before sialendoscopy may be justi-
fied in cases of unexplained, recurrent swelling of a
salivary gland to rule out pathologies that are intra-
glandular and, therefore, unreachable by the sialendo-
scope. Patients could, thus, be spared the unpleasantness
of an unsuccessful intervention.
An advantage of CBCT over other radiologic imag-
ing, such as CT and MRI, is its lower cost. The amount
of exposure to radiation is less than with CT and is
comparable with conventional sialography.
Controlled studies with higher case numbers and
follow-up examinations are planned to further verify the
value of CBCT examinations for the diagnosis of sali-
vary gland pathologies.
REFERENCES
1. Onkar PM, Ratnaparkhi C, Mitra K. High-frequency ultrasound in
parotid gland disease. Ultrasound Quart. 2013;29:313-321.
2. Bozzato A, Hertel V, Bumm K, Iro H, Zenk J. Salivary simulation
with ascorbic acid enhances sonographic diagnosis of obstructive
sialadenitis. J Clin Ultrasound. 2009;37:329-332.
ORIGINAL ARTICLE
Kroll et al. 775
3. Kroll T, Helbig M, Klussmann JP, Wittekindt C. Intraductal
application of Levovist(R) in salivary glands of animals. Lar-
yngorhinootologie. 2012;91:229-232 [in German].
4. Zengel P, Berghaus A, Weiler C, Reiser M, Clevert DA. Intra-
ductally applied contrast-enhanced ultrasound (IA-CEUS) for
evaluating obstructive disease and secretory dysfunction of the
salivary glands. Eur Radiol. 2011;21:1339-1348.
5. Kroll T, Finkensieper M, Hauk H, Guntinas-Lichius O,
Wittekindt C. Sialendoscopydlearning curve and nation-wide
survey in German ENT-departments. Laryngorhinootologie.
2012;91:561-565 [in German].
6. Kraff O, Theysohn JM, Maderwald S, et al. High-resolution MRI
of the human parotid gland and duct at 7 Tesla. Invest Radiol.
2009;44:518-524.
7. Abdel-Wahed N, Amer ME, Abo-Taleb NS. Assessment of the
role of cone beam computed sialography in diagnosing salivary
gland lesions. Imaging Sci Dent. 2013;43:17-23.
8. Jadu FM, Lam EW. A comparative study of the diagnostic ca-
pabilities of 2 D plain radiograph and 3 D cone beam CT sia-
lography. Dentomaxillofac Radiol. 2013;42:20110319.
9. Jadu F, Yaffe MJ, Lam EW. A comparative study of the effective
radiation doses from cone beam computed tomography and plain
radiography for sialography. Dentomaxillofac Radiol. 2010;39:
257-263.
10. Varoquaux A, Larribe M, Chossegros C, Cassagneau P, Salles F,
Moulin G. Cone beam 3 D sialography: preliminary study. Rev
Stomatol Chir Maxillofac. 2011;112:293-299 [in French].
11. Koch M, Iro H, Kunzel J, Psychogios G, Bozzato A, Zenk J.
Diagnosis and gland-preserving minimally invasive therapy for
Wharton’s duct stenoses. Laryngoscope. 2012;122:552-558.
12. Koch M, Kunzel J, Iro H, Psychogios G, Zenk J. Long-term re-
sults and subjective outcome after gland-preserving treatment in
parotid duct stenosis. Laryngoscope. 2014;124:1813-1818.
13. Benson B. Salivary gland radiology. In: White SC, Pharoah MJ,
eds. Oral Radiology: Priciples and Interpretation. 6th ed. St.
Louis, MO: Mosby; 2009:578-598.
14. Rzymska-Grala I, Stopa Z, Grala B, et al. Salivary gland calcu-
lidcontemporary methods of imaging. Polish J Radiol. 2010;75:
25-37.
15. Shahidi S, Hamedani S. The feasibility of cone beam computed
tomographic sialography in the diagnosis of space-occupying
lesions: report of 3 cases. Oral Surg Oral Med Oral Pathol
Oral Radiol. 2014;117:e452-e457.
Reprint requests:
Tobias Kroll, MD
Department of Otorhinolaryngology
University Hospital Gießen
Justus Liebig University
Klinikstraße 33 D-35392 Gießen
Germany
tobiaskroll75@gmail.com