Int. J. Oral Maxillofac. Surg.

2009; 38: 346–349

doi:10.1016/j.ijom.2008.12.007, available online at http://www.sciencedirect.com

Clinical Paper
Oral Medicine

Retention of lipiodol after J. Schortinghuis1,2, J. Pijpe1,

F. K. L. Spijkervet1, A. Vissink1
1
Department of Oral and Maxillofacial

parotid gland sialography Surgery, University of Groningen and

University Medical Center Groningen,
Groningen, The Netherlands; 2Department of
Oral and Maxillofacial Surgery, Scheper
Hospital Emmen, Emmen, The Netherlands
J. Schortinghuis, J. Pijpe, F. K. L. Spijkervet, A. Vissink: Retention of lipiodol after
parotid gland sialography. Int. J. Oral Maxillofac. Surg. 2009; 38: 346–349. # 2008
International Association of Oral and Maxillofacial Surgeons. Published by Elsevier
Ltd. All rights reserved.

Abstract. There is limited information about the retention of lipiodol in the parotid
gland after parotid gland sialography. This study assesses the prevalence of lipiodol
retention after parotid sialography and determines if retention of lipiodol is related
to the sialography technique or the underlying salivary gland pathology. Using the
electronic hospital database (1996–2006), 66 out of 565 patients were identified
who had additional maxillofacial radiographic examinations after the initial
sialography. Additional radiographs up to October 2007 were included; these were
orthopantomographic radiographs in all cases. In 28 patients (42%) signs of lipiodol
retention were observed (mean radiographic follow-up: 15  13 months). Retention
was characterized by small radiopaque spots in the periphery of the gland. Lipiodol
retention was predominantly associated with a fausse route (n = 8) or the presence
of salivary gland disease (sialectasia; n = 17). In 9 patients with signs of lipiodol
retention, a series of radiographs was available. Lipiodol radiodensities decreased in
Keywords: contrast sialography; parotid gland;
size during 28 months, and could disappear gradually (follow-up 14–57 months). lipiodol; complications.
Despite the high frequency of retention of small depots of lipiodol for years after
sialography in patients subjected to additional radiographic examinations, no Accepted for publication 9 December 2008
clinically adverse effects were observed. Available online 20 January 2009

Sialography allows visualization of the
ductal architecture of the salivary gland
radiographically. A small amount of con-
trast fluid is infused into the main duct; it
can be oil- or water-based. Lipiodol is a
commonly use oil-based contrast medium;
it is an iodized poppy seed oil. SICARD and
FORESTIER17 first used lipiodol for contrast
myelography by introducing it into the
subdural space. In the 1950s, a water-
based contrast fluid was introduced as
an alternative for sialography procedures7.
There is controversy whether to use oil- or
water-based contrast fluid; both are in use.
Oil-based contrast fluid seems to render
sharper images than water-based alterna-
tives4,8,10. One argument in favour of
water-based contrast media is that oil-
based media are poorly eliminated from
the body. There are case reports of oil-
based contrast media being present a long
time after sialography2,12. We found 2
cases of lipiodol retention in their depart-
ment and wondered how often this
occurred. The purpose of this study was
to assess the prevalence of lipiodol reten-
tion in the parotid gland after sialography,
and to assess whether there was any rela-
tionship between lipiodol retention and the
sialography technique and/or salivary
gland function and disease.
Materials and methods
From 1 January 1996 until 31 December
2006, all patients who underwent a sialo-
gram of the parotid gland were identified
using the electronic hospital diagnostic
database. Patients were included when
they had been subjected to additional
radiographic examinations in the follow-
ing months or years, to assess possible

0901-5027/040346 + 04 $36.00/0 # 2008 International Association of Oral and Maxillofacial Surgeons. Published by Elsevier Ltd. All rights reserved.

retention of lipiodol (radiographs taken up
to 31 October 2007 were assessed). Useful
additional radiographs were considered
orthopantomograms, parotid radiographs,
and lateral or anteroposterior skull radio-
graphs. These radiographs were usually
taken for reasons other than salivary gland
disease, such as the evaluation of impacted
teeth or for the purpose of implantology.
The sialography technique involved can-
nulation of the parotid main duct, followed
by retrograde infusion of lipiodol under
low manual pressure using a 2-ml Corn-
wall syringe. The patient’s sensation of a
sudden increase of preauricular pressure is
used to estimate filling of the gland, after
which an anteroposterior and a lateral
radiograph are taken. Premature leakage
of contrast fluid is prevented through main
duct ligation under local anesthesia.
A database was set up to assess the
incidence of lipiodol retention and to
assess any relationship with sialography
technique or underlying salivary gland
disease. The following items were scored:
age, sex, date of sialogram, amount of
lipiodol (ml) inserted, complications dur-
ing sialography (fausse route; an iatro-
genic rupture of Stensen’s duct during
sialography), presence of salivary gland
abnormality on sialogram (sialectasia),
parotid gland function (ml/min stimulated
salivary flow rate), date of additional
radiographs, signs and location of lipiodol
retention, additional parotid disease, addi-
tional systemic disease, new complaints
after sialography.
Lipiodol retention was defined as punc-
tate, globular or cavitary radiodensities
located in the region of the parotid gland
parenchyma after contrast sialography or
as an irregular radiodensity in the anterior
region of Stenson’s duct in the case of a
fausse route. Care was taken to ensure that
the location and pattern of the radioden-
sities found at follow-up matched the
location and pattern of the radiodensity
of the initial sialogram. To gain an impres-
sion of the extend of lipiodol retention, it
was scored whether the retention consisted
of more or less than 10 radiopaque ‘spots’
on the radiograph. Confluent spots were
counted as ‘one’. Statistical analysis was
Table 1. Retention of lipiodol after parotid gland sialography.
Number of sialograms
Amount of contrast medium (ml), mean  SD
Radiographic follow-up (months), mean  SD (median)
Normal sialography
Fausse route*
Sialectasia**
Pearson’s x2 *p < 0.01 and **p < 0.04 (retention versus no retention).
Lipiodol retention after sialography
performed using Pearson’s x2 test with a
0.05 significance level.
Results
556 patients, with suspected salivary
gland disease, were identified who under-
went sialography of the parotid gland. Of
these patients, 66 had one or more addi-
tional orthopantomographic radiographs
taken in the subsequent years, and fulfilled
the study inclusion criteria. In 28 of these
66 patients, signs of lipiodol retention in
the parotid gland could be observed
(Table 1). In 8/28 cases a fausse route
had occurred during the sialographic pro-
cedure (p < 0.01), and in 17/28 cases the
sialogram showed signs of salivary gland
disease (p < 0.04). Of these 17 patients,
13 were diagnosed with Sjögren’s syn-
drome, the remaining patients experienced
a banal parotitis. Retention of lipiodol was
generally visible as a small cloud of round
radiopaque spots, about 0.5–1 mm in dia-
meter, located in the periphery of the
gland (Figs. 1 and 2). In 10/28 cases fewer
than 10 spots were visible, in the remain-
ing 18 cases there were more than 10
spots. In the case of a fausse route, reten-
tion was observed as an irregular ‘bleb’ on
the radiograph (Fig. 3). Retention of lipio-
dol was not related to the amount inserted
or to decreased parotid gland function
(mean stimulated parotid gland salivary
flow rate was 0.15 ml/min/gland ( 0.18)
at baseline). There was no relationship
with age or sex. In 9/28 patients with
lipiodol retention, a series of orthopanto-
mograms was available for follow-up. In
4/9 patients the first post-sialogram ortho-
pantomogram (average follow-up 14
months, range 9–21 months) showed signs
of retention; these signs had disappeared at
a subsequent orthopantomogram (average
follow-up 57 months, range 26–100
months). In the other 5 patients the lipio-
dol radiodensities found at the first ortho-
panthomogram had decreased in size
during an average follow-up of 28 months
(range 7–65 months), but had not disap-
peared. Among these 9 cases, there was
only one case with a fausse route.
Retention (n = 28) No retention (n = 38)
28
0.75  0.11
15  13 (9) 33  28 (32)
3 22
8 2
17
347
Fig. 1. Lipiodol retention 2 years after initial
sialography of the left parotid gland. Lipiodol
retention is represented by the punctate radio-
densities in the periphery of the parotid gland.
There were no clinical signs of salivary gland
disease. The patient suffered from a myalgia
of the masseter muscle.
Discussion
This study indicates that in 42% of the
parotid sialographic imaging cases
assessed, retention of small amounts of
lipiodol seemed to have occurred. The
amount of lipiodol retention was minor
and was mainly observed as small radio-
paque ‘spots’ in the periphery of the gland.
The high frequency of retention was sur-
prising, since it is commonly assumed that
lipiodol is completely ‘washed out’ by the
flow of saliva within hours after the sialo-
graphic procedure. No similar study is
available in the literature with which to
compare these results.
One reason explaining the high fre-
quency of lipiodol retention may be a
selection bias in the patients studied.
Patients with salivary or other pathology
may be overrepresented because they may
be prone to having additional future radio-
graphs. The retrospective design of this
Total (n = 66)
38 66
0.72  0.11 0.73  0.11
24  24 (16)
25
10
14 31
348 Schortinghuis et al.
Fig. 2. Lipiodol retention 1 month after the initial sialography of the right parotid gland in a
patient diagnosed with Sjögrens’ syndrome.
Fig. 3. (A) Lipiodol retention 8 months after a fausse route during sialography of the right
parotid gland. (B) The same patient, 49 months after sialography. The amount of lipiodol has
decreased. The radiographs were taken for the purpose of a mandibular advancement procedure.
study and the selection bias are limitations
that indicate the need for a more controlled
study.
Lipiodol retention was observed in
cases with or without a fausse route. In
the case of a fausse route, the syringe is
displaced through the main duct into the
soft tissue. As a consequence, lipiodol is
inadvertently placed inside the soft tissue
outside the ductal parenchyma, located in
the region of the main duct. This lipiodol
cannot be washed out by the salivary
secretion and has to be eliminated by
resorption. In the 10 fausse routes encoun-
tered, there was retention of contrast
media in eight. In one patient with a fausse
route, additional radiographs were avail-
able and it appeared that in this case the
radiodensity decreased in size over 49
months (Fig. 3). This high occurrence of
fausse routes might be attributed to the
fact that the sialography procedure was
performed by both residents and more
experienced surgeons. This is in contrast
to the study by KALK et al., in which all the
sialograms were made by the same clin-
ician, a resident oral and maxillofacial
surgeon with substantial experience in
sialography5. Lipiodol retention was also
observed in cases of a normal sialography
procedure without a fausse route. How
retention occurs in these cases is not clear.
When finer ductules are filled, oil-based
media are poorly eliminated and may
cause ductile obstruction3. In a normal
gland, lipiodol may be pushed through
the acinar structures into the soft tissue
surrounding the gland14. In rabbit experi-
ments, overfilling of the submandibular
gland with lipiodol led to parenchymal
destruction, resulting in vacuoles filled
with contrast medium surrounded by
macrophages18. This may be the case in
the presence of parotid gland disease
where the gland parenchyma may have
become fragile. This phenomenon may
be related to the present results: in 63%
of the cases in which retention was
observed there were indications of gland
destruction. Patients diagnosed with Sjög-
ren’s syndrome are characterized by
decreased salivary gland function, and it
might be more difficult for these patients
to ‘wash out’ the lipiodol.
In the literature, there is limited infor-
mation about the clinical significance of
lipiodol retention. Two cases could be
traced in which a fausse route during
parotid sialography resulted in lipiodol
retention that was detected 4 years2 and
70 months12 after the initial sialography
procedure. These patients had no com-
plaints. Another case involved a fausse
route into the soft tissue that led to the
formation of lipogranulomas. Because of
swelling and inflammation, the mass had
to be surgically removed15. MACAN et al.
recently reported a case where lipiodol
leaked from an iatrogenic perforation of
Stensen’s duct, which constituted a for-
eign body in the cheek9. In the present
series, no adverse responses were
observed after the sialography procedure
with lipiodol, not even in the cases of a
fausse route. These data were extracted
from the hospital electronic database,
however, and the patients visited the hos-
pital for reasons other than complaints
after sialography. It is not clear whether
these patients were questioned about any
possible complaints after the initial sialo-
graphy procedure. Only patients with
additional radiographic images were
included, so patients who underwent sia-
lography without additional imaging were
not evaluated.
Although a longer follow-up seems
necessary, as complaints might occur
many years after imaging, the authors
observed that lipiodol slowly disappears
from the gland and related structures dur-
ing the following years. The clinical sig-
nificance of lipiodol retention seems
small, but a more controlled study should
be undertaken to confirm this.
There is controversy whether oil-based
contrast media, such as lipiodol, or water-
based contrast media should be used for
sialography. Oil-based contrast fluid gives
sharper images because of its hydrophobic

nature4, but it is not readily absorbed in
cases of overfilling or a fausse route,
unlike water-based contrast fluid14. PET-
TINI et al. compared the use of oil- and
water-based media in dogs and concluded
that oil-based lipiodol is an irritant, espe-
cially if it passes beyond the salivary
excretory tree, but not to such a degree
that it should not be used.13
In the authors’ department, lipiodol is
used for its excellent visualisation of the
ductile system of the gland14 and its low
morbidity5. Although retention of small
amounts of lipiodol seems to occur fre-
quently, the authors think that this should
not limit its use because no morbidity has
been associated with this retention. One
study even showed a reduction of pain days
after lipiodol sialography, which might
indicate an extra benefit of the procedure1.
An important disadvantage of sialogra-
phy is that it involves exposure to ionizing
radiation and infusion of iodized contrast
fluid into the salivary gland. Another
drawback is that the correct interpretation
of the images may require a certain level
of expertise6. Magnetic resonance ima-
ging (MRI) and sonography are consid-
ered biologically harmless. Some patients
are not suitable for MRI (they have metal
implants or are claustrophobic). Sonogra-
phy has no known contraindications; it is a
quick and generally well-accepted proce-
dure. Disadvantages of sonography are the
limited tissue penetration depth and the
requirement for an experienced observer
to interpret the sonograms11. Scintigraphy
can be used for measuring gland function,
for example in the diagnosis of Sjogren’s
syndrome, however, it does not visualise
the ductal system as the other modalities
do16.
In conclusion, retention of minor
amounts of lipiodol seemed to occur in
up to 42% of sialography procedures of the
parotid gland in patients who had addi-
tional radiographs in the following
months. It is mainly associated with a
fausse route or underlying salivary gland
disease. Retention of lipiodol did not
result in clinically adverse effects, but
more controlled studies and longer fol-
low-up is necessary. Although oil-based
media provide superior image quality,
water-based media are an alternative in
less experienced hands to minimize the
Lipiodol retention after sialography
risk of an iatrogenic ‘fausse route’ and
subsequent lipiodol retention.
Competing interests
None declared.
Funding
None.
Ethical approval
Not required.
References
1. Drage NA, Brown JE, Wilson RF. Pain
and swelling after sialography: is it a
significant problem? Oral Surg Oral
Med Oral Pathol Oral Radiol Endod
2000: 90: 385–388.
2. El-Hadary A, Ruprecht A. Long-term
retention of contrast medium in sialogra-
phy: a case report. Dentomaxillofac
Radiol 1986: 15: 41–44.
3. Feinberg SE. The diagnosis and surgi-
cal management of salivary gland dis-
orders. In: Peterson LJ, et al., eds:
Principles of Oral and Maxillofacial
Surgery, 2. Philadelphia: Lippincott
Company 1992 : 841–879.
4. Holtgrave E, Elke M, Lüthy H. Sia-
lography with Telebrix 38. A compara-
tive study with Lipiodol-UF in the
diagnosis of salivary gland diseases. Den-
tomaxillofac Radiol 1973: 2: 68–72.
5. Kalk WWI, Vissink A, Spijkervet
FKL, Möller JM, Roodenburg JLN.
Morbidity from parotid sialography. Oral
Surg Oral Med Oral Pathol Oral Radiol
Endod 2001: 92: 572–575.
6. Kalk WW, Vissink A, Spijkervet FK,
Bootsma H, Kallenberg CG, Rooden-
burg JL. Parotid sialography for diagnos-
ing Sjögren syndrome. Oral Surg Oral
Med Oral Pathol Oral Radiol Endod
2002: 94: 131–137.
7. Katzberg RW. Urography into the 21 st
century: new contrast media, renal hand-
ling, imaging characteristics, and nephro-
toxicity. Radiology 1997: 204: 297–312.
8. Luyk NH, Doyle T, Ferguson MM.
Recent trends in imaging the salivary
glands. Dentomaxillofac Radiol 1991:
20: 3–10.
9. Macan D, Hat J, Lukšić. Lipiodol ultra-
fluid – foreign body in the cheek. Dento-
maxillofac Radiol 2008: 37: 171–174.
349
10. Nicholson DA. Contrast media in sialo-
graphy: a comparison of Lipiodol Ultra
Fluid and Urografin 290. Clin Radiol
1990: 42: 423–426.
11. Niemelä RK, Takalo R, Pääkkö E,
Suramo I, Päivänsalo M, Salo T,
Hakala M. Ultrasonography of salivary
glands in primary Sjogren’s syndrome. A
comparison with magnetic resonance
imaging and magnetic resonance sialo-
graphy of parotid glands. Rheumatology
2004: 43: 875–879.
12. Ozdemir D, Polat NT, Polat S. Lipio-
dol UF retention in dental sialography. Br
J Radiol 2004: 77: 1040–1041.
13. Pettini PL, Laforgia PD. Reactions of
the submandibular salivary gland caused
by sialography. Experimental research in
dogs. Minerva Stomatol 1977: 26: 175–
184.
14. Qwarnström E. Experimental sialo-
graphy. The effects of retrograde infu-
sion of radiographic contrast media on
salivary gland morphology and func-
tion. A review article. Oral Surg Oral
Med Oral Pathol Oral Endod 1986: 62:
668–682.
15. Yanagisawa S, Shimizu M,
Fukuyama Y, Takahashi Y. Parotid
and pterygomaxillary lipogranuloma
caused by oil-based contrast medium
used for sialography: report of a case. J
Oral Maxillofacial Surg 1996: 54: 350–
353.
16. Saito T, Fukuda H, Horikawa M,
Ohmori K, Shindoh M, Amemiya A.
Salivary gland scintigraphy with
99mTc-pertechnetate in Sjögren’s syn-
drome: relationship to clinicopathologic
features of salivary and lacrimal glands. J
Oral Pathol Med 1997: 26: 46–50.
17. Sicard MM. Forestier. Méthode générale
d’exploration radiologique par l’huile
iodée (lipiodol). Bull Soc Méd Hóp Paris
1922: 46: 463–468.
18. Verhoeven JW. Choice of contrast med-
ium in sialography. Oral Surg 1984: 57:
323–337.
Address:
Jurjen Schortinghuis
Department of Oral and Maxillofacial
Surgery
University Medical Center Groningen
(UMCG)
PO Box 30.001
9700 RB Groningen
The Netherlands
Tel: +31 50 3612567
Fax: +31 50 3611136.
E-mail: j.schortinghuis@kchir.umcg.nl