Kamulegeya1 The Journal of Contemporary Dental Practice, Volume 5,
No. 3, August 15, 2004
Neoplasms of the Salivary Glands: A Descriptive Retrospective Study
of 142
Cases-Mulago Hospital Uganda
Objective: This study was aimed at establishing the prevalence,
sex, age, and site distribution of salivary gland neoplasms as seen
in the Department of Oral Surgery Mulago Hospital Uganda.
Design: Descriptive retrospective study.
Materials and Methods: Histopathological results from the
Department of Oral Surgery were retrieved and analyzed for
histological diagnosis, age, sex, and site. Records retrieved
included those from January 1988 to December 2000. Student’s t-test
and z-test were used for data analysis.
Results: Of all neoplasms seen in the Department between 1988 and
2000, salivary gland tumors com- prised 7.8% of which 77 (54.23%)
were benign and 65 (45.77%) were malignant. Pleomorphic adenoma was
the most prevalent over all. Among the malignant group,
adenocarcinoma was the most prevalent of those where the palate was
the most affected site. The age range was 1-80 years with the
overall mean age being 33.83 ± 18. 83. The mean age for the benign
lesions was 26.74 ± 16.0 and for the malignant lesions was 42.49 ±
19.15.
Conclusion: The findings were in agreement with most African
studies but differed from studies from the other continents.
Pleomorphic adenoma was the most prevalent salivary gland neoplasm
with both males and females affected equally. More research is
needed on specific salivary gland neoplasms in Uganda and
preferentially on a regional basis geographically. However, if
valid data on the diagnosis as well as age, site, and sex
distribution are to be obtained, an improvement in the
standardization of diagnosis and reporting of patient data will be
necessary.
Keywords: Salivary gland neoplasms, pleomorphic adenoma,
adenocarcinoma, Uganda
Citation: Kasangaki A, Kamulegeya A. Neoplasms of the Salivary
Glands: A Descriptive Retrospective Study of 142 Cases-Mulago
Hospital Uganda. J Contemp Dent Pract 2004
August;(5)3:016-027.
Abstract
© Seer Publishing
2 The Journal of Contemporary Dental Practice, Volume 5, No. 3,
August 15, 2004
Introduction Different diseases affect- ing the salivary glands
affect the efficiency and normal function of the glands. The
diseases range from infections, inflammation, obstruction,
functional disorders, and neoplasms. Although in the general
popula- tion, the prevalence of salivary gland neoplasms is
relatively low in relation to other salivary gland diseases, they
do constitute a significant proportion of oral tumors after
squamous cell carcinoma.1 A study from Nigeria reports salivary
gland neoplasms as 2.8% of all head and neck tumors.2 In Kenya they
are among the most common oral tumors.3 While in Uganda, they
ranked next to sarcomas, adamantinomas (ameloblastoma), and
carcinomas.4 Despite the fact that globally benign salivary gland
tumors predominate over malignant ones5-10, there is a variation in
the prevalence of the specific salivary gland neoplasms from study
to study worldwide. In the state of Oklahoma in the USA5, Jamaica6,
and Japan11, muco-epidermoid carcinoma is the most prevalent
malignant neoplasm, while in Britain12 adenoid cystic carcinoma
predominates. In South Africa10 polymorphous low-grade adeno-
carcinoma predominates. The most afflicted sites by these tumors
also vary from study to study. For example, pleomorphic adenoma is
the most common tumor in the parotid gland. Seventy five percent of
these tumors are present in the parotid gland as compared to 39.8%
in the palate.13 The situation in Uganda is not yet clearly known
since we could only find one article4 on this subject. Sing et al.
reports of a palatal location but the findings are about tumors of
the jaws which may have included pleomorphic adenoma of the minor
salivary glands of the palate.
The sex distribution of these neoplasms is also reported
differently by different authors. A report from Brazil7 reports a
slight female preponder- ance of minor salivary gland neoplasms,
while the Jordanian study reports a slight male preponder- ance.8
Cawson1 reports salivary gland carcinoma preponderance for females,
and Aritoba14 reports several sex preponderancies of different
salivary gland neoplasms in relation to their location. The age
distribution of patients with salivary neo- plasms
differ.10,14,15,16
The purpose of this study was to establish the dif- ferent biopsed
and histopathologically confirmed salivary gland neoplasms seen at
the Department of Oral Surgery Mulago Hospital, their age, sex
distribution, and site for a period between January 1988 to
December 2000.
Materials and Methods A total of 142 salivary gland neoplasm
reports were retrieved from the histopathological results files of
the Department of Oral Surgery Mulago Hospital Kampala Uganda
stretching from the year 1988-2000.
Clinical information including, age, sex, and site from which the
biopsy was taken were recorded. The data was compiled manually onto
a master sheet from which tallying was done. Both the t-test and
z-test were used for data analysis.
We also went on to ask the pathologists for the criteria they use
in the diagnosis of salivary gland tumors. A summary is provided in
Table 1.
Results The distribution of the various salivary gland neoplasms is
presented in Table 2. They com- prised 7.5% of all oral neoplasms.
From the Department’s biopsy results file, (77) 54.23% of the
lesions were benign and (65) 45.77% were malignant.
Of all the benign lesions, the most frequent was pleomorphic
adenoma constituting 77.9%, fol- lowed by salivary gland adenoma
18.18%, and then oxyphil cell adenoma at 3.9%. Among the malignant
salivary gland neoplasms, adenocarci- noma was the most prevalent
at 36.92%, followed by carcinoma at 27.69%, then muco-epidermoid
carcinoma 9.23%, mucus secreting carcinoma constituted 6.15%,
acinic cell carcinoma 4.62%, and carcinoma ex pleomorphic adenoma
3.08%. Of the 142 cases, 64 (45.07%) were male, 76 (53.52% were
female, and in 2 cases (1.41%) sex was not specified. The
male:female ratio was 1: 1.18. Forty-three (55.84%) of the benign
neo- plasms cases were female; while 33 (50.77%) of malignant cases
were female and 2 (3.08%) of the malignant cases were unspecified.
The gen- der distribution of these salivary gland neoplasms is
shown in Table 2, and the age distribution of these cases is shown
in Table 3. The mean age
3 The Journal of Contemporary Dental Practice, Volume 5, No. 3,
August 15, 2004
Table 1. Summary of diagnostic criteria.
4 The Journal of Contemporary Dental Practice, Volume 5, No. 3,
August 15, 2004
Table 1 (cont.). Summary of diagnostic criteria.
Table 2. The different salivary gland neoplasms; distribution by
sex, diagnosis and working abbreviation.
5 The Journal of Contemporary Dental Practice, Volume 5, No. 3,
August 15, 2004
for this patient population was 33.83 ± 18.83 with an age range of
1-80 years. The mean age of the benign neoplasms was 26.74 ± 16.0,
while that of the malignant neoplasms was 42.49 ± 9.15. Most cases
were in the second, third, fourth, and fifth decade of life with a
peak incidence in the third decade (see Table 3 and Figure
1).
The mean age of the most prevalent neoplasms, i.e., pleomorphic
adenoma had a mean age of 29 ± 15.74 with a range of 5-69 years and
a peak in the third decade of life (Figure 1). The mean age of
patients with salivary gland adenoma was 19.59 ± 1.73. Among the
malignant salivary gland neoplasms, adenocarcinoma had a mean age
of 47.17 ± 19.09, while adenoid cystic car- cinoma had a mean of
40.56 ± 17.09; salivary duct carcinoma and mucoe-pidermoid
carcinoma had a mean age of 46.8 ± 18.28 and 38 ± 18.74,
respectively.
The distribution of each type of salivary gland neoplasms as per
site of biopsy is shown in Table 4. Of all the histopathological
report slips, 47.1% did not have the site of biopsy specified.
Among those whose site of biopsy was available, the pal- ate and
parotid area were the most common sites 49.33% and 26.67%,
respectively. (See Table 3 and Figure 2).
Discusion There are only a few reports available that have reviewed
large studies of salivary gland neo- plasms worldwide. Eneroth
reviewed 2328 cases between the years 1919-1969 and Chaudary et al.
reviewed 1320 cases.17,18 More recently, addi- tional case studies
have been reported: Williams et al., Maita et al., Abiose et al.,
Vargas et al., Chidzonga et al., and Onyango et
al.2,6,8,16,19,21
Although there have been some reports on sali- vary gland neoplasms
as a component of tumors of the oral facial area from Uganda4, no
specific report is available from this part of the world on
salivary gland neoplasms independent from other classification of
tumors.
From Table 2 benign salivary gland neoplasms were more prevalent
(54.3%), which is consis- tent with several other studies as can be
seen in Table 5. Similar to Neely et al. our study shows an almost
equal prevalence of benign and malig- nant tumors.5 In contrast,
Williams et al. and
Vargas et al. reported significantly different preva- lences,
although benign lesions still predominate.6,16
Among the benign group and on the over all, pleo- morphic adenoma
was the most common salivary gland tumor. This agrees with previous
studies (Table 5) and with more recent studies6,8,14,16,21, hence
it seems the world over, Uganda inclusive, pleomorphic adenoma is
the most common salivary gland tumor. Slide 6 shows one of the
cases diag- nosed as pleomorphic adenoma.
Among the benign group, salivary gland adeno- ma was the second
most common diagnosis. Unfortunately while Cawson reports it1, this
diagno- sis does not exist in the World Health Organization (WHO)
classification20 but could be the equivalent of monomorphic
adenomas inclusive of adeno- lymphomas, a diagnosis not seen in
these studies or other types since at least three cases of oxy-
philic adenoma are reported. It must be noted in the WHO
classification20 salivary gland adenoma comprises a group of
fourteen tumors. The only subtype found in this study were
pleomorphic ade- noma (42.3%) and oxyphylic cell adenoma (oncon-
cytoma) 2.1%; the remaining adenomas (9.9%) were not further
classified. The pathologists helped us review some adenomas and we
discovered that adenomas as diagnosed varied from cannalicular,
warthin’s tumor, to cystadenoma as can be seen in slides 3, 4, and
5; slide 6 shows pleomorphic adenoma. Hence,there is a need to
review and standardize the diagnoses of the different patients’
specimens seen in the Department of Pathology according to the
current International Classification of Diseases (ICD)/World Health
Organization (WHO) classification.
6 The Journal of Contemporary Dental Practice, Volume 5, No. 3,
August 15, 2004
The results of our study are in agreement with the study done in
South Africa10; was found to be the most common salivary gland
malig- nancy. However, other studies are in contrast with these
findings2,5,6,7,11,12,16 with some rarely reporting this
tumor.18,19 Furthermore in this study some adenocarcinomas were not
sub typed; yet on review, variations were clear as can be seen in
slides 1 and 2 showing unspecified and acinic cell adenocarcinoma,
respectively. Thus, a combined effort to review African studies is
needed to estab- lish whether this is an African phenomenon or just
a regional one or whether reporting is standardized to a specific
classification. Adenoid cystic carcinoma in this study was
found to be the second most prevalent tumor fol- lowed by
muco-epidermoid carcinoma at 27.69% and 12.31%, respectively. This
is similar to a report from Nigeria, which ranks both as the most
common salivary gland neoplasms.2 Unlike the Jamaican and Brazilian
studies6,16, adenoid cystic carcinoma predominates over
muco-epidermoid carcinoma in our study. In this study salivary
gland carcinoma was the third most common lesion at 9.23%. This
diagnosis is not reported in other studies as can be seen from
Table 5 and is not a diagnosis cited in the WHO classifica- tion20
further stressing the need for standardiza- tion of reporting.
Mucus secreting carcinoma was reported in our study. As seen in
Table 5, some researchers, but not all, reported this category
of
Figure 1.
Figure 2. par = parotid gland, pal = palate, smf = submandibular
gland & fl oor of the mouth, uns = unspecifi ed.
7 The Journal of Contemporary Dental Practice, Volume 5, No. 3,
August 15, 2004
Slides 1 and 2 show two different adenocarcinomas whose sub type
was not specifi ed. However, on review they were sub typed as
unspecifi ed and acinic cell adenocarcinoma, respectively.
Slide 1. Slide 2.
Slide 4. Slide 3.
Slide 5. Slide 6.
Slides 3, 4, and 5 were all diagnosed as adenomas, while slide 6
was diagnosed as pleomorphic adenoma. On review, slides 3, 4, and 5
were sub-typed as cannalicular, Warthin’s tumor, and cyst
adenoma.
8 The Journal of Contemporary Dental Practice, Volume 5, No. 3,
August 15, 2004
Table 3. Age distribution of salivary gland neoplasms.
Table 4. Shows the sites from which the biopsy was taken and the
diagnosis.
9 The Journal of Contemporary Dental Practice, Volume 5, No. 3,
August 15, 2004
tumor. Cawson1 argues this is likely to belong to the epidermoid
tumor group and could be consid- ered either benign or malignant.
Acinic cell carci- noma constituted 4.6% of all salivary gland
malig- nancy and 2.1% overall which is in agreement with other
studies1,17,18 as being a rare salivary gland neoplasm. Carcinoma
ex pleomorphic constituted (2 cases) 1.4% overall and 3.08% of all
malignant lesions. Unfortunately, the references available to us
hardly mentioned this tumor. Cawson attributes its rarity to
carcinomatous changes in pleomor- phic adenoma occurring late,
thus, rarely given a chance to get to this stage1. However, Lewis
et al.22 reports non specified adenocarcinoma as one of the
sub-types of this tumor, and Varga et al.16
also notes all the cases they saw had adenocar-
cinoma as the malignant component; it is unclear whether
adenocarcinomas seen in this study might have been alternatively
classified as carcinoma ex pleomorphic adenoma.
Of all cases in this study 53.5% were female, 45.1 were specified
as male, and 1.4% were unspeci- fied. Thus, an almost equal sex
distribution (p=0.05) which is in agreement with several other
studies.2,4,5 The male to female ratio was 1:1.19, thus, no
statistically significant difference in sex distribution. These
statistics differ from those of Jordan8; hence, the question of
possible geographical differences in their prevalence. Both
malignant and benign neoplasms had equal sex distribution
(p=0.05).
Table 5. Salivary gland neoplasms prevalence from present study
along side other research findings.
10 The Journal of Contemporary Dental Practice, Volume 5, No. 3,
August 15, 2004
Pleomorphic adenoma had a slight preponder- ance for females with
the male to female sex ratio being 1:1.3, which is in agreement
with other find- ings.13 Unlike a report from Nigeria14, for all
age groups, female distribution predominated over males. Salivary
gland adenoma had equal sex distribution, but because no other
study we came across recognized this entity we had nothing to
compare with. All of the three cases of oxyphil cell adenoma were
female though Cawson1 writes of a female preponderance; in this
study we could not establish any sex preponderance due to the few
cases found. Adenocarcinoma had equal sex distribution (p=0.05).
However, for poorly dif- ferentiated adenocarcinoma, there was male
pre- ponderance over female (p=0.05). Adenoid cystic carcinoma had
female preponderance over male (p=0.05). The numbers of salivary
gland carcino- ma, muco-epidermoid carcinoma, and acinic cell
carcinoma were too few to comfortably establish sex preponderance.
Carcinoma ex pleomorphic adenoma had only two cases, both being
male.
The distribution of the salivary gland neo- plasms as per site of
biopsy is shown in Table 4. Unfortunately, many of the biopsy sites
in this study’s cases 47.18% were not recorded. Of those recorded,
the palate featured greatly con- stituting 45.94%, followed by the
parotid area 26.67%, and sub-mandibular area 16% (Table 4 and
Figure 2). These results agree with a few other studies from
Africa. The palatal neo- plasms had a preponderance for females
64.7% (p=0.005), a fact that was reported by Onyango et al.21 The
parotid region was equally affected by both malignant and benign
tumors, which is in contrast with other reports17,18, therefore,
geo- graphical differences may exist.
Though pleomorphic adenoma is known to mainly affect the parotid
gland1,17, in this study it affected the palate (p=0.05). The lip
was affected in just three cases. Of the three lip lesions, two
were malignant, i.e., adenoid cystic carcinoma and muco-epidermoid
carcinoma with the third being salivary gland adenoma. Owens et
al.24 reports adenoid cystic carcinoma and muco-epidermoid
carcinoma as the most frequent malignant salivary gland neoplasms
of the lip.
As reported by other authors elsewhere, our results agree with the
findings that pleomorphic adenoma is common among adults and in the
fact salivary neoplasms appear in the third, fourth, and fifth
decade of life with a peak in the third decade (Table 3 and Figure
1). Though not reflected in either the tables or figures, the
general overall presentation between the sexes, in terms of age at
the time of reporting, revealed there were many more females with
pleomorphic adenoma than males in the peak group.
Conclusion There are significant reporting differences by diag-
nosis and a lack of information regarding the sites of
lesions.
There is a geographic variation in the prevalence of salivary gland
neoplasms according to sex, age, and site. More research into
specific salivary gland neoplasms inclusive of treatment
modalities, prognosis, and causative factors are needed in
Uganda.
While the objective of this study was to establish the prevalence
and site distribution of salivary gland neoplasms as seen in the
Department of Oral Surgery Mulago Hospital, we realize a need for
standardization and improved reporting by both the surgeons and
pathologists.
11 The Journal of Contemporary Dental Practice, Volume 5, No. 3,
August 15, 2004
References 1. Cawson RA. Essentials of dental surgery and pathology
4th edition. Churchill Livingstone Medical
Division of Longman Group UK Ltd. 1984. 298-311. ISBN 044304015X.
2. Abiose BO, Oyejide O, Ogunniyi J. Salivary gland tumours in
Ibadan, Nigeria: a study of 295 cases.
Afr J Med Med Sci. 1990 Sep;19(3):195-9. 3. Onyango JF, Awange DO,
Wakiaga JM. Oral tumours and tumour-like conditions in Kenya:
I.
Histological distribution. East Afr Med J. 1995 Sep;72(9):560-3. 4.
Pritam Singh and J.Cook. Tumors of the jaws.East African Medical
Journal Oct. 1995 33(10): 383-
390 5. Neely MM, Rohrer MD, Young SK. Tumors of minor salivary
glands and the analysis of 106 cases. J
Okla Dent Assoc. 1996 Spring;86(4):50-2. Review. 6. Williams NP,
Boyd DL, Choy L, et. al. Salivary gland lesions: a Jamaican
perspective. West Indian
Med J. 2001 Mar;50(1):62-5. 7. Loyola AM, de Araujo VC, de Sousa
SO, et. al. Minor salivary gland tumours. A retrospective
study
of 164 cases in a Brazilian population. Eur J Cancer B Oral Oncol.
1995 May;31B(3):197-201. 8. Jasser K Ma’aita, Nabih Al-Kaisi,
Shawkat Al- Tamimi, Abdulwahab Wraikat. Salivary gland tumors
in
Jordan: A retrospective study of 221 patients. The Croatian Medical
Journal March 1999 Volume 40 Number 4 (Medline).
9. Idris AM, Ahmed HM, Mukhtar BI, et. al. Descriptive epidemiology
of oral neoplasms in Sudan 1970- 1985 and the role of toombak. Int
J Cancer. 1995 Apr 10;61(2):155-8.
10. van Heerden WF, Raubenheimer EJ. Intraoral salivary gland
neoplasms: a retrospective study of seventy cases in an African
population. Oral Surg Oral Med Oral Pathol. 1991
May;71(5):579-82.
11. Kusama K, Iwanari S, Aisaki K, et. al. Intraoral minor salivary
gland tumors: a retrospective study of 129 cases. J Nihon Univ Sch
Dent. 1997 Sep;39(3):128-32.
12. Eveson JW, Cawson RA. Tumours of the minor (oropharyngeal)
salivary glands: a demographic study of 336 cases. J Oral Pathol.
1985 Jul;14(6):500-9.
13. Chidzonga MM, Lopez Perez VM, Portilla Alvarez AL Pleomorphic
adenoma of the salivary glands. Clinicopathologic study of 206
cases in Zimbabwe. Oral Surg Oral Med Oral Pathol Oral Radiol
Endod. 1995 Jun;79(6):747-9.
14. Arotiba GT. Salivary gland neoplasms in Lagos, Nigeria. West
Afr J Med. 1996 Jan-Mar;15(1):11-7. 15. J.V. Soames and
J.C.Southam. Oral Pathology 2nd edition Oxford Medical Publications
1993, 245-
253. ISBN 0192622153 16. Vargas PA, Gerhard R, Araujo Filho VJ, et.
al. Salivary gland tumors in a Brazilian population: a ret-
rospective study of 124 cases. Rev Hosp Clin Fac Med Sao Paulo.
2002 Nov-Dec;57(6):271-6. Epub 2003 Feb 17.
17. Eneroth CM. Salivary gland tumors in the parotid gland,
submandibular gland, and the palate region. Cancer. 1971
Jun;27(6):1415-8. No abstract available.
18. Chaundry AP, Vickers RA, Gorlin RJ. Intraoral minor salivary
gland tumors. An analysis of 1,414 cases. Oral Surg Oral Med Oral
Pathol. 1961 Oct;14:1194-226. No abstract available.
19. Chidzonga MM, Lopez Perez VM, Portilla-Alvarez AL. Salivary
gland tumours in Zimbabwe: report of 282 cases. Int J Oral
Maxillofac Surg. 1995 Aug;24(4):293-7.
20. Seifert G, Brocheriou C, Cardesa A, et. al. WHO International
Histological Classification of Tumours. Tentative Histological
Classification of Salivary Gland Tumours. Pathol Res Pract. 1990
Oct;186(5): 555-81.
21. Onyango JF, Awange DO, Muthamia JM, et. al. Salivary gland
tumours in Kenya. East Afr Med J. 1992 Sep;69(9):525-30. 69(9)
525-530.
22. Lewis JE, Olsen KD, Sebo TJ. Carcinoma ex pleomorphic adenoma:
pathologic analysis of 73 cases. Hum Pathol. 2001
Jun;32(6):596-604.
23. Ezeanolue BC. Salivary gland neoplasms: a descriptive analysis
of the pattern seen in Enugu. West Afr J Med. 1999
Jul-Sep;18(3):179-82.
24. Owens OT, Calcaterra TC. Salivary gland tumors of the lip. Arch
Otolaryngol. 1982 Jan;108(1):45-7.
12 The Journal of Contemporary Dental Practice, Volume 5, No. 3,
August 15, 2004
About the Authors