journal of dentistry 38 (2010) 83–95

available at www.sciencedirect.com

journal homepage: www.intl.elsevierhealth.com/journals/jden

Review

The association between periodontal disease and cancer:

A review of the literature

Sarah G. Fitzpatrick *, Joseph Katz

Department of Oral and Maxillofacial Diagnostic Sciences, University of Florida College of Dentistry, Gainesville, FL, USA

article info abstract

Article history: Objectives: Periodontal disease has long been linked to many systemic diseases, and recently
Received 11 June 2009 a link between periodontal disease and cancer has been established. The purpose of this
Received in revised form paper is to review the literature to explore the evidence to date of a relationship between
14 October 2009 periodontal disease and cancer. In addition, the main hypotheses for the association are
Accepted 27 October 2009 discussed along with challenges in evaluating the evidence.
Data/sources/study selection: In this review, English-language papers studying the relation-
ship between periodontal disease or tooth loss in humans and increased risk of several types
Keywords: of cancers along with overall cancer risk between 1990 and April 2009 were reviewed.
Risk factors Conclusions: The most consistent increased risk was noted in studies of oral and esophageal
Periodontitis cancers and periodontal disease. Gastric and pancreatic cancers had an association in most
Oncology but not all studies. Lung, prostate, hematologic and other cancers were less consistently
associated or did not have sufficient studies to determine a predictable pattern. Studies to
date indicate a positive correlation between several forms of cancer and periodontal disease.
# 2009 Elsevier Ltd. All rights reserved.

1. Introduction destruction.5 Many cases have been reported of gingival

squamous cell carcinoma presenting clinically similar to
Periodontal disease has been linked to many systemic inflammatory periodontal or periodontal/endodontic
conditions such as, cardiovascular disease, low-birth weight lesions.6–12 Cases of other types of cancer mistaken for
complications in pregnancy, diabetes, and, pulmonary dis- periodontal disease such as metastatic pancreatic cancer13
ease.1 More recently, a link between periodontal disease and and osteogenic sarcoma14 have also been reported. These
cancer has been suggested through several studies looking examples hint that a similar underlying mechanism may be
both at specific types of cancers and at the overall total cancer responsible for both periodontal disease and cancer.
rate and the relationship to periodontal disease. The scientific Review of the pertinent literature reveals major differences
rationale behind the proposed association is that inflamma- in criteria used by the various investigators for the diagnosis
tion is a major factor in both periodontal disease and cancer.2–4 and grading of periodontal disease. Furthermore, many
Oral cancer, gingival squamous cell carcinoma in parti- potential confounding factors that may influence the validity
cular, has been known to mimic advanced periodontal disease of the study have not always been considered. The purpose of
in clinical appearance showing similar symptoms of swelling, this paper is to critically review the current literature reporting
bleeding, tooth mobility, deep periodontal pockets, and bone the association between periodontal disease and several types

* Corresponding author at: Department of Oral and Maxillofacial Diagnostic Sciences, PO Box 100414, Gainesville, FL 32610-0416, USA.
Tel.: +1 352 273 6683; fax: +1 352 392 2507.
E-mail address: sfitzpatrick@dental.ufl.edu (S.G. Fitzpatrick).
0300-5712/$ – see front matter # 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.jdent.2009.10.007
84 journal of dentistry 38 (2010) 83–95

of cancer along with overall cancer risk and discuss the main strong trends when these two groups have been separated
hypotheses for the association. from each other.

3.1. Oral cancer

2. Materials and methods
Oral cancer and a link to tooth loss or periodontal disease has
In this review, articles were identified by searches on PubMed been evaluated in several studies.2,4,15–24 Most found a
from 1990 through April 2009 in the English language including significant increase of oral cancer risk with patients with
humans. We used the following search terms: ‘‘periodontal increased tooth loss or other parameters of periodontal
disease’’, ‘‘periodontitis’’, and ‘‘tooth loss’’ linked with disease even after adjustment for tobacco and alcohol.
‘‘cancer studies’’, ‘‘oral cancer studies’’, ‘‘gastrointestinal Perhaps the most informative studies of these are two by
cancer studies’’, ‘‘lung cancer studies’’, ‘‘pancreatic cancer Tezal et al., since objective measures of periodontal disease
studies’’, ‘‘prostate cancer studies’’, and ‘‘hematologic cancer were used in these as opposed to tooth loss as a measurement
studies’’. Studies were also identified from the reference lists tool.4,23 The first reported a 5.23-fold increase in risk of tongue
of retrieved papers. Case reports were excluded from this cancer with each millimeter of bone loss in periodontal
study. disease patients in a study of men with tongue cancer in
Buffalo, New York, and in fact found no association with other
parameters of dental disease such as caries or significant prior
3. Results dental work.4 In an earlier cohort study of a NHANES III data,
Tezal et al. found an increased incidence of oral tumors or pre-
The studies evaluated in this review varied greatly in design, cancerous oral lesions in patients with greater than 1.5 mm
criteria, and evaluation methods. Several factors were taken clinical attachment loss.23 The only other study that used a
into consideration to derive conclusions from these avail- specific periodontal disease indicator was a large cohort study
able studies. Study design and sample size were the first by Michaud et al. and was in fact the only study not finding a
factor taken into consideration. All of the studies included significant association; however, two factors should be
fell into the categories of either cohort or case–control considered in evaluating this.2 First of all, oral cancer was
studies. Most of the cohort studies included very large grouped in with all oropharyngeal cancers in this study and
numbers of potential subjects; however, some of the thus no true determination of the risk of oral cancer alone
numbers of individual cases of specific cancers were very could be determined.2 Secondly, the parameter measuring
small, numbering in some instances less than 10 cases in periodontal disease was less strong than objective measure of
tens of thousands of overall subjects. Therefore, though the clinical attachment loss because it relied on patient reported
large cohort studies should be considered to be perhaps history of periodontal disease verified by radiographic bone
more informative than the case–control studies, the small loss.2 Of the studies in this review looking at oral cancer and
numbers of individual cancers should provoke some periodontal disease or missing teeth, eleven found a sig-
reservation. It then seems to follow that when looking at nificant increase in risk of oral cancer (odds ratio [OR] ranging
the large cohort information available, the data may be from 1.55 to 8) and one found no significant association (see
more reliable for subsets such as overall cancer rates where Table 1). This may indicate that in the case of oral cancer, a
the cancer cases numbered in the hundreds or thousands valid association is likely present for both tooth loss and
than for very specific cancers where the numbers of cases periodontal disease.
are much smaller within the cohort. A second consideration
should be the adjustment for confounding factors. Most 3.2. Esophageal cancer
studies attempted to at least control for major risk factors
such as tobacco and alcohol use, therefore for the most part Several studies have found a link between esophageal cancer
studies could be compared relatively evenly in regards to and missing teeth.2,3,18,19,25–29 It is difficult to determine,
this factor. however, if this translates to a direct link to periodontal
The most important variable in the comparison of the disease because only one of these studies, Michaud et al., used
studies available was the periodontal disease criteria used and a specific measure of periodontal disease rather than missing
whether or not it specifically measured some form of period- teeth, and their large scale cohort study failed to find an
ontal disease value or merely relied on missing teeth as a association with esophageal cancer either with patient self-
surrogate of presumed presence of periodontal disease. Earlier reported periodontal disease verified by radiographic bone
studies all relied on missing teeth alone as parameters, which loss or with missing teeth.2 Hiraki et al. reported a significant
of course may be misleading due to the differing reasons that association with esophageal cancer similar to their previously
teeth can be lost other than periodontal disease including mentioned work with head and neck cancer.19 A recent study
caries activity and trauma. However, the presence of several by Abnet et al. in Iran found that increasing numbers of teeth
large studies using specific parameters measuring periodontal lost had a direct relationship on the risk of esophageal
disease gave the most valuable information to draw conclu- squamous cell carcinoma noting that edentulous subjects had
sions from. When analyzing the studies as a whole, striking twice the risk as compared to subjects who lost fewer than 13
patterns can be seen differentiating the studies measuring teeth.27 Guha et al. showed that missing between 6 and 15
specifically for periodontal disease against the studies teeth doubled the risk of esophageal cancer when compared
measuring for tooth loss, and in many cases the results show with subjects missing less than 5 in two concurrent studies in
Table 1 – Studies linking oral cancer with periodontal disease.
Author Year Population Criteria Results Type of Study
a
Bundgaard 1986–1990 161 cases oral SCCA and Loss of teeth Significantly increased risk (OR 2) of developing Case–control
et al.15 400 controls in Denmark oral SCCA for patients with fewer than 15 teeth
after smoking/EtOH adjustment
de Rezende 2003–2004 50 patients with oral and CPITNb and DMFTc 76% of subjects in cancer group showed >6 mm Case–control
et al.16 oropharyngeal squamous pockets compared to 10% of control group
cell carcinoma and (No OR calculated) No association with DMFT
50 cancer-free controls
Fernandez- 1996–1999 200 cases oral cancer and Loss of teeth Significantly increased risk (OR 2.7) of oral Case–control
Garrote controls in Cuba cancer for patients missing 16 or more teeth
et al.17 after smoking/EtOH adjustment
Guha Europe 1998–2002 924 cases head and neck and Dental condition by exam (good, average, Poor oral condition significant increased risk Case–control
et al.18 Latin America esophageal SCCA/928 controls poor as measured with tartar, bleeding, (Europe OR 2.89, LA OR 1.89) of head and
1998–2003 in Europe and 2286 cases and mucosal irritation, and caries present) neck cancer

journal of dentistry 38 (2010) 83–95

1824 controls in Latin America and missing teeth
Hiraki 2001–2005 429 cases head and neck Loss of teeth Significantly increased risk of head and neck Case–control
et al.19 cancer out of 5240 cancer cancer to decreased number remaining teeth
patients and 10,480 control (OR 1.68 for 0 remaining teeth) after smoking/
patients in Japan EtOH adjustment
Marshall 1975–1983 290 subjects and controls Loss of teeth Significantly increased risk of oral cancer Case–control
et al.20 in the United States (OR 2.7) with loss of 11 or more teeth after
smoking and EtOH adjustment
Michaud 1986–2004 118 cases of oropharyngeal Patient reported history of periodontal No significant increase in risk for oropharyngeal Cohort
et al.2 cancer out of 48,328 cohort disease (validated by radiographic cancer with history of periodontal disease or
in United States measure of bone loss) and loss of teeth increased number of teeth lost
Rosenquist 2000–2004 132 cases and 320 Loss of teeth and radiographic evaluation Significantly increased risk (OR 3.4) of oral and Case–control
et al.21 controls in Sweden (amount of marginal bone loss) oropharyngeal cancer for missing over 20 teeth
after adjustment for smoking and EtOH
Radiographic evidence of increased risk also
but not after adjustment for confounders
Talamini 1996–1999 132 cases and 148 Loss of teeth and poor general oral No significant increase in risk of oral cancer to Case–control
et al.22 controls in Italy condition (tartar, decayed teeth, and increased missing 16 or more teeth found but poor
mucosal irritation) condition of mouth was associated (OR 4.5) after
smoking and EtOH control
Tezal 1988–1994 131 oral tumors and 323 oral Clinical attachment loss (CAL) Significantly increased risk of oral tumor (OR 4.57) and Cohort
et al.23 pre-cancerous lesions out pre-cancerous lesion (OR 1.55) with >1.5 mm CAL both
of 13,798 subjects still significant in never smokers (tumors OR 1.63,
pre-cancerous OR 1.32)
Tezal 1999–2005 51 cases and 54 controls Mm of alveolar bone loss Significantly increased risk (OR 5.23) of tongue Case–control
et al.4 in the United States cancer with each mm of bone lost after
smoking adjustment
Zheng 1989–1990 404 subjects and control Loss of teeth Significantly increased risk of oral cancer with Case–control
et al.24 pairs in China increased missing teeth (OR 2–3 men, 5–8 women)
controlling for smoking and EtOH
a
Odds ratio.
b
WHO guidelines of Community Periodontal Index of Treatment Needs (depth of periodontal pockets on scale of 1–4).

85
c
WHO guidelines of Decayed, Missing, and Filled Teeth.
 86
Table 2 – Studies linking esophageal cancer and upper GI with periodontal disease.
Author Year Population Criteria Results Type of study
3
Abnet et al. 1985–1999 49 esophageal cancer cases and Loss of teeth No significant association seen for esophageal or Cohort
66 esophageal/gastric cardia cases gastric cardia cancer with increased loss of teeth
out of 29,124 cohort in Finland
Abnet et al.25 1986–1991 620 cases of esophageal cancer Loss of teeth Significantly increased risk of esophageal cancer Cohort
in a 28,868 cohort in China (OR 1.3) with increasing tooth loss adjusted for
tobacco and EtOH
Abnet et al.26 1986–1991 2625 upper GI deaths in 29,584 Loss of teeth Significant increase in risk of upper GI deaths Cohort

journal of dentistry 38 (2010) 83–95

cohort in China (OR 1.25 under age 50) with increasing loss of teeth
especially in a younger age controlled for smoking
and EtOH
Abnet et al.27 2003–2007 283 esophageal SCCA and 560 DMFT and poor hygiene Significantly increased risk of esophageal SCCA with Case–control
controls in Iran Loss of teeth 32 DMFT compared to <15 (OR 2.10) Also found
significant risk with poor oral hygiene and increasing
numbers of teeth lost—controlled for smoking
Guha et al.18 Europe (EU) 924 cases head and neck and Loss of teeth and dental Significantly increased risk of esophageal cancer Case–control
1998–2002 esophageal SCCA/928 controls in condition by exam (good, when missing between 6 and 15 teeth (EU OR 3.4,
Latin America Europe and 2286 cases and 1824 average, poor as measured LA OR 2.18) controlled for smoking and EtOH
(LA) 1998–2003 controls in Latin America with tartar, bleeding,
mucosal irritation, and
caries present)
Hiraki et al.19 2001–2005 354 esophageal cancers out of Loss of teeth Significantly increased risk for esophageal cancer Case–control
5240 cancer patients and 10,480 with increased number of teeth lost (OR 2.36 for
control patients in Japan 0 teeth remaining) after smoking/EtOH adjustment
Michaud et al.2 1986–2004 131 cases of esophageal cancer Patient reported history of No significant increase in risk for esophageal cancer Cohort
out of 48,328 cohort in United States periodontal disease with history of periodontal disease or increased
(validated by radiographic number of teeth lost
measure of bone loss) and
loss of teeth
Sepehr et al.28 1995–1996 124 cases of precursor esophageal Poor oral health (rated good, Significantly increased risk (OR 5) for dysplasia in Case–control
dysplasia vs. 50 control in Iran fair, poor, very poor, or edentulous patients over good oral health
prostheses) and loss of teeth
Wei et al.29 2002 724 subjects in China Loss of teeth Significant increased risk for esophageal dysplasia Cohort
in patients having lost more but not all
teeth -controlled for smoking
Table 3 – Studies linking gastric cancers with periodontitis.
Author Year Population Criteria Results Type of study

Abnet et al.3 1985–1999 179 gastric non-cardia and 66 Loss of teeth Significantly increased risk of gastric non-cardia cancer (OR 1.65) Cohort
esophageal/gastric cardia in Finland for loss of all teeth compared to losing less than 10 after
adjustment for smoking/EtOH. No significant increased risk for
gastric cardia cancer

journal of dentistry 38 (2010) 83–95

Abnet et al.25 1986–1991 431 gastric cardia and 102 gastric Loss of teeth Significantly increased risk for gastric cardia (OR 1.3) and Cohort
non-cardia cancers in China non-cardia cancer (OR 1.8) with increasing tooth loss,
controlled for smoking and EtOH
Demirer et al.30 1987–1988 100 cases of adenocarcinoma of Deficient teeth and Stomach cancer patients were more likely to brush their teeth Case–control
stomach and 100 controls in Turkey poor oral hygiene less frequently and have more deficient teeth than control
Hiraki et al.19 2001–2005 702 stomach cancers out of 5240 Loss of teeth No significant association between gastric cancer and Case–control
cancer patients and 10,480 control increased loss of teeth
patients in Japan
Hujoel et al.31 1971–1975, 8 cases of stomach cancer in 11,328 Periodontal exam by Russell No significant increase in stomach cancer risk for patients Cohort
follow-up cohort in United States index (classified as periodontitis, with periodontitis found
until 1992 gingivitis, no periodontitis,
or edentulous)
Michaud et al.2 1986–2004 106 cases of stomach cancer out of Patient reported history of No significant association between stomach cancer and Cohort
48,328 cohort in United States periodontal disease (validated by history of periodontal disease or loss of teeth
radiographic measure of bone
loss) and loss of teeth
Watabe et al.32 1996–1997 242 cases gastric cancer and 484 Loss of teeth Significant increased risk of gastric cancer with more than Case–control
controls in Japan 20 teeth lost when compared to none lost (OR 3–4),
controlled for smoking

87
88 journal of dentistry 38 (2010) 83–95

Europe and Latin America.18 In the same study, other

Type of study

Case–control
indicators of oral health other than missing teeth were not
significantly associated.18 Overall, of studies examining

Cohort

Cohort

Cohort
esophageal cancer and periodontal disease or tooth loss,
seven found a significant association (OR from 1.25 to 5) (see
Table 2), and two studies failed to find a significant increase in

disease and persisted after control smoking

risk. This shows that while an association may be present for

Significantly increased risk of lung cancer

Significantly increased risk of lung cancer

Significantly increased risk of lung cancer

Significantly increased risk of lung cancer

(OR 1.94) but not significant when limited

No significant association noted between

with increased loss of teeth (OR 1.54 for
an increased risk of esophageal cancer for patients with an

0 teeth remaining) after adjustment for

25–32 but did not persist after smoking

(OR 1.70) with 1–16 teeth compared to
overall poor oral condition and increased numbers of missing

(OR 1.36) with history of periodontal

teeth, the evidence does not yet support a specific association
to periodontal disease indicators.

lung cancer and loss of teeth

Results
3.3. Upper GI and gastric cancers

controlled (OR 0.96)

The association between upper GI and gastric cancers has

to non-smokers

smoking/EtOH
been evaluated in the context of periodontal disease.2,3,19,25,30–
32
Increased tooth loss as a risk factor seems to exhibit a mixed
result in the literature. Abnet et al. found a significantly
elevated risk between tooth loss and esophageal squamous
cell carcinoma, gastric cardia adenocarcinoma, and gastric
non-cardia adenocarcinoma. The strongest association was

Patient reported history of periodontal

noted between gastric non-cardia adenocarcinoma and tooth

(classified as periodontitis, gingivitis,

loss in a study originating in China.25 However, in another

Periodontal exam by Russell index

disease (validated by radiographic

no periodontitis, or edentulous)
study in Finland, Abnet et al. found a significant association
between tooth loss and gastric non-cardia adenocarcinoma
but not with esophageal squamous cell carcinoma or gastric
Criteria

measure of bone loss)

cardia adenocarcinoma.3 Of the studies, there were two large
cohort studies that evaluated the relationship between
specific periodontal indicators and stomach cancer.2,31 As in
Loss of teeth

Loss of teeth

Loss of teeth
oropharyngeal and esophageal cancers, Michaud et al.
reported no significant association between stomach cancers
and history of periodontal disease verified by radiographic
bone loss or with tooth loss as well.2 Hujoel et al. in the
NHANES I Epidemiologic Follow-up Study found no significant
191 lung and bronchus cancer deaths

678 lung cancers out of 48,328 cohort

association between stomach cancer and periodontitis as

in 11,328 cohort in the United States

patients and 10,480 control patients

909 lung cancers out of 5240 cancer

Lung cancer number unspecified in

defined by clinical examination but had a small number of
cases within their cohort.31 In review of gastric cancer and
periodontal disease or tooth loss studies, four studies found an
Table 4 – Studies linking lung cancer and periodontal disease.

12,223 cohort in Scotland

Population

increased risk of cancer (OR ranging from 1.3 to 1.8), and three
did not (see Table 3). Again, as the studies specifically
measuring parameters of periodontal disease did not show
in United States

an association, the evidence as of yet does not support a link

from gastric or upper GI cancer with periodontal disease.
in Japan

3.4. Lung cancer

The relationship between lung cancer and periodontitis or

follow-up until 1992

follow-up until 2005

missing teeth has also been evaluated several times as well in

the literature.2,19,31,33 Again, two studies specifically exam-
1948–1968 and

ined the link using periodontal measures. Hujoel et al.

Year

1971–1975,

2001–2005

1986–2004

examined the relationship between periodontal disease by

clinical examination and lung cancer and reported a strong
initial association.31 However, the result was not significant
after limiting the effects of never smokers31 which tempers
the strength of the association as smoking is such a strong
Michaud et al.2
31

risk factor for lung cancer. Michaud et al. found a significant

Hiraki et al.19
Hujoel et al.

increased risk in patients with a history of periodontal

Tu et al.33
Author

disease that persisted after controlling for smoking, but for

patients with tooth loss they noted a significant increase only
before controlling for smoking.2 The other two studies used
 journal of dentistry 38 (2010) 83–95 89

tooth loss as their measure with Hiraki et al. finding a

Type of study

Case–control
significantly increased risk of lung cancer with increased loss
of teeth after adjusting for tobacco and alcohol19 and Tu et al.

Cohort

Cohort

Cohort
in the Glasgow Alumni Cohort failing to find an association
between increased numbers of missing teeth and lung cancer
after controlling for tobacco use.33 Therefore, three out of four
authors studying lung cancer and periodontal disease or

Increased risk noted for pancreatic

No significant increase in risk for

pancreatic cancer for edentulous

tooth loss found some level of association (OR ranged from

No significant association found

between pancreatic cancer and

pancreatic cancer with history

patients compared to patients

of periodontitis (OR 1.64) after
1.36 to 1.94) although in most cases it did not remain after

periodontal disease (OR 1.77)

missing 1–10 teeth (OR 1.63)

Significant increased risk of

Significant increased risk of

adjustment for smoking (see Table 4). Overall, this suggests

cancer with patients with

adjustment for smoking

increasing loss of teeth

that smoking is a very high confounder of this relationship,

increased loss of teeth

Results

adjusted for smoking

and thus there is some reservation in interpreting these
results as proof of a link between lung cancer and periodontal
disease.

3.5. Pancreatic cancer

The link between pancreatic cancer and periodontal disease

or tooth loss has had mixed results in several studies.19,31,34,35

measure of bone loss) and loss of teeth

Two large cohort studies that used a specific periodontal

Patient reported history of periodontal

(classified as periodontitis, gingivitis,
measure both found significant associations.31,34 Michaud

Periodontal exam by Russell index

disease (validated by radiographic

et al. looked at a link between pancreatic cancer and history of

no periodontitis, or edentulous)
periodontal disease and found a significant association
between the two, in fact finding a higher association among

Criteria
the non-smokers than the smokers.34 Hujoel et al. found a
significant association between pancreatic cancer and per-
iodontitis measured by examination but again had a Loss of teeth

Loss of teeth
relatively small number of cases within their cohort.31 Of

Loss of teeth
the other two studies involving loss of teeth, the results were
again mixed. A significant association was found between
pancreatic cancer and tooth loss by a Finnish study.35 Hiraki
et al., however, found no association between pancreatic
cancer and tooth loss.19 Overall, for pancreatic cancer and
patients and 10,480 control patients in Japan

12 pancreatic cancer deaths in 11,328 cohort

periodontitis, three studies found an increase in risk (OR

178 cases pancreatic cancer of 5240 cancer

253 cases in 48,328 cohort in United States

ranging from 1.63 to 1.77) and one did not (see Table 5). The
evidence favoring an association specifically with period-
Table 5 – Studies linking pancreatic cancer and periodontal disease.

ontal disease appears promising for pancreatic cancer when 174 cases in 29,104 cohort in Finland
the two studies by Michaud et al. and Hujoel et al. are taken
Population

into account.31,34

3.6. Prostate cancer

in United States

Several investigators examined the relationship between

prostate cancer and tooth loss or periodontal disease2,19,31
(see Table 6). Interestingly the two studies which used tooth
loss as an indicator actually found an inverse relationship.2,19
Hiraki et al. found that subjects with greater tooth loss actually
demonstrated a decreased risk of prostate cancers.19 They
speculated that socio-economic status may explain this
1971–1975,
2001–2005

1986–2004

1985–1988
until 1992
follow-up

finding, as a negative association between prostate cancer

Year

and socio-economic status has been previously suggested.36 A

similar inverse relationship was also reported by Michaud
et al.2 Hujoel et al. was the only study to use direct assessment
of periodontal state and established a slight positive associa-
Solomon et al.35

tion between prostate cancer and periodontal disease; how-

Michaud et al.34
Hujoel et al.31

ever, they had a low sample size (20 cases of patients with
19

Stolzenberg-
Hiraki et al.

prostate cancer) in comparison to the aforementioned

Author

studies.31 It does not appear that as of yet a strong association

can be made at this time between periodontal disease and
prostate cancer.
90 journal of dentistry 38 (2010) 83–95

Table 6 – Studies linking prostate cancer and periodontitis.

Author Year Population Criteria Results Type of study
19
Hiraki et al. 2001–2005 136 cases of 5240 cancer Loss of teeth Negative (inverse) association Case–control
patients and 10,480 to risk of prostate cancer
control patients in Japan for increased number of
teeth lost
Hujoel et al.31 1971–1975, 20 deaths in 11,328 cohort Periodontal exam Increased risk of prostate Cohort
follow-up in United States by Russell index cancer with patients with
until 1992 (classified as periodontitis, periodontitis (OR 1.81)
gingivitis, no periodontitis,
or edentulous)
Michaud et al.2 1986–2004 541 out of 48,328 cohort Loss of teeth Negative (inverse) association Cohort
in United States between risk of prostate cancer
and number of teeth lost

3.7. Hematologic cancers periodontitis by examination and found no significant

association.31
Hematologic cancers in association with periodontal disease
have had limited study2,19 (see Table 7). Michaud et al. looked 3.10. Overall cancer
at hematologic malignancy for an association with history of
periodontal disease with verified radiographic bone loss and Several cohort studies have looked at overall cancer rates for a
reported an increase with non-Hodgkin lymphoma (NHL), possible link to periodontal disease2,31,33,37 (see Table 9). In
leukemia, and myelomas but only NHL yielded a significant these, an interesting pattern has emerged. Again, the two
relationship.2 Michaud et al. also evaluated loss of teeth in studies which specifically used periodontal measures as
relationship to hematologic malignancies and found no criteria both found a positive association with overall cancer
significant increase in risk.2 Hiraki et al. could not demon- rates, and lending more credibility was the large sample size
strate an association between lymphomas in general and for both.2,31 The Hujoel et al. study found an increase in overall
tooth loss.19 As of yet, there has been insufficient study to cancer rates for patients with periodontal disease by clinical
conclude if a relationship exists, but as the Michaud study was examination,31 and Michaud et al. found a small but
a cohort study with quite large numbers and found a significant increase in overall cancer risk even when adjusted
relationship with NHL after separating out periodontal for smoking when looking at patients with reported history of
patients from general tooth loss, the suggestion of a link is periodontal disease as verified by radiographic bone loss.2 The
an interesting course for future study. other two studies both measured missing teeth in relation to
overall cancer rates, and neither found an association.33,37
3.8. Breast cancer This seems to suggest that while missing teeth may not be an
independent risk factor for overall cancer risk, periodontal
Breast cancer was investigated in two studies (see Table 8) disease may play a significant part.
with Hujoel finding a significant association between breast
cancer risk and periodontal disease by examination31 but
Hiraki finding no association between breast cancer risk and 4. Discussion
tooth loss.19 The Hiraki et al. study was not specific for
periodontal indicators but did have a much higher numbers of Several hypotheses are of interest in the potential etiology of a
breast cancer patients, whereas Hujoel et al. used a specific link between, periodontal disease and cancer.
measure of periodontal disease but had rather low numbers.
This may reflect a difference between the tooth loss group and 4.1. Alteration of the oral flora
true periodontal disease groups; however, the data seems
insufficient to draw a strong conclusion at this point. It has been suggested that carcinogenic metabolic by-products
of periodontal disease might account for the relationship
3.9. Other cancers between the two diseases. When considering gastric cancers,
Abnet et al. argue that the mechanism of increased cancer risk
Hiraki et al. also investigated liver, ovary, uterus, bladder, may be an increased production of nitrosamines in situations
thyroid, and colon cancer in their study and found no of poor oral hygiene and that these by-products may function
significant associations for these cancers and missing teeth.19 as GI organ specific carcinogens.25 Nitrosamines have been
Michaud et al. also studied several other cancers in relation- linked to cancers of the stomach and esophagus.38 Heliobacter
ship to a history of periodontal disease and missing teeth and pylori infection also plays a role in stomach cancers.39,40 The
found rectal, bladder, and brain with no significant associa- carcinogenic ethanol by-product acetaldehyde has also been
tion found. They also found kidney cancer to have a studied, as there is a clear link between upper GI cancers and
significant association with history of periodontal disease, alcohol.41 A study by Homann et al. showed that poor dental
and noted an inverse relationship between melanoma and status was shown to lead to a 2-fold increase in salivary
tooth loss.2 Hujoel et al. also looked at colon cancer and acetaldehyde production by oral microbes.42 In another study,
Table 7 – Studies linking hematologic cancers and periodontitis.
Author Year Population Criteria Results Type of study
19
Hiraki et al. 2001–2005 232 cases of lymphoma of 5240 Loss of teeth No significant increase in risk of lymphoma Case–control
cancer patients and 10,480 with increased number of missing teeth
control patients in Japan
Michaud et al.2 1986–2004 934 cases of hematologic cancers out Patient reported history of Significant increased risk of NHL with patients Cohort
of 48,328 cohort in United States periodontal disease (validated reporting periodontal disease history (OR = 1.31)
by radiographic measure of bone loss) (only NHL was significant out of all heme cancers
Loss of teeth represented but also had the greatest number
of cases)

journal of dentistry 38 (2010) 83–95

No significant increase in risk noted for hem.
cancers with increased loss of teeth

Table 8 – Studies linking breast cancer and periodontitis.

Author Year Population Criteria Results Type of study
19
Hiraki et al. 2001–2005 756 cases of 5240 cancer patients Loss of teeth No significant increase in risk of breast Case–control
and 10,480 control patients in Japan cancer with increased number of
missing teeth
Hujoel et al.31 1971–1975, 19 cancer deaths in 11,328 cohort Periodontal exam by Russell index Increased risk of breast cancer with Cohort
follow-up in United States (classified as periodontitis, gingivitis, patients with periodontitis (OR 1.32)
until 1992 no periodontitis, or edentulous)

91
92 journal of dentistry 38 (2010) 83–95

Type of study
Homann et al. showed that certain oral bacterial species had a
higher acetaldehyde production capacity including Streptococ-

Cohort

Cohort

Cohort

Cohort
cus salivarius, Corynebacterium, Stomatococcus, and alpha-hemo-
lysing Streptococcus.43
Other microorganisms have been studied as well as
potential carcinogenic agents. There is also some evidence
that some strains of candidiasis have been seen in higher

cancer deaths and increased missing teeth

frequency in oral cancer patients.44 Viruses may play a role in

No significant increase in risk between

this also. A suggestion has been made that increased period-

Significantly increased risk of overall

Increased risk of cancer deaths with

cancer for patients with a history of

patients with periodontitis (OR 1.55)
No significant association between

periodontal disease (OR 1.14) after

ontal disease may be associated with infection with cytome-
galovirus and/or Epstein-Barr Virus 1 with mixed results.45–47
cancer mortality and increased

EBV of course has been linked to cancer including lymphoma

Results

number of missing teeth

and nasopharyngeal carcinoma.39

controlling for smoking

4.2. Increase in systemic circulatory inflammatory
markers

This is the basis for the link between periodontal disease and
cardiovascular disease and diabetes.4 The presence of inflam-
matory cells and mediators such as chemokines, cytokines,
and prostaglandins associated with tumors represent the
main indicators.48 Myeloperoxidase and superoxide dismu-
index (classified as periodontitis,

tase help to regulate inflammation and are found to be

periodontal disease (validated
Periodontal exam by Russell

elevated in periodontitis, and polymorphisms of these genes

by radiographic measure of
gingivitis, no periodontitis,

Patient reported history of

have been associated with elevated pancreatic cancer risk.49

This dual condition result of genetic polymorphisms has been
Criteria

suggested for gastric cancer and periodontitis as well.3 Many

or edentulous)

cancers are linked to local chronic inflammation, including

Loss of teeth

Loss of teeth

inflammatory bowel disease and colon cancer, Hepatitis C

bone loss)

inflammation and liver carcinoma, and H. pylori-associated

ulcers and gastric cancer.39,48,50 Also, the immune response
mounted to a chronic periodontal infection has been proposed
as a potential carcinogenic etiologic factor.2 Also of interest is
the relationship between the pro-inflammatory expression of
5720 cancer cases out of 48,328

the receptor for advanced glycation end products (RAGE) and

884 cancer deaths of 11,328

549 cancer deaths of 12,223

esophageal, gastric, colon, biliary, pancreatic, and prostate

patient cohort in Scotland
68 cancer deaths of 1462

cancers.51 RAGE has been shown to play a role in the

cohort in United States

cohort in United States

Population
Table 9 – Studies linking overall cancer rates with periodontitis.

inflammatory processes of oral infections including period-

women in Sweden

ontal disease.52
There are several obstacles in accurately determining a
relationship between, periodontal disease and cancer.

4.3. Methodology

First, there are many potential statistical problems, involved

in the measurement of periodontal disease. Studies vary
1948–1968 and follow-up

between inclusion, criteria for patients with periodontal

1967–1969, follow-up

1971–1975, follow-up

disease—some rely on tooth loss as a marker and, others

until 1992–1993
Year

use patient history or clinical or radiographic examination as

criteria. The, difficulty in using tooth loss as a variable is that
1986–2004
until 1992

until 2005

teeth can be lost due either to caries, trauma, or periodontal

disease, thus it is difficult to remove caries as a confounding
factor in these relationships. A major limiting factor in
establishing a solid link between periodontitis and cancer is
that many of these studies exclusively used tooth loss as a
Michaud et al.2
37

measure of periodontal disease. Tu et al. point out in a study

Hujoel et al.31
Cabrera et al.

on overall cancer risk and tooth loss that in fact as the number
Tu et al.33
Author

of teeth that are lost increases, the exposure to periodontal

disease may decrease (as edentulous patients no longer have
active periodontal disease).33 This is supported in Guha et al.
 journal of dentistry 38 (2010) 83–95 93

where they found that if the missing teeth were over 15 in risk levels. This may include the possibility of a more frequent
number the association with esophageal cancer disap- periodontal recall schedule to keep the level of disease and
peared.18 However, some aspects of these studies support inflammation to an absolute minimum and stressing the
the differentiation of periodontal disease from general poor importance of maintaining regular dental visits and meticu-
oral health as an independent risk factor for cancer. A study in lous oral hygiene. In addition, the patients should be
Brazil comparing the oral health status of patients with encouraged to work with their other medical providers to
untreated oral and oropharyngeal squamous cell carcinoma comply with all appropriate schedules of screening examina-
with cancer-free patients showed a higher rate of periodontal tions for early detection of malignancy including frequent oral
disease in the cancer patients but similar rates of caries as cancer screenings in the dental health setting. Finally, these
measured by DMFT and oral hygiene habits.16 This is also patients should be encouraged and supported in making
supported by the Tezal et al. study in 2007 where other positive lifestyle changes that may reduce their cancer risk
indicators of oral health (i.e. DMFT) were not significantly levels such as tobacco cessation, maximal nutrition and
associated with tongue cancer while millimeters of alveolar exercise, and elimination of environmental risks. Dental
bone loss were.4 Even amongst studies that specifically health practitioners have a great opportunity and obligation
measured for periodontal disease, differences in measure- to assist in preventative medicine by helping their patients
ment criteria existed: periodontal disease was measured in obtain the best possible oral health.
varied ways including by clinical attachment loss,23 Russell
index31 (which categorized patients into periodontitis, gingi-
references
vitis, no periodontitis, and edentulous), and patient reported
history verified with radiographic bone loss.2 Of these
methods the most objective and thus most reliable would
be the direct clinical measurement of attachment loss. 1. Pihlstrom BL, Michalowicz BS, Johnson NW. Periodontal
diseases. Lancet 2005;366:1809–20.
2. Michaud DS, Liu Y, Meyer M, Giovannucci E, Joshipura K.
4.4. Confounding factors Periodontal disease, tooth loss, and cancer risk in male
health professionals: a prospective cohort study. The Lancet
Smoking appeared to be the main confounding factor among Oncology 2008;9:550–8.
these studies, especially for cancers strongly linked to tobacco 3. Abnet CC, Kamangar F, Dawsey SM, Stolzenberg-Solomon
use such as lung cancer. Other potential confounding factors RZ, Albanes D, Pietinen P, et al. Tooth loss is associated with
increased risk of gastric non-cardia adenocarcinoma in a
are socio-economic status, diabetes, age, gender and ethni-
cohort of Finnish smokers. Scandinavian Journal of
city,40 along with genetics.27 Dietary factors associated with
Gastroenterology 2005;40:681–7.
advanced tooth loss have also been proposed as a potential 4. Tezal M, Sullivan MA, Reid ME, Marshall JR, Hyland A, Loree
confounding factor as a lack of fruit and vegetable intake has T, et al. Chronic periodontitis and the risk for tongue cancer.
been shown to be increased with a higher risk of some cancers Archives of Otolaryngology-Head & Neck Surgery 2007;133:
and patients who have lost many teeth may be less able to 450–4.
include these foods in their diets.40 5. Seoane J, Varela-Centelles PI, Walsh TF, Lopez-Cedrun JL,
Vazquez I. Gingival squamous cell carcinoma: diagnostic
delay or rapid invasion? Journal of Periodontology
2006;77:1229–33.
5. Conclusions 6. Gallagher Jr CS, Svirsky JA. Misdiagnosis of squamous cell
carcinoma as advanced periodontal disease. Report of a
In summary, the current literature shows that the strongest case. Journal of Oral Medicine 1984;39:35–8.
degree of evidence of association to periodontal disease is with 7. Levi Jr PA, Kim DM, Harsfield SL, Jacobson ER. Squamous cell
oral cancer. There is also suggestive positive evidence for carcinoma presenting as an endodontic-periodontic lesion.
Journal of Periodontology 2005;76:1798–804.
pancreatic cancer as well as overall cancer rates. Mixed results
8. Yoon TY, Bhattacharyya I, Katz J, Towle HJ, Islam MN.
appear to be present for esophageal, gastric, lung, prostate, Squamous cell carcinoma of the gingival presenting as
breast, hematologic and other cancers in regards to a link localized periodontal disease. Quintessence International
specifically to periodontal disease. 2007;38:97–102.
Additional studies are needed to explore these suggested 9. Kirkham DB, Hoge HW, Sadeghi EM. Gingival squamous cell
associations. Consistent and standardized criteria periodontal carcinoma appearing as a benign lesion: report of a case.
Journal of the American Dental Association 1985;111:767–9.
disease (such as millimeters of alveolar bone loss by clinical
10. Craig Jr RM, Vickers CA, Correll RW. Erythroplastic lesion on
examination) and consistent rigorous control of the con-
the mandibular marginal gingival. Journal of the American
founding factors such as smoking and alcohol use will be Dental Association 1989;119:543–4.
helpful in establishing a definitive link between periodontal 11. Heller AN, Klein A, Barocas A. Squamous cell carcinoma of
disease and cancer. the gingival presenting as an endoperiodontic lesion. Journal
Given the evidence to date, it would be beneficial to the of Periodontology 1991;62:573–5.
dental provider to be aware of the real potential that 12. Lee JJ, Cheng SJ, Lin SK, Chiang CP, Yu CH, Kok SH. Gingival
squamous cell carcinoma mimicking a dentoalveolar
periodontal disease may exist as a risk factor for many forms
abscess: report of a case. Journal of Endodontia 2007;33:
of malignancy. Thus it may be of real benefit to patients 177–80.
exhibiting periodontal disease who also have concurrent 13. Selden HS, Manhoff DT, Hatges NA, Michel RC. Metastatic
known risk factors for certain types of cancer such as lifestyle carcinoma to the mandible that mimicked pulpal/
factors or family history to consider several steps to reduce the periodontal disease. Journal of Endodontia 1998;24:267–70.
94 journal of dentistry 38 (2010) 83–95

14. Solomon MP, Biernacki J, Slippen M, Rosen Y. Parosteal 30. Demirer T, Icli F, Uzanalimoglu O, Kucuk O. Diet and
osteogenic sarcoma of the mandible. Existence masked by stomach cancer incidence. A case–control study in Turkey.
diffuse periodontal inflammation. Archives of Otolaryngology Cancer 1990;65:2344–8.
1975;101:754–60. 31. Hujoel PP, Drangsholt M, Spiekerman C, Weiss NS. An
15. Bundgaard T, Wildt J, Frydenberg M, Elbrond O, Nielsen JE. exploration of the periodontitis-cancer association. Annals
Case–control study of squamous cell cancer of the oral of Epidemiology 2003;13:312–6.
cavity in Denmark. Cancer Causes & Control 1995;6:57–67. 32. Watabe K, Nishi M, Miyake H, Hirata K. Lifestyle and gastric
16. de Rezende CP, Ramos MB, Daguila CH, Dedivitis RA, cancer: a case–control study. Oncology Reports 1998;5:1191–4.
Rapoport A. Oral health changes in patients with oral and 33. Tu YK, Galobardes B, Smith GD, McCarron P, Jeffreys M,
oropharyngeal cancer. Brazilian Journal of Otolarngology Gilthorpe MS. Associations between tooth loss and
2008;74:596–600. mortality patterns in the Glasgow Alumni Cohort. Heart
17. Fernandez-Garrote L, Herrero R, Ortiz-Reyes RM, Vaccarella 2007;93:1098–103.
S, Lence-Anta J, Ferbeye L, et al. Risk factors for cancer of the 34. Michaud DS, Joshipura K, Giovannucci E, Fuchs CS. A
oral cavity and oro-pharynx in Cuba. British Journal of Cancer prospective study of periodontal disease and pancreatic
2001;85:46–54. cancer in US male health professionals. Journal of the
18. Guha N, Boffetta P, Filho VW, Neto JE, Shangina O, Zaridze National Cancer Institute 2007;99:171–5.
D, et al. Oral health and risk of squamous cell carcinoma of 35. Stolzenberg-Solomon RZ, Dodd KW, Blaser MJ, Virtamo J,
the head and neck and esophagus: results of two Taylor PR, Albanes D. Tooth loss, pancreatic cancer, and
multicentric case–control studies. American Journal of Helicobacter pylori. The American Journal of Clinical Nutrition
Epidemiology 2007;166:1159–73. 2003;78:176–81.
19. Hiraki A, Matsuo K, Suzuki T, Kawase T, Tajima K. Teeth 36. Mackillop WJ, Zhang-Solomons J, Boyd CJ, Groome PA.
loss and risk of cancer at 14 common sites in Japanese. Associations between community income and cancer
Cancer Epidemiology Biomarkers & Prevention A Publication of the incidence in Canada and the United States. Cancer
American Association for Cancer Research cosponsored by the 2000;89:901–12.
American Society of Preventive Oncology 2008;17:1222–7. 37. Cabrera C, Hakeberg M, Ahlqwist M, Wedel H, Bjorkelund C,
20. Marshall JR, Graham S, Haughey BP, Shedd D, O’Shea R, Bengtsson C, et al. Can the relation between tooth loss and
Brasure J, et al. Smoking, alcohol, dentition and diet in the chronic disease be explained by socioeconomic status? A
epidemiology of oral cancer. European Journal of Cancer Part B 24-year follow-up from the population study of women in
Oral Oncology 1992;28B:9–15. Gothenburg, Sweden. European Journal of Epidemiology
21. Rosenquist K, Wennerberg J, Schildt EB, Bladstrom A, Goran- 2005;20:229–36.
Hansson B, Andersson G. Oral status, oral infections and 38. Mirvish SS. Role of N-nitroso compounds (NOV) and N-
some lifestyle factors as risk factors for oral and nitrosation in etiology of gastric, esophageal,
oropharyngeal squamous cell carcinoma. A population- nasopharyngeal and bladder cancer and contribution to
based case–control study in southern Sweden. Acta Oto- cancer of known exposures to NOC. Cancer Letters
laryngologica 2005;125:1327–36. 1995;93:17–48.
22. Talamini R, Vaccarella S, Barbone F, Tavani A, La Vecchia C, 39. Coussens LM, Werb Z. Inflammation and cancer. Nature
Herrero R, et al. Oral hygiene, dentition, sexual habits and 2002;420:860–7.
risk of oral cancer. British Journal of Cancer 2000;83:1238–42. 40. Meyer M, Joshipura K, Giovannucci E, Michaud DS. A review
23. Tezal M, Grossi SG, Genco RJ. Is periodontitis associated of the relationship between tooth loss, periodontal disease,
with oral neoplasms? Journal of Periodontology 2005;76:406– and cancer. Cancer Causes & Control 2008;19:895–907.
10. 41. Salaspuro MP. Acetaldehyde, microbes, and cancer of the
24. Zheng TZ, Boyle P, Hu HF, Duan J, Jian PJ, Ma DQ, et al. digestive tract. Critical Reviews of Clinical Laboratory Sciences
Dentition, oral hygiene, and risk of oral cancer: a case– 2003;40:183–208.
control study in Beijing, People’s Republic of China. Cancer 42. Homann N, Tillonen J, Rintamaki H, Salaspuro M, Lindqvist
Causes & Control 1990;1:235–41. C, Meurman JH. Poor dental status increases acetaldehyde
25. Abnet CC, Qiao YL, Mark SD, Dong ZW, Taylor PR, Dawsey production from ethanol in saliva: a possible link to
SM. Prospective study of tooth loss and incident esophageal increased oral cancer risk among heavy drinkers. Oral
and gastric cancers in China. Cancer Causes & Control Oncology 2001;37:153–8.
2001;12:847–54. 43. Homann N, Tillonen J, Meurman JH, Rintamaki H, Lindqvist
26. Abnet CC, Qiao YL, Dawsey SM, Dong ZW, Taylor PR, Mark C, Rautio M, et al. Increased salivary acetaldehyde levels in
SD. Tooth loss is associated with increased risk of total heavy drinkers and smokers: a microbiological approach to
death and death from upper gastrointestinal cancer, heart oral cavity cancer. Carcinogenesis 2000;21:663–8.
disease, and stroke in a Chinese population-based cohort. 44. Tillonen J, Homann N, Rautio M, Jousimies-Somer H,
International Journal of Epidemiology 2005;34:467–74. Salaspuro M. Role of yeasts in the salivary acetaldehyde
27. Abnet CC, Kamangar F, Islami F, Nasrollahzadeh D, Brennan production from ethanol among risk groups for ethanol-
P, Aghcheli K, et al. Tooth loss and lack of regular oral associated oral cavity cancer. Alcoholism Clinical and
hygiene are associated with higher risk of esophageal Experimental Research 1999;23:1409–15.
squamous cell carcionoma. Cancer Epidemiology Biomarkers & 45. Chalabi M, Moghim M, Mogharehabed A, Najafi F, Rezaie F.
Prevention A publication of the American Association for Cancer EBV and CMV in chronic periodontitis: a prevalence study.
Research cosponsored by the American Society of Preventive Archives of Virology 2008;153:1917–9.
Oncology 2008;17:3062–8. 46. Slots J, Kamma JJ, Sugar C. The herpesvirus – Porphyromonas
28. Sepehr A, Kamangar F, Fahimi S, Saidi F, Abnet CC, Dawsey gingivalis – periodontitis axis. Journal of Periodontal Research
SM. Poor oral health as a risk factor for esophageal 2003;38:318–23.
squamous dysplasia in northeastern Iran. Anticancer 47. Slots J, Sugar C, Kamma JJ. Cytomegalovirus periodontal
Research 2005;25:543–6. presence is associated with subgingival Dialister
29. Wei WQ, Abnet CC, Lu N, Roth MJ, Wang GQ, Dye BA, et al. pneumosintes and alveolar bone loss. Oral Microbiology and
Risk factors for oesophageal squamous dysplasia in adult Immunology 2002;17:369–74.
inhabitants of a high risk region of China. Gut 2005;54: 48. Mantovani A, Allavena P, Sica A, Balkwill F. Cancer-related
759–63. inflammation. Nature 2008;454:436–44.
 journal of dentistry 38 (2010) 83–95 95

49. Wheatley-Price P, Asomaning K, Reid A, Zhai R, Su L, Zhou glycation end products (RAGE) is related to prognosis
W, et al. Myeloperoxidase and superoxide dismutase for patients with esophageal squamous cell
polymorphisms are associated with an increased risk of carcinoma. Annals of Surgical Oncology 2009;16:
developing pancreatic adenocarcinoma. Cancer 440–6.
2008;112:1037–42. 52. Lalla E, Lamster IB, Stern DM, Schmidt AM. Receptor for
50. Karin M, Lawrence T, Nizet V. Innate immunity gone awry: advanced glycation end products, inflammation, and
linking microbial infections to chronic inflammation and accelerated periodontal disease in diabetes: mechanisms
cancer. Cell 2006;124:823–35. and insights into therapeutic modalities. Annals of
51. Tateno T, Ueno S, Hiwatashi K, Matsumoto M, Okumura Periodontology/The American Academy of Periodontology
H, Setoyama T, et al. Expression of receptor for advanced 2001;6:113–8.