Rostrum
Can oral pathogens influence allergic disease?
Samuel J. Arbes, Jr, DDS, MPH, PhD,a and Elizabeth C. Matsui, MD, MHSb
The hygiene hypothesis contends that fewer opportunities for
infections and microbial exposures have resulted in more
widespread asthma and atopic disease. Consistent with that
hypothesis, decreases in infectious oral diseases over the past
half century have coincided with increases in the prevalence of
asthma and other allergic diseases. This observation has led
some researchers to speculate that exposures to oral bacteria,
including pathogens associated with periodontal diseases, such
as gingivitis and periodontitis, might play a protective role in
the development of asthma and allergy. Colonization of the oral
cavity with bacteria, including some species of periodontal
pathogens, begins shortly after birth, and the detection of serum
antibodies to oral pathogens in early childhood provides
evidence of an early immune response to these bacteria. Current
knowledge of the immune response to oral bacteria and the
immunologic pathogenesis of periodontal diseases suggests
biologically plausible mechanisms by which oral pathogens
could influence the risk of allergic disease. However, studies
investigating the association between oral pathogen exposures
and allergic disease are few in number and limited by cross-
sectional or case-control design, exclusion of young children,
and use of surrogate measures of oral bacterial colonization.
Additional studies, particularly well-designed case-control
studies among very young children and prospective birth
cohort studies, are needed. (J Allergy Clin Immunol
2011;127:1119-27.)
Key words: Oral microflora, oral pathogens, gingivitis, periodonti-
tis, periodontal diseases, allergy, asthma, hay fever, allergic rhinitis,
hygiene hypothesis
Over the past half century, dental professionals and pharma-
ceutical companies have waged an increasingly successful war
against the oral pathogens that cause tooth decay, gingivitis, and
periodontitis, as evidenced by a steady decrease in rates of tooth
decay and periodontal disease in the United States and other
developed countries.1-4 However, researchers have recently be-
gun to ask whether achieving ‘‘axenic and inflammation-free
oral conditions’’ has had unintended consequences.5 An impetus
for raising this question has been the increasing popularity of the
From aRho, Inc, Chapel Hill, NC, and bJohns Hopkins University School of Medicine,
Baltimore, Md.
Disclosure of potential conflict of interest: S. J. Arbes is employed by Rho, Inc. E. C.
Matsui receives research support from the National Institutes of Health.
Received for publication November 17, 2010; revised March 11, 2011; accepted for pub-
lication March 15, 2011.
Reprint requests: Samuel J. Arbes, Jr, DDS, MPH, PhD, Rho, Inc, 6330 Quadrangle
Dr, Rho Building, Chapel Hill, NC 27517. E-mail: sam_arbes@rhoworld.com.
0091-6749/$36.00
Ó 2011 American Academy of Allergy, Asthma & Immunology
doi:10.1016/j.jaci.2011.03.023
Chapel Hill, NC, and Baltimore, Md
Abbreviations used
NHANES III: Third National Health and Nutrition Examination
Survey
OVA: Ovalbumin
Treg: Regulatory T
hygiene hypothesis, which contends that reduced opportunities
for infection in early childhood have resulted in more widespread
asthma and atopic disease. Originally proposed in 1989 by D. P.
Strachan to explain his observation of an inverse relationship be-
tween family size and the prevalence of hay fever,6 the hygiene
hypothesis has gained widespread attention among scientists
and has been extended to examine the effects of a variety of mi-
crobial exposures on the development of asthma and other atopic
diseases. These exposures include bacterial, viral, and parasitic
infections; living on farms; gut microflora; probiotics; vaccina-
tions; and oral pathogens. The purpose of this article is to review
the ecology of oral microflora, review published studies examin-
ing associations between oral pathogen–related exposures and
allergic disease, and make recommendations for future research.
ECOLOGY OF ORAL MICROFLORA
Approximately 280 bacterial species have been cultured from
the oral cavity and formally named, and studies using culture-
independent molecular methods suggest the existence of 500 to
700 common oral species.7 These indigenous oral bacteria typi-
cally have a commensal relationship with the host; however,
under certain circumstances, some oral bacteria overcome host
defenses and become pathogenic. Oral bacteria colonize on oral
mucosa and teeth and are present in saliva. Colonization begins
shortly after birth, and studies of some bacterial species have
shown that transmission occurs from parent to child and child
to child, most likely through contact with saliva and the sharing
of objects, such as cups or spoons.8-11 Oral pathogens associated
with periodontal disease, including 2 of the most studied patho-
gens, Porphyromonas gingivalis and Actinobacillus actinomyce-
temcomitans, are commonly found in the mouths of healthy
infants and children.12-14 In a study of 222 healthy Ohio children
aged 0 to 18 years, A actinomycetemcomitans and P gingivalis
were detected in 48% and 36% of the children, respectively,
and both species were detected in infants as young as 20 days
old.15 Serum IgM antibodies to P gingivalis (formerly called
Bacteroides gingivalis) have been detected in children less than
6 months of age, indicating a very early systemic response to
oral pathogens and perhaps an early opportunity for oral patho-
gens to influence the immune system with respect to allergic
disease.16
The species present in the oral cavity and their relative
proportions differ within subjects over time, between subjects
1119
1120 ARBES AND MATSUI
in a population, and between populations.17 Reasons for these dif-
ferences are not fully understood, although a reasonable assump-
tion is that patterns of colonization are influenced by the timing of
and opportunities for bacterial exposures, diet, oral hygiene prac-
tices, and interactions between the bacteria and host defenses.
Oral bacteria on teeth reside in a biofilm known as dental
plaque. Supragingival plaque (plaque at or above the gum line) is
associated with the development of tooth decay and gingivitis, a
reversible inflammatory condition of the gums. Subgingival
plaque (plaque below the gum line) is associated with periodon-
titis, an inflammatory disease characterized by the irreversible
destruction of the epithelium, connective tissue, and bone
supporting the teeth. Subgingival plaque resides in a gingival
crevice surrounding the tooth. In health that crevice is about 3 mm
deep; however, with periodontitis, which can affect a single tooth
or multiple teeth, the crevice deepens and widens with the
destruction of the periodontal tissues. Dental plaque begins to
form on cleaned tooth surfaces within a few hours, and in the
absence of further oral hygiene measures, the plaque increases in
mass and transitions from mostly gram-positive facultative spe-
cies to mostly gram-negative anaerobic species.18 Gingivitis is as-
sociated with a mixture of gram-positive and gram-negative
species, as well as facultative and anaerobic species. In contrast,
periodontitis is predominantly associated with gram-negative an-
aerobes.18 Dental hygiene practices, such as tooth brushing, floss-
ing, and professional dental cleaning, can remove dental plaque
and also influence its bacterial composition. The extensive use
of fluorides in developed countries has contributed much to the
decrease of tooth decay. Although fluoride makes tooth enamel
more resistant to demineralization, fluoride can also inhibit the
growth of many plaque microorganisms.18 Whether the wide-
spread use of fluorides has affected patterns of oral pathogen col-
onization at a population level is not known.
Gingivitis and periodontitis, along with other conditions of the
periodontium, are collectively known as periodontal diseases.
Gingivitis is more common among children, whereas periodon-
titis is more common among adults. In a US national survey, the
prevalence of gingival bleeding (an objective sign of gingivitis)
among children aged 13 to 17 years was 63%,19 and the preva-
lence of periodontitis among dentate adults aged 30 years or older
was estimated to be at least 35%.20 Risk factors for gingivitis in-
clude age, sex, and oral hygiene status; and risk factors for peri-
odontitis include age, sex, cigarette smoking, diabetes, and
socioeconomic status.19 When periodontitis occurs in susceptible
subjects, it starts at a young age, often in the teenage years.2 The
prevalence and severity of periodontitis increase with age, but
these increases are thought to reflect more of a cumulative effect
of disease progression than an increased susceptibility to dis-
ease.19 On the other hand, in most persons gingivitis does not pro-
gress to periodontitis.
In periodontally healthy and diseased mouths, pathogens in
subgingival plaque elicit both local and systemic immune
responses. In periodontally healthy mouths, intact gingival epi-
thelium and neutrophils and serum antibodies in the gingival
crevicular fluid keep pathogens in check.21 However, if conditions
in the gingival crevice favor the proliferation of pathogens, the re-
lease of proinflammatory cytokines and matrix metalloprotei-
nases by host cells can lead to inflammation and irreversible
destruction of periodontal tissues.21 Robust antibody responses
are mounted against oral pathogens in subjects with periodontitis,
as evidenced by high serum IgG titers to specific periodontal
J ALLERGY CLIN IMMUNOL
MAY 2011
pathogens in subjects with periodontitis, although low titers are
often seen in periodontally healthy children and adults, indicating
past or current exposure to the pathogens.16,21 In subjects with
periodontitis, the local bacteria, bacterial antigens, and inflamma-
tory cytokines can enter the circulation and trigger systemic in-
flammation.22 Markers of systemic inflammation associated
with periodontal disease include increased numbers of peripheral
leukocytes and increased levels of serum antibodies to the bacte-
ria, circulating proinflammatory cytokines, and acute-phase pro-
teins, such as C-reactive protein, fibrinogen, soluble CD14, and
LPS-binding protein.22 Markers of systemic inflammation are as-
sociated with cardiovascular disease, adverse pregnancy out-
comes, and diabetes mellitus, and increasing evidence suggests
that systemic inflammation caused by periodontitis might in-
crease a subject’s risk for these conditions.22
WHAT IS THE EVIDENCE?
Overview of selected studies
Human clinical studies published since 1990 and investigating
the association between oral bacteria or periodontal diseases and
allergic disease or asthma were identified in Medline. Because of
a purported link between dental caries and asthma medications,
the search excluded studies of dental caries unless the study also
examined periodontal disease. A large, although inconclusive,
literature has investigated the effects of asthma pharmacotherapy
on dental caries.23 Twelve studies were identified, reviewed, and
classified as being either supportive (Table I) or nonsupportive
(Table II) of a beneficial association between oral pathogens
and allergy-related outcomes.5,24-34
None of the 12 studies evaluated the association between
allergic disease and oral bacteria sampled directly from the
mouth: either periodontal disease (gingivitis, periodontitis, or
both) or serum antibodies were assessed as the pathogen-related
exposure. Each of the reviewed studies had either a cross-
sectional or case-control design. How the authors conceptualized
and analyzed the potential association between the oral heath and
allergy-related variables differed across the 12 studies. Five
studies (Hujoel et al,5 Friedrich et al,24 Arbes et al,25 Friedrich
et al,26 and von Hertzen et al27) mentioned the hygiene hypothesis
and investigated oral pathogens or periodontal disease as potential
mitigators of allergic diseases. Six studies (Laurikainen and Kuu-
sisto,28 McDerra et al,29 Shulman et al,30 Eloot et al,31 Mehta
et al,32 and Stensson et al33) mentioned the potential effects of
asthma or asthma treatments on oral health in the study’s intro-
duction and conceptualized asthma as a risk factor for oral health.
The remaining study (Grossi et al34) referenced the potential role
of systemic diseases in periodontal disease in the introduction and
also conceptualized asthma as a risk factor for periodontal dis-
ease. For each study, the summary tables indicate whether the
oral health variable was the independent (exposure or explana-
tory) or dependent (response) variable in analyses.
Supportive studies
Supportive studies are listed in Table I. The first report of an in-
verse relationship between periodontal disease and an allergy-
related outcome was a study by Grossi et al34 in 1994. This
cross-sectional study of adult residents in and around Erie County,
New York, evaluated a variety of potential risk factors for perio-
dontal disease and found that subject-reported allergy, but not
J ALLERGY CLIN IMMUNOL ARBES AND MATSUI 1121
VOLUME 127, NUMBER 5

TABLE I. Summary of human studies that support a beneficial association between oral pathogens or periodontal disease and allergy-
related outcomes
Oral pathogen
or periodontal Allergy or asthma
Study Design Participants disease variable(s) variable(s) Covariables Results

Grossi Cross-sectional d 1426 residents Periodontal disease Subject-reported ORs were adjusted for d In a final ordinal
et al,34 in and around severity defined as allergy and asthma/hay sex, race, education, logistic regression
1994 Erie County, mean periodontal fever (asthma income, SES, and oral model of perio-
New York attachment loss and and hay fever were hygiene status. dontal disease
d 25-74 y of age categorized as healthy, combined)  severity, allergy
low, moderate, high, was inversely as-
and severe* sociated with peri-
odontal disease
(OR, 0.77; 95% CI,
0.58-1.00).
d Asthma/hay fever
did not remain in
the final model.
Friedrich Cross-sectional d 2837 persons Periodontal disease Subject-reported hay ORs were adjusted for d Relative to perio-
et al,24 survey sampled from the extent defined as fever, house dust mite sex, age, education, dontal health, the
2006 adult population of the percentage of allergy, and asthma* smoking, alcohol odds of hay fever
northeast Germany periodontal sites with consumption, family and mite allergy,
d 20-59 y of age attachment loss >3 mm history for allergies or but not asthma,
and categorized as asthma, and number of decreased signifi-
healthy, low, teeth. cantly with in-
moderate, and severe  creasing periodon-
titis categories
(Ptrend 5 .01, .02,
and .11,
respectively).
d ORs (95% CIs)
comparing severe
periodontitis to
health were as fol-
lows: 0.53 (0.3-
0.9) for hay fever,
0.39 (0.2-0.9) for
mite allergy, and
0.48 (0.2-1.0) for
asthma.
Arbes Cross-sectional d 9385 persons Serum IgG Subject-reported, ORs were adjusted d For a 1-log-unit
et al,25 survey surveyed in concentrations doctor-diagnosed for age, sex, race- increase in P gin-
2006 NHANES III (continuous, log- asthma, wheeze, ethnicity, education, givalis antibody
d 12-90 y of age transformed) to 2 oral and hay fever* census region, concentration,
pathogens: P gingivalis urbanization, serum ORs (95% CI)
and A cotinine level, and were as follows:
actinomycetemcomitans  body mass index. 0.41 (0.20-0.87)
for asthma, 0.43
(0.23-0.78) for
wheeze, and 0.45
(0.23-0.93) for
hay fever.
d For A actinomyce-
temcomitans,
ORs (95% CIs)
were as follows:
0.56 (0.19-1.72)
for asthma, 0.39
(0.17-0.86) for
wheeze, and 0.48
(0.23-1.03) for
hay fever.

(Continued)
1122 ARBES AND MATSUI J ALLERGY CLIN IMMUNOL
MAY 2011

TABLE I. (Continued)
Oral pathogen
or periodontal Allergy or asthma
Study Design Participants disease variable(s) variable(s) Covariables Results
Friedrich Cross-sectional d 170 type 1 Periodontal disease Subject-reported hay ORs were adjusted Relative to periodontal
et al,26 diabetics in northeast extent defined as fever, house dust mite for sex, age, smoking, health, the odds of
2008 Germany the percentage of allergy, and asthma and diabetes duration respiratory allergies
d 17-80 y of age periodontal sites with dichotomized as (education, alcohol decreased with
attachment loss >3 mm respiratory allergies, consumption, family increasing periodontitis
and categorized as yes or no* history of allergies, and categories (Ptrend < .01).
healthy, low, moderate, hemoglobin A1c levels ORs (95% CIs) were as
and severe  did not remain in the follows: 0.30 (0.08-
final model). 1.07) for mild, 0.12
(0.02-0.60) for
moderate, and 0.05
(0.01-0.35) for severe.
Hujoel Cross-sectional d 12,631 persons Periodontal disease Number of years RRs were adjusted Relative to persons with
et al,5 review of 7 y with periodontal severity categorized during which at least for age and sex. early periodontitis,
2008 of HMO data disease (assessed as early, moderate, 1 prescription for a persons with advanced
(1996-2002) between 1999 and advanced (no therapy group was periodontitis had
and 2002) periodontally healthy filled (fill rate); significantly lower fill
d Health care coverage persons were included)  respiratory agents rates for respiratory
provided by the were one of 16 broad agents. RRs (95% CIs)
HMO for all 7 y therapy groups.* were as follows: 0.57
d 45-61 y of age (0.41-0.81) for
antihistamines, 0.46
(0.34-0.61) for
decongestants, 0.67
(0.55-0.81) for cold/
cough agents, and 0.68
(0.49-0.96) for
antiasthmatics.

OR, Odds ratio; RR, relative risk; SES, socioeconomic status.

*Dependent (response) variable(s) in statistical analyses.
 Independent (exposure or predictor) variable(s) in statistical analyses.

asthma and hay fever (a composite variable), was significantly
and inversely associated with periodontal disease severity. In their
discussion of the allergy finding, the authors, who were investigat-
ing asthma and allergy as risk factors for periodontal disease,
speculated on a possible role of allergy medication in the inverse
association, but they did not discuss the possibility that periodon-
tal disease could be a protective factor for asthma and allergy. Not
until later in that decade and the beginning of the next, when the
hygiene hypothesis became well known, did researchers begin to
conceptualize oral pathogen exposure as a potential protective
factor.
The 4 remaining supportive studies were conducted from the
perspective of the hygiene hypothesis. In 2006, Friedrich et al,24
who hypothesized that colonization by periodontal pathogens
might protect against allergic disease, reported on a population-
based survey of adult residents in the northeast of Germany. Be-
fore and after adjustment for confounders, periodontal disease
was inversely and significantly associated with self-reported
house dust mite allergy and hay fever. Asthma showed an inverse
but nonsignificant association.
The same group of investigators conducted a second cross-
sectional study limited to subjects with type 1 diabetes mellitus.26
That study showed an inverse association between periodontal
disease and the risk of respiratory allergies (a composite variable
of self-reported outcomes of dust mite allergy, hay fever, and/or
asthma) among diabetic subjects.
Arbes et al25 investigated associations between serum IgG an-
tibody levels to A actinomycetemcomitans and P gingivalis and
subject-reported asthma, wheeze, and hay fever among a repre-
sentative sample of the US population aged 12 years and older.
The authors reported inverse associations between P gingivalis
and the outcomes asthma, wheeze, and hay fever and between
A actinomycetemcomitans and wheeze but not asthma or hay fe-
ver. Although the authors did not provide an explanation for the
difference in results by pathogen, they stated that ‘‘A actinomyce-
temcomitans has been implicated in an aggressive form of perio-
dontal disease, whereas P gingivalis has been implicated in a
chronic form, though both forms of periodontal disease are con-
sidered mixed bacterial diseases.’’ That statement suggests a dif-
ference in virulence between or host immune response to the
pathogens.
The most recent of the supportive studies, Hujoel et al,5 was an
analysis of 7 years of data for 12,631 adults with periodontal dis-
ease who had medical, dental, and pharmaceutical coverage by
the Kaiser Permanente Northwest health maintenance organiza-
tion. The objective of the study was to relate patterns of medica-
tion use to the severity of periodontal disease. A primary finding
was that persons with more severe periodontitis had significantly
lower prescription fill rates for antihistamines, decongestants,
cough or cold medications, and asthma medications. In their arti-
cle the authors discussed the consistency of their findings with the
hygiene hypothesis, while acknowledging other potential
J ALLERGY CLIN IMMUNOL ARBES AND MATSUI 1123
VOLUME 127, NUMBER 5

TABLE II. Summary of human studies that do not support a beneficial association between oral pathogens or periodontal disease and
allergy-related outcomes
Oral pathogen
or periodontal Allergy or asthma
Study Design Participants disease variable(s) variable(s) Covariables Results

Laurikainen Case-control d 33 asthmatic Periodontal status Diagnosed asthma* d Subjects were Asthmatic subjects had
and and 33 nonasth- index (percentage matched on a higher mean
Kuusisto,28 matic adults in of teeth with gingival age and sex. periodontal status
1998 Finland bleeding, calculus, d Education was index than
d 25-50 y of age or periodontal a covariate in nonasthmatic
pockets)  the analysis. subjects (P < .05).
McDerra Case-control d 100 asthmatic d Gingivitis score Asthma (doctor- d Subjects were
For both age groups,
d

et al,29 and 149 nonasth- d Dental plaque diagnosed and matched on asthmatic subjects had
1998 matic children score  currently using an age, sex, race,
higher gingivitis
in England inhaler for asthma and SES. scores than nonasth-
d 4-16 y of age control)* d Analyses were
matic subjects
stratified by
(P < .01).
d For the younger group
age (4-10 y
and 11-16 y)only, asthmatic sub-
jects had higher pla-
and by 6 areas
que scores than
of the mouth.
nonasthmatic subjects
(P < .05).
Shulman Cross-sectional d 1596 d Gingivitis (bleeding, d Parent-reported, ORs were adjusted In separate logistic
et al,30 survey adolescents yes/no) doctor-diagnosed for sex, race, regression models
2003 surveyed in d Periodontitis (2 out- asthma categorized income, medications, for each periodontal
NHANES III comes): loss of attach- as healthy, mild, tobacco exposure, disease outcome,
d 13-17 y of age ment (yes/no) and and moderate and dental neither asthma nor
periodontal pockets or severe examination in past asthma medication
(yes/no)* d Asthma medication year. use was significant
use  (odds ratios were not
reported in the
article).
Eloot et al,31 Cross-sectional, d 140 asthmatic d Dental plaque index Duration of
d None (nonparametric No significant
2004 case only children in d Gingivitis (bleeding asthma correlation analyses correlations between
Belgium index)* d Severity of asthma were conducted; the oral health
d 3-17 y of age d Exposure time logistic regression measures and asthma
to medication  analyses modeled duration, asthma
dental caries but not severity, or exposure
gingivitis or plaque). time to medication
were found.
von Hertzen Cross-sectional d 790 Finns Seropositivity (yes/no) Atopy (positive d ORs were adjusted d Adjusted ORs (95%

et al,27 survey sampled from to antibodies against skin prick test for age and sex. CIs) for P gingivalis
2006 North Karelia, 7 pathogens, including response to at d Analyses were strati- were as follows: 0.95
Finland P gingivalis and A least 1 of 11 fied by country. (0.65-1.39) for Finns
d 387 Russians sam- actinomycetemcomitans  common airborne and 1.44 (0.79-2.64)
pled from the Re- allergens, yes vs for Russians.
public of Karelia, no)* d Adjusted ORs (95%

Russia CIs) for A actinomy-

d 25-54 y of age cetemcomitans were
as follows: 0.76 (0.54-
1.06) for Finns and
1.22 (0.69-2.13) for
Russians.
Mehta Case-control d 80 asthmatic d Dental plaque index Asthma defined d Subjects were Asthmatic subjects had
et al,32 and 80 nonasth- d Gingivitis index  as currently matched on age, higher mean plaque
2009 matic subjects on asthma sex, and SES. and gingivitis scores
in Mangalore, medication* than nonasthmatic
India subjects (t test, P <
d 11-25 y of age .001 for each test).
(Continued)
1124 ARBES AND MATSUI
TABLE II. (Continued)
Study Design Participants
Stensson Case-control with d 70 asthmatic d Gingivitis (bleeding, Diagnosed
et al,33 assessments at and 70 nonasth-
2010 ages 3 and 6 y matic children d Dental plaque on more
in Sweden
d 3 y of age at
study entry d Saliva sampling for
OR, Odds ratio; RR, relative risk; SES, socioeconomic status.
*Dependent (response) variable(s) in statistical analyses.
 Independent (exposure or predictor) variable(s) in statistical analyses.
explanations. In concluding their article, Hujoel et al stated that
the findings ‘‘indicate the need to evaluate the potentially adverse
effects of increasingly sophisticated mechanical and anti-
infective lifelong daily oral hygiene routines in the same fashion
as any other intervention.’’
These 5 supportive studies were limited by their cross-sectional
designs, making the directionality of the association unclear. In
addition, none of the studies investigated young children. Early-
life exposures are thought to be the most relevant exposures in the
hygiene hypothesis. Other limitations to these studies included
the use of either periodontal disease or serum antibodies as a
surrogate for oral pathogen exposure and the reliance on either
subject-reported allergy-related outcomes or, in the case of the
study by Hujoel et al,5 prescription use rates. The main strength of
the studies, with the exception of the 2008 study by Friedrich
et al,26 was the large sample size.
Nonsupportive studies
Nonsupportive studies are listed in Table II. Four of the nonsup-
portive studies reported a positive (direct or in the same direction)
association between asthma and periodontal disease, whereas 3
reported a null association.
The 4 positive studies, Laurikainen and Kuusisto,28 McDerra
et al,29 Mehta et al,32 and Stensson et al,33 were case-control stud-
ies that evaluated periodontal disease (gingivitis in 3 studies and a
periodontal disease score in the fourth) in asthmatic and nonasth-
matic subjects. Two of the studies, McDerra et al29 and Stensson
et al,33 included young children. McDerra et al29 compared gingi-
vitis and plaque scores between asthmatic children aged 4 to 16
years and nonasthmatic children matched on age, sex, race, and
socioeconomic status. The authors reported that asthmatic chil-
dren had more plaque and gingivitis than nonasthmatic children.
Stensson et al,33 compared the prevalence of gingivitis and plaque
in asthmatic children at ages 3 and 6 years with that in nonasth-
matic children matched on age and sex. The authors found that
gingivitis was more prevalent among asthmatic children at age
3 years but not at age 6 years and that plaque measures did not dif-
fer between groups. Three of the articles briefly discussed
J ALLERGY CLIN IMMUNOL
MAY 2011
Oral pathogen
or periodontal Allergy or asthma
disease variable(s) variable(s) Covariables Results
d Subjects were d Asthmatic subjects
yes/no) asthma* matched on age were more likely than
and sex nonasthmatic subjects
than 4 tooth surfaces d Separate analyses to have gingivitis at
(yes/no) were conducted at age 3 y (x2 test, P 5
ages 3 and 6 y .005) but not at age 6
cariogenic bacteria y (P 5 .18).
(detectable, yes/no)  d The prevalence of
plaque and detectable
cariogenic bacteria
did not significantly
differ between groups
at either age.
possible explanations for the positive associations between peri-
odontal disease and asthma. Laurikainen and Kuusisto28 attrib-
uted the positive association to similar inflammatory processes
between the 2 diseases, and McDerra et al29 and Mehta et al32
attributed the association to immunologic processes and mouth-
breathing habits in asthmatic subjects, which, according to the
authors, can cause dehydration of the mouth and gingivitis. Al-
though Stensson et al33 did not address the reason for the associ-
ation, they did discuss more frequent mouth breathing and higher
consumption of sugary drinks as an explanation for their finding
of more dental caries among asthmatic children.
The 3 studies that reported a null association, Shulman et al,30
Eloot et al,31 and von Hertzen et al,27 were cross-sectional studies.
Shulman et al30 analyzed data from the Third National Health and
Nutrition Examination Survey (NHANES III) for adolescents
aged 13 to 17 years, and in that analysis neither asthma nor the cu-
mulative use of antiasthma drugs was significantly associated
with gingivitis or periodontitis. This is in contrast to the study
by Arbes et al,25 who, using the NHANES III database, found
an inverse association between serum antibody levels to P gingi-
valis and asthma (but not between A actinomycetemcomitans and
asthma). A potential explanation for the difference in findings be-
tween the 2 studies is that persons without periodontal disease can
have increased serum antibody levels to oral pathogens and not
everyone with periodontal disease has increased serum antibody
levels to the 2 oral pathogens measured in NHANES III. An anal-
ysis of adults 40 years and older in NHANES III indicated that
increased serum antibody titers to P gingivalis and A actinomyce-
temcomitans, respectively, were seen in 16.1% and 24.3% of
persons with periodontal disease and 9.8% and 9.7% of persons
without periodontal disease.35 The study by von Hertzen et al27
examined serum antibodies to 7 pathogens, including 2 periodon-
tal pathogens, in an attempt to explain differences in the
prevalence of atopy in eastern Finland and western Russia, 2
neighboring areas with profound differences in living conditions
and lifestyles. In adjusted statistical models neither of the 2 oral
pathogens was associated with atopy; however, Helicobacter
pylori and atopy showed a significant inverse association. In the
study by Eloot et al,31 which was a case-only investigation of
J ALLERGY CLIN IMMUNOL
VOLUME 127, NUMBER 5
asthmatic subjects, gingivitis was not significantly associated
with asthma duration, asthma severity, or asthma medication use.
DISCUSSION
Of the 12 reviewed studies that examined the association
between oral pathogens and allergy-related outcomes, 5 reported
an inverse association, 3 reported a null association, and 4
reported a positive association. Each of the studies used either a
cross-sectional or case-control design, only 3 included young
children, and none investigated the association between an
allergy-related outcome and bacteria measured directly by means
of oral sampling. In studies evaluating the effects of oral bacteria
on allergic disease, periodontal disease might be a poor surrogate
for oral pathogen colonization because oral pathogens are some-
times present in healthy mouths and can stimulate local and
systemic immune responses in the absence of periodontal disease.
Although all of the studies controlled for confounding to some
degree, either through matching or statistical analyses (Tables I
and II), control of confounding might have been incomplete in
these studies. For example, none of the studies controlled for
diet, which could potentially influence both oral health and al-
lergy or asthma, and only 1 study controlled for dental care.
The results of this review, coupled with the lack of information
from longitudinal studies, studies of young children, and studies
measuring oral pathogens directly, do not allow any conclusions
to be made about the presence or nature of an association between
oral pathogens and allergic disease. However, local and systemic
immune responses to oral bacteria provide biologically plausible
mechanisms by which oral bacteria could affect allergic disease.
Those mechanisms are discussed in the remainder of this article.
Two biological mechanisms have been proposed to explain the
hygiene hypothesis. The first was that reduced early-childhood
exposures to microbes prevent a shifting from a TH2 to a TH1 phe-
notype.36 A TH2 pattern of cytokine release, which predominates
at birth, is involved in immune responses to parasitic infections
and allergic disease, whereas a TH1 pattern is involved in cellular
immune responses to bacterial and viral pathogens and autoim-
mune diseases. However, the discovery of regulatory T (Treg)
cells and observations contradictory to the proposed TH1/TH2
shift mechanism, such as the concomitant increase in the preva-
lence of both TH1- and TH2-mediated diseases in developed coun-
tries, led to a second proposed mechanism: that reduced microbial
exposure leads to reduced TH cell suppression by Treg cells.36 In
either paradigm, however, the shift away from an allergic immune
phenotype is thought to be mediated by stimulation of the innate
immune system, which in turn can lead to a predominant TH1 phe-
notype or perhaps upregulation of Treg cells.37 Because the innate
immune system is the first line of defense against oral microbes
and oral microbes can cause systemic inflammation, it is plausible
that oral pathogens can influence the phenotype of the adaptive
immune response to nonoral pathogen antigens. Should this be
the mechanism by which oral microbes influence the phenotype
of the immune response, differences in the distribution of poly-
morphisms in innate immunity genes among the study popula-
tions could explain some of the apparent discrepancies in
findings among the studies we reviewed.
In the innate response oral microbial pathogen-associated
molecular patterns bind to pattern-recognition receptors on host
cells, including dendritic cells, which then activate the inflam-
matory response with a release of proinflammatory cytokines,
ARBES AND MATSUI 1125
such as IL-1b, IL-6, IL-12, and TNF-a.22,38 Initially, only gingi-
vitis might develop, and if the immune response is successful, the
gingivitis will resolve; however, if the infection persists and fur-
ther proliferation of the bacteria occurs, intensification of the
inflammatory response can lead to the destruction of periodontal
tissue in susceptible subjects.22,38 Although control of the
immune response in patients with periodontal disease is not fully
understood, CD41CD251 Treg cells have been found in both gin-
givitis and periodontitis lesions, and there is speculation that Treg
cells are the mechanism by which immune responses to oral
pathogens are controlled and periodontal tissue destruction is
prevented.39 Whether the innate immune response to oral bacteria
or the stimulation of Treg cells in periodontal tissues affect the
TH1/TH2 balance or downregulate TH2 cells in a way that would
influence allergic disease is not known, but either mechanism
appears plausible.
Within the context of the hygiene hypothesis, the most relevant
microbial exposures are thought to occur during early childhood,
and as described previously in this article, evidence suggests that
colonization by oral bacteria, including periodontal pathogens,
begins soon after birth. Moreover, some researchers of the
hygiene hypothesis have speculated that mechanisms acting in
utero might be important.37 A significant body of literature has
investigated the role of maternal periodontal disease on birth out-
comes. In a study of 640 human umbilical cord blood samples,
Boggess et al40 found that 35.2% of the samples were IgM posi-
tive to at least 1 oral pathogen, and 26.6% were positive to more
than 1 oral pathogen. Fetal exposure to oral pathogens, as evi-
denced by a pathogen-specific IgM response, was associated
with preterm birth, and the risk for preterm birth was greatest
among fetuses that also demonstrated an inflammatory response
measured by C-reactive protein, 8-isoprostane, or prostaglandin
E2, although it was not clear from the study that these inflamma-
tory mediators were the result of the oral pathogen exposures. The
authors speculated that fetal exposure to maternal oral pathogens
or their bacterial products occurs as the result of systemic dissem-
ination of the organisms during periods of periodontal disease ex-
acerbation. It is conceivable that the fetal inflammatory response
to maternal oral pathogens could promote a more complete TH2 to
TH1 shift in infancy. Although the hygiene hypothesis is typically
framed around early-life exposures, and certainly early childhood
and birth cohort studies are critical for addressing the proposed
hypothesis, adults do have incident allergic disease, as is the
case with some occupational allergies, and it is conceivable that
oral pathogen exposures in adulthood could influence adult onset
of allergy.
In 2010, a report of an animal study investigating whether
infection with the oral pathogen P gingivalis could modify aller-
gic airway hyperresponsiveness and inflammation was pub-
lished.41 Infections with P gingivalis were established in a
subcutaneous chamber implanted on the backs of mice either
before or after sensitization to ovalbumin (OVA). P gingivalis
does not naturally colonize in the mouse, and the subcutaneous
chamber model provides a reproducible, low-level systemic chal-
lenge intended to mimic a naturally occurring oral infection. After
subsequent challenges with OVA, the mice were assessed for air-
way responsiveness to methacholine, serum levels of total IgE and
OVA-specific IgE, and measures of inflammation in bronchoalve-
olar fluid, serum, and lung tissue. The authors reported that P gin-
givalis infection after sensitization to OVA significantly reduced
airway responsiveness but had no effect on inflammation. In
1126 ARBES AND MATSUI
contrast, infection before sensitization significantly reduced air-
way inflammation but had no effect on airway responsiveness. Al-
though the discordance between airway responsiveness and
inflammation in this study is hard to reconcile, the study’s find-
ings support the notion that infection with a periodontal pathogen
can protect against airway responsiveness and inflammation, de-
pending on the timing of infection.
FUTURE DIRECTIONS
The study of oral bacteria in the context of the hygiene hypothesis
is a relatively new area of research, and more evidence of a
protective association between oral bacteria and allergy or asthma is
needed. The next step would be to conduct well-designed case-
control studies that examine associations between oral bacteria
sampled directly from the mouth and atopy and allergic disease
(allergic rhinitis, atopic dermatitis, food allergy, and asthma)
among very young children. If those studies supported an associ-
ation, then other questions could be investigated with prospective
cohort studies. These questions include the following: How early in
a child’s development does the oral microbial exposure have to
occur? Does in utero exposure to maternal oral microbes or micro-
bial products influence a child’s immune system? What microbial
species or groups of species are involved? Can colonization by
commensal oral bacteria influence allergy? Does gingivitis or
periodontitis have to be present, or is colonization sufficient?
What allergy-related outcomes are influenced: atopy, allergic rhini-
tis, atopic dermatitis, food allergy, or asthma? A prospective cohort
study, particularly a birth cohort study, could address most of these
questions and provide convincing evidence of an association, if one
exists. Although a cohort study is expensive and time consuming,
oral bacteria and serum antibody sampling and oral health examina-
tions could be added to a cohort study that is investigating other
hypotheses, making the cost less prohibitive.
Questions are often raised about the practical implications of
this research, and no one would suggest that individuals should be
given periodontal disease to protect them from asthma or allergy.
Future research might indicate that certain commensal oral
bacteria or groups of commensal bacteria, as opposed to perio-
dontal pathogens, can provide a beneficial stimulus to the immune
system. Even if pathogens are identified as protective agents, it
might be possible to colonize subjects with less pathogenic strains
(eg, some strains of A actinomycetemcomitans are known to be
less virulent than others42) or to expose them to immunostimula-
tory components of the bacteria if such components are identified.
In allergy research the administration of bacterial DNA immunos-
timulatory sequences containing CpG motifs is providing promis-
ing results. For example, immunostimulatory DNA combined
with ragweed allergen and given as a vaccine offered long-term
efficacy in a study of 25 adults with ragweed allergy.43 At the
least, further investigations of an association between oral patho-
gens and allergy-related disease would shed additional light on
the immunopathology of periodontal disease and asthma, 2 of
the most prevalent chronic inflammatory diseases.
We thank medical writer Patricia Edkins, MD (Rho, Inc, Chapel Hill, NC),
who reviewed drafts of the article and provided useful edits and comments.
REFERENCES
1. Petersen PE. Challenges to improvement of oral health in the 21st century—the
approach of the WHO Global Oral Health Programme. Int Dent J 2004;54:329-43.
J ALLERGY CLIN IMMUNOL
MAY 2011
2. Burt BA, Eklund SA. Dentistry, dental practice, and the community. 6th ed. St
Louis (MO): Elsevier Saunders; 2005.
3. Dye BA, Tan S, Smith V, Lewis BG, Barker LK, Thornton-Evans G, et al. Trends
in oral health status: United States, 1988-1994 and 1999-2004. Vital Health Stat 11
2007;(248):1-92.
4. Borrell LN, Burt BA, Taylor GW. Prevalence and trends in periodontitis in the
USA: the [corrected] NHANES, 1988 to 2000. J Dent Res 2005;84:924-30.
5. Hujoel PP, Cunha-Cruz J, Maupome G, Saver B. Long-term use of medications and
destructive periodontal disease. J Periodontol 2008;79:1330-8.
6. Strachan DP. Hay fever, hygiene, and household size. BMJ 1989;299:1259-60.
7. Dewhirst FE, Chen T, Izard J, Paster BJ, Tanner AC, Yu WH, et al. The human oral
microbiome. J Bacteriol 2010;192:5002-17.
8. Carletto-Korber FP, Gonzalez-Ittig RE, Jimenez MG, Cornejo LS. Initial acquisi-
tion and genetic identity of Streptococcus mutans of mother-child Pairs. Pediatr
Dent 2010;32:205-11.
9. Domejean S, Zhan L, DenBesten PK, Stamper J, Boyce WT, Featherstone JD. Hor-
izontal transmission of mutans streptococci in children. J Dent Res 2010;89:51-5.
10. Van Winkelhoff AJ, Boutaga K. Transmission of periodontal bacteria and models
of infection. J Clin Periodontol 2005;32(suppl 6):16-27.
11. Darby I, Curtis M. Microbiology of periodontal disease in children and young
adults. Periodontol 2000 2001;26:33-53.
12. Soncini JA, Kanasi E, Lu SC, Nunn ME, Henshaw MM, Tanner AC. Oral micro-
biota of children in a school-based dental clinic. Anaerobe 2010;16:278-82.
13. Kanasi E, Johansson I, Lu SC, Kressin NR, Nunn ME, Kent R Jr, et al. Microbial
risk markers for childhood caries in pediatricians’ offices. J Dent Res 2010;89:
378-83.
14. Tanner AC, Milgrom PM, Kent R Jr, Mokeem SA, Page RC, Riedy CA, et al. The
microbiota of young children from tooth and tongue samples. J Dent Res 2002;81:
53-7.
15. Lamell CW, Griffen AL, McClellan DL, Leys EJ. Acquisition and colonization sta-
bility of Actinobacillus actinomycetemcomitans and Porphyromonas gingivalis in
children. J Clin Microbiol 2000;38:1196-9.
16. Mouton C, Hammond PG, Slots J, Genco RJ. Serum antibodies to oral Bacteroides
asaccharolyticus (Bacteroides gingivalis): relationship to age and periodontal dis-
ease. Infect Immun 1981;31:182-92.
17. Haffajee AD, Japlit M, Bogren A, Kent RL Jr, Goodson JM, Socransky SS. Differ-
ences in the subgingival microbiota of Swedish and USA subjects who were peri-
odontally healthy or exhibited minimal periodontal disease. J Clin Periodontol
2005;32:33-9.
18. Haake SK, Newman MG, Nisengard RJ, Sanz M. Periodontal microbiology. In:
Newman MG, Takei HH, Carranza FA, editors. Carranza’s clinical periodontology.
Philadelphia: W.B. Saunders Co; 2002. p. 96-112.
19. Beck JD, Arbes SJ Jr. Epidemiology of gingival and periodontal diseases. In: New-
man MG, Takei HH, Carranza FA, editors. Carranza’s clinical periodontology.
Philadelphia: W.B. Saunders Co; 2002. p. 74-94.
20. Albandar JM, Brunelle JA, Kingman A. Destructive periodontal disease in adults
30 years of age and older in the United States, 1988-1994. J Periodontol 1999;
70:13-29.
21. Haake SK, Nisengard RJ, Newman MG, Miyasaki KT. Microbial interactions with
the host in periodontal diseases. In: Newman MG, Takei HH, Carranza FA, editors.
Carranza’s clinical periodontology. Philadelphia: W.B. Saunders Co; 2002. p.
132-52.
22. Madianos PN, Bobetsis YA, Van Dyke TE. Infection and Inflammation. In: Genco
RJ, Williams RC, editors. Periodontal disease and overall health: a clinician’s
guide. Yardley (PA): Professional Audience Communications, Inc; 2010. p. 24-41.
23. Maupome G, Shulman JD, Medina-Solis CE, Ladeinde O. Is there a relationship
between asthma and dental caries?: a critical review of the literature. J Am Dent
Assoc 2010;141:1061-74.
24. Friedrich N, Volzke H, Schwahn C, Kramer A, Junger M, Schafer T, et al. Inverse
association between periodontitis and respiratory allergies. Clin Exp Allergy 2006;
36:495-502.
25. Arbes SJ Jr, Sever ML, Vaughn B, Cohen EA, Zeldin DC. Oral pathogens and al-
lergic disease: results from the Third National Health and Nutrition Examination
Survey. J Allergy Clin Immunol 2006;118:1169-75.
26. Friedrich N, Kocher T, Wallaschofski H, Schwahn C, Ludemann J, Kerner W, et al.
Inverse association between periodontitis and respiratory allergies in patients with
type 1 diabetes mellitus. J Clin Periodontol 2008;35:305-10.
27. von Hertzen LC, Laatikainen T, Makela MJ, Jousilahti P, Kosunen TU, Petays T,
et al. Infectious burden as a determinant of atopy—a comparison between
adults in Finnish and Russian Karelia. Int Arch Allergy Immunol 2006;140:
89-95.
28. Laurikainen K, Kuusisto P. Comparison of the oral health status and salivary flow
rate of asthmatic patients with those of nonasthmatic adults—results of a pilot
study. Allergy 1998;53:316-9.
J ALLERGY CLIN IMMUNOL
VOLUME 127, NUMBER 5
29. McDerra EJ, Pollard MA, Curzon ME. The dental status of asthmatic British
school children. Pediatr Dent 1998;20:281-7.
30. Shulman JD, Nunn ME, Taylor SE, Rivera-Hidalgo F. The prevalence of
periodontal-related changes in adolescents with asthma: results of the Third Annual
National Health and Nutrition Examination Survey. Pediatr Dent 2003;25:279-84.
31. Eloot AK, Vanobbergen JN, De Baets F, Martens LC. Oral health and habits in
children with asthma related to severity and duration of condition. Eur J Paediatr
Dent 2004;5:210-5.
32. Mehta A, Sequeira PS, Sahoo RC, Kaur G. Is bronchial asthma a risk factor for
gingival diseases? A control study. N Y State Dent J 2009;75:44-6.
33. Stensson M, Wendt LK, Koch G, Nilsson M, Oldaeus G, Birkhed D. Oral health in
pre-school children with asthma—followed from 3 to 6 years. Int J Paediatr Dent
2010;20:165-72.
34. Grossi SG, Zambon JJ, Ho AW, Koch G, Dunford RG, Machtei EE, et al. Assess-
ment of risk for periodontal disease. I. Risk indicators for attachment loss.
J Periodontol 1994;65:260-7.
35. Dye BA, Choudhary K, Shea S, Papapanou PN. Serum antibodies to periodontal
pathogens and markers of systemic inflammation. J Clin Periodontol 2005;32:
1189-99.
36. Romagnani S. Coming back to a missing immune deviation as the main explana-
tory mechanism for the hygiene hypothesis. J Allergy Clin Immunol 2007;119:
1511-3.
ARBES AND MATSUI 1127
37. Schaub B, Lauener R, von Mutius E. The many faces of the hygiene hypothesis.
J Allergy Clin Immunol 2006;117:969-78.
38. Gu Y, Ryan ME. Overview of periodontal disease: causes, pathogenesis, and char-
acteristics. In: Genco RJ, Williams RC, editors. Periodontal disease and overall
health: a clinician’s guide. Yardley (PA): Professional Audience Communications,
Inc; 2010. p. 5-23.
39. Nakajima T, Ueki-Maruyama K, Oda T, Ohsawa Y, Ito H, Seymour GJ, et al.
Regulatory T-cells infiltrate periodontal disease tissues. J Dent Res 2005;84:
639-43.
40. Boggess KA, Moss K, Madianos P, Murtha AP, Beck J, Offenbacher S. Fetal im-
mune response to oral pathogens and risk of preterm birth. Am J Obstet Gynecol
2005;193:1121-6.
41. Card JW, Carey MA, Voltz JW, Bradbury JA, Ferguson CD, Cohen EA, et al. Mod-
ulation of allergic airway inflammation by the oral pathogen Porphyromonas gin-
givalis. Infect Immun 2010;78:2488-96.
42. Haake SK, Huang GTJ. Molecular biology of the host-microbe interaction in
periodontal diseases: selected topics. In: Newman MG, Takei HH, Carranza FA,
editors. Carranza’s clinical periodontology. Philadelphia: W.B. Saunders Co;
2002. p. 153-67.
43. Creticos PS, Schroeder JT, Hamilton RG, Balcer-Whaley SL, Khattignavong AP,
Lindblad R, et al. Immunotherapy with a ragweed-toll-like receptor 9 agonist vac-
cine for allergic rhinitis. N Engl J Med 2006;355:1445-55.