Acta Odontologica Scandinavica, 2013; 71: 553–559

ORIGINAL ARTICLE

Evaluation of periodontal pathogens in amniotic fluid and the role

of periodontal disease in pre-term birth and low birth weight

ESRA ERCAN1, KENAN ERATALAY2, OZGUR DEREN3, DENIZ GUR4,

OZGUR OZYUNCU3, BELGIN ALTUN4, CEYDA KANLI2, PINAR OZDEMIR5 &
HAKAN AKINCIBAY2
Acta Odontol Scand Downloaded from informahealthcare.com by University of Pennsylvania on 06/21/13

1
Department of Periodontology, Karadeniz Technical University, Turkey, 2Department of Periodontology, Hacettepe
University, Turkey, 3Maternal Fetal Medicine Unit, Department of Obstetrics and Gynecology, Hacettepe University
Faculty of Medicine, Turkey, 4Department of Internal Medicine, Infectional Disease Unit, Hacettepe University, Turkey,
and 5Department of Biostatistics, Hacettepe University Faculty of Medicine, Turkey

Abstract
Background. Pre-term birth and/or low birth weight (PTLBW) is a serious problem in developing countries. The absence of
known risk factors in ~ 50% of PTLBW cases has resulted in a continued search for other causes. The aim of this study was to
examine the effect of periodontitis on pregnancy outcomes. Methods. Samples were taken from 50 pregnant women who
underwent amniocentesis. Polymerase chain reaction was performed on amniotic fluid samples obtained during amniocentesis
For personal use only.

and on subgingival plaque samples to determine the presence of Porphyromonas gingivalis, Aggregatibacter actinomycetemco-
mitans, Treponema denticola, Tannerella forsythia, Fusobacterium nucleatum, Prevotella intermedia, Campylobacter rectus and
Eikenella corrodens. Plaque index, gingival index, bleeding on probing, probing depth and clinical attachment level were
evaluated. Medical records were obtained after birth. Results. Social and demographic variables were similar among the
Gingivitis (G), Localized Periodontitis (LP) and Generalized Periodontitis (GP) groups. Four subjects gave birth to PTLBW
neonates. Campylobacter rectus, T. forsythia, P. gingivalis and F. nucleatum were detected in the amniotic fluid and subgingival
plaque samples of three patients who gave birth to PTLBW neonates. The amniotic fluid sample from the fourth patient was
not positive for any of the tested pathogens. Conclusion. These findings suggest that the transmission of some periodontal
pathogens from the oral cavity of the mother may cause adverse pregnancy outcomes. The results contribute to an
understanding of the association between periodontal disease and PTLBW, but further studies are required to better clarify
the possible relationship.

Key Words: periodontitis, microbiology, pregnancy complications

Introduction neonatal mortality and long-term sequelae [3]. Factors

such as smoking, alcohol or drug use during preg-
Periodontitis can present a systemic inflammatory nancy, inadequate pre-natal care, race, low socioeco-
challenge because of the large epithelial surfaces that nomic status, hypertension, high or low maternal age,
can ulcerate in the periodontal pockets. Ulceration can diabetes and genitourinary tract infections increase the
allow bacteria and their products to reach other parts of risk for PTLBW [2]. The role of infection has gained
the body [1]. Cross-sectional, case-control and cohort importance in efforts to determine the etiology of
studies have indicated that periodontitis may confer a PTLBW. The strongest data for infection-induced
2-fold risk for cardiovascular disease and a 7-fold risk pre-term labor indicate an association between bac-
for premature birth [2]. terial vaginosis and premature birth [4]. Although the
Pre-term and/or low birth weight (PTLBW) is a etiology of ~ 50% of pre-term deliveries remains
major medical, social and economic problem and unknown, increasing evidence supports an effect of
accounts for a large proportion of maternal and non-genital maternal infections [5].

Correspondence: Dr Esra Ercan, Karadeniz Technical University, Department of Periodontology, 61080 Trabzon, Turkey. Fax: +90 462 3253017.
E-mail: [email protected]

(Received 20 October 2011; revised 3 April 2012; accepted 3 May 2012)

ISSN 0001-6357 print/ISSN 1502-3850 online  2013 Informa Healthcare
DOI: 10.3109/00016357.2012.697576
 554 E. Ercan et al.

Cross-sectional and prospective data, as well as the history and defined by ultrasonographic fetal biome-
results of animal experiments, continue to bolster the try between weeks 16–18. Full-term birth was defined
hypothesis that periodontal infection is an indepen- as giving birth in week 37 or later to a child weighing
dent risk factor for prematurity and growth restriction more than 2500 g.
[2,6–8]. Periodontal infections may mediate PTLBW The demographic characteristics of the partici-
through contamination of the fetoplacental unit by pants, such as age, marital status and educational
periodontal pathogens or through effects of inflam- status were obtained using a detailed questionnaire.
matory mediators on the fetoplacental unit. A third Gestational week and birth weight data were obtained
potential mechanism involves the effects of lipopoly- from medical records.
saccharides of periodontal pathogens [9]. Infection of During amniocentesis, 2-mL samples of amniotic
mice with Fusobacterium nucleatum, Campylobacter fluid were collected from each participant for micro-
rectus and Porphyromonas gingivalis has induced pre- biological examination. The samples were stored at
term delivery [8,10,11]. Porphyromonas gingivalis has 80
C until analysis.
also been isolated from amniotic fluid and subgingival
samples from pregnant women diagnosed with threa- Clinical periodontal measurements
Acta Odontol Scand Downloaded from informahealthcare.com by University of Pennsylvania on 06/21/13

tened or premature labour [12]. However, in a study

by Katz et al. [13], P. gingivalis was detected in the After the amniocentesis, all participants were invited
placenta in a case of normal birth. In another study, to the Department of Periodontology, Hacettepe
Dörtbudak et al. [14] examined the amniotic fluid University for periodontal examination and subgingi-
of 36 women at risk for pregnancy complications. val plaque sampling to be performed on the same day as
Bacteria were never found in the amniotic fluid the amniocentesis procedure. The following clinical
samples studied, but bacteria from subgingival plaque periodontal parameters were measured by a skilled
samples, including orange and red complex bacteria, clinician (EE) for all teeth except third molars: probing
were found in 18% of full-term and 100% of pre-term depth (PD), clinical attachment level (CAL), plaque
neonates. In a study by Leon et al. [12], 26 women index (PI) [15], gingival index (GI) [16] and bleeding
with threatened premature labor were examined and on probing (BOP) [17]. Six sites were examined on
amniotic fluid samples were collected. Microbial inva- each tooth: mesiobuccal, buccal, distobuccal, distolin-
For personal use only.

sion of the amniotic cavity was indicated in 30.8% gual, lingual and mesiolingual. PD and CAL measure-
of the samples based on P. gingivalis polymerase chain ments were recorded to the nearest millimeter using a
reaction (PCR), although bacterial cultures were manual periodontal probe (Hu-Friedy, Chicago, IL).
negative. Radiographic exposure was avoided because the
The aim of this study was to examine the microbes women were pregnant. Periodontitis was defined
present in amniotic fluid and subgingival plaque according to the extent of the disease [18]. Localized
samples from pregnant women, using PCR. periodontitis (LP) was defined as PD ‡4 mm and CAL
‡3 mm at the same site on two or three teeth; and
Materials and methods generalized periodontitis (GP), as PD ‡4 mm and CAL
‡3 mm at the same site on four or more teeth [18]. All
This study was approved by the Ethics Committee of teeth were examined with the exceptions of third
Hacettepe University, Ankara, Turkey. The study molars, incompletely erupted teeth, teeth in areas
population was drawn from women who underwent with extensive caries lesions, teeth with fractures or
amniocentesis due to double or triple screening test iatrogenically damaged restorations and teeth with
positivity. In total, 50 women undergoing amniocen- surfaces where the cement–enamel margins could
tesis at the Department of Obstetrics and Gynecology, not be distinguished. Periodontal treatment was
Hacettepe University between January and December performed during the post-partum period.
2009 consented to participate in the study. All
subjects were informed about the aim of the study Microbiological sampling and processing
and all provided written consent approved by the
Hacettepe University Ethics Committee. Amniocen- Subgingival bacterial samples were taken after the
tesis was performed by two gynaecologists (OD, OO) clinical measurements. A total of four periodontal sites
between 16–18 weeks of pregnancy. Women with with the deepest PDs, representing one site on each of
known congenital uterine or vaginal malformations, four different teeth in different quadrants, were chosen
fetal malformation, multiple gestation, a chronic for microbial sampling. Supragingival plaque was
disease (diabetes, hypertension, epilepsy, cardiac removed, the sample site was isolated from saliva
disease, lung disease, renal disease or a positive test and a sterile endodontic paper point was inserted in
for human immunodeficiency virus), a history of the periodontal pocket until resistance was felt. The
antibiotic use or periodontal treatment during preg- point was left in place for 20 s and then placed in a vial
nancy or fewer than 20 teeth were excluded. Gesta- containing 0.1 mL of Tris-EDTA buffer (1 M TE) for
tional age assessment was based on menstrual cycle storage at 80
C until analysis [19].
 The role of periodontitis in pregnancy 555

The patients were informed of their periodontal (Gel Logic 200; Eastman Kodak, New York, NY)
status and treatment requirements. Professional oral and photographed.
prophylaxis, including supragingival scaling, was
applied, but other treatments were postponed until Statistical analyses
after the patients gave birth.
The presence of eight periodontitis-related species The descriptive statistics of continuous variables are
(P. gingivalis, Aggregatibacter actinomycetemcomitans, expressed as means, standard deviations and median
Treponema denticola, Tannerella forsythia, F. nucleatum, values (interquartile ranges: IQR) and those of
Prevotella intermedia, C. rectus, Eikenella corrodens) was categorical variables are expressed as frequencies
determined by 16S ribosomal DNA (rDNA) ampli- and percentages. Inter-group differences of catego-
fication using the PCR method described by rical variables were analysed with x 2 tests. Welch’s
Slots et al. [20]. The samples were analyzed in the analysis of variance (ANOVA) and Kruskal-Wallis
microbiology laboratory of Hacettepe University tests were used to examine differences among groups
Children’s Hospital. in continuous variables. The groups were compared
DNA extraction was performed using the QIAGEN with independent-sample t-tests. Stepwise logistic
Acta Odontol Scand Downloaded from informahealthcare.com by University of Pennsylvania on 06/21/13

DNA mini-kit (QIAGEN, Hilden, Germany) accor- regression analysis was used to identify factors that
ding to the manufacturer’s instructions. The oligonu- may have caused the bacterial presence. The x2 test
cleotide primers used for amplification are shown was used to examine the frequency of PTLBW. Box
in Table I. Amplification reactions were performed plots were used to show differences among groups. All
in a final volume of 50 L, containing 1  reaction statistical analyses were performed using the PASW
buffer, 2.5 mM magnesium chloride, 0.2 mM each Statistics 18 software (SPSS Inc., Chicago, IL).
dNTP, 0.5 U of Taq DNA polymerase (MBI Fermen- Values of p < 0.05 were deemed to indicate statistical
tas, Burlington, ON, Canada), 30 pM each of the significance.
primers and 5 L of DNA sample. The amplification
reactions were performed in an automated thermal Results
cycler (Flexigene; Techne, Inc., Burlington, NJ) pro-
grammed for denaturation at 95
C for 1 min, followed Participants were grouped according to periodontal
For personal use only.

by 36 cycles of denaturation at 95
C for 30 s, annealing status: gingivitis (G; n = 16), localized periodontitis
at 60
C for 1 min and extension at 72
C for 1 min, with (LP; n = 10) and generalized periodontitis (GP;
a final extension at 72
C for 2 min. n = 24). Table II provides the sociodemographic
The PCR products (10 mL of each reaction mixture) variables and periodontal parameters of the partici-
were separated by electrophoresis in 1.5% agarose gels pants. A small proportion of women had experienced
with 1  Tris-acetic acid-EDTA (TAE) running previous pre-term births (6.3% in group G, 30% in
buffer, using x174 as molecular weight standards. After group LP, 8.3% in group GP). A small proportion
the gels were stained with ethidium bromide, the bands (12.5%) of women in group GP were current light
were visualized on an ultraviolet transilluminator smokers (<5 cigarettes/day).

Table I. Primer sequences.

Bacterium Primer sequence (5¢!3¢) Amplicon length

A. actinomycetemcomitans GCT AAT ACC GCG TAG AGT CGG 443 bp

ATT TCA CAC CTC ACT TAA AGG T
T. forsythia GCG TAT GTA ACC TGC CCG CA 641 bp
TGC TTC AGT GTC AGT TAT ACC T
C. rectus TTT CGG AGC GTA AAC TCC TTT TC 598 bp
TTT CTG CAA GCA GAC ACT CTT
F. nucleatum GAA GAA ACA AAT GAC GGT AAC AAC 705 bp
GTC ATC CCC ACC TTC CTC CT
E. corrodens CTA ATA CCG CAT ACG TCC TAA G 688 bp
CTA CTA AGC AAT CAA GTT GCC C
P. gingivalis AGG CAG CTT GCC ATA CTG CG 404 bp
ACT GTT AGC AAC TAC CGA TGT
P. intermedia AAC GGC ATT ATG TGC TTG CAC 589 bp
CTC AAG TCC GCC AGT TCG CG
T. denticola TAA TAC CGA ATG TGC TCA TTT ACA T 316 bp
TCA AAG AAG CAT TCC CTC TTC TTC TTA
 556 E. Ercan et al.

Table II. Sociodemographic variables and periodontal parameters of the participants.

Parameter G (n = 16) LP (n = 10) GP (n = 24) p-value

Age (years) (median–IQR) 35–8 35–11 34–8

Educational status:
Elementary 32.2% 40% 16.7%
High school 18.8% 10% 16.7%
University 50% 50% 66.6%
Previous pre-term birth 6.3% 30% 8.3%
Smoker 0% 0% 12.5%
PI (mean ± SD) 0.45 ± 0.34 0.79 ± 0.3 0.83 ± 0.4* < 0.05
GI (median–IQR) 0.35–0.23* 0.68–0.22 0.59–0.49 < 0.05
BOP (%) (median–IQR) 27.5–28** 45.5–39 48–55** < 0.05
PD (mm) (mean ± SD) 1.76 ± 0.24*** 2.03 ± 0.11*** 2.27 ± 0.36*** < 0.001
Acta Odontol Scand Downloaded from informahealthcare.com by University of Pennsylvania on 06/21/13

CAL (mm) (mean ± SD) 1.8 ± 0.24* 2.08 ± 0.15 2.29 ± 0.37 < 0.001

IQR, interquartile range; SD, standard deviation.

*significantly different from the other two groups; **significantly different from each other; ***all three groups differ significantly.

The mean PI values of groups G, LP and GP were and GP were 1.76 ± 0.24, 2.03 ± 0.11 and 2.27 ±
0.45 ± 0.34, 0.79 ± 0.3 and 0.83 ± 0.4, respectively. 0.36 mm, respectively. The PD values differed
The PI of group GP was significantly higher than significantly among all groups. The mean CAL values
those of groups G and LP (p < 0.05). The median GI of groups G, LP and GP were 1.8 ± 0.24, 2.08 ±
values of groups G, LP and GP were 0.35–0.23, 0.15 and 2.29 ± 0.3 mm, respectively. The CAL value
For personal use only.

0.68–0.22 and 0.59–0.49, respectively. The GI of of group G was significantly lower than those of
group G was significantly lower than those of the groups LP and GP.
other two groups (p < 0.05). The median BOP
percentages of groups G, LP and GP were 27.5–28, Microbiological assessment
39–45.5 and 48–55, respectively. The BOP percent-
age differed significantly between groups G and Table III presents the PCR positive determination
GP (p < 0.05). The mean PDs of groups G, LP rates of the eight periodontal pathogens in the

Table III. Frequencies (%) of bacteria in subgingival plaque (sp) and amniotic fluid (af) samples.

Bacterium G group LP group GP group p-value Total

A.a (sp) 31.3 50 54.2 > 0.05 46

A.a (af) 0 0 0 > 0.05 0
T. forsythia (sp) 43.8* 80 95.8 < 0.05 76
T. forsythia (af) 0 0 8.3 > 0.05 4
C. rectus (sp) 43.8* 80 83.3 < 0.05 70
C. rectus (af) 0 0 12.5 > 0.05 6
T. denticola (sp) 12.5 30 37.5 > 0.05 28
T. denticola (af) 0 0 0 > 0.05 0
P. intermedia (sp) 31.3 60 62.5 > 0.05 52
P. intermedia (af) 0 0 0 > 0.05 0
P. gingivalis (sp) 18.8* 70 62.5 < 0.05 50
P. gingivalis (af) 0 0 4.2 > 0.05 2
E. corrodens (sp) 31.3 50 54.2 > 0.05 46
E. corrodens (af) 0 0 0 > 0.05 0
F. nucleatum (sp) 50* 90 95.8 < 0.05 80
F. nucleatum (af) 0 0 8.3 > 0.05 4

*differs significantly from the other two groups.

 The role of periodontitis in pregnancy 557

subgingival dental plaque and amniotic fluid samples. neonates were all between 30–35 years old and
T. forsythia, C. rectus, P. gingivalis and F. nucleatum none were smokers. Several published studies support
were detected less frequently in the subgingival plaque the role of intra-amniotic infection in the pathogenesis
samples of the women in group G. These four patho- of pre-term labor [8,10–12]. Thus, the presence of
gens were also detected in the amniotic fluid samples bacteria in the amniotic fluid may be one reason for
of three women in group GP. None of the evaluated giving birth to a PTLBW neonate.
bacteria was found in the amniotic fluid samples of Previous investigations of periodontal disease and
women in groups G or LP. adverse pregnancy outcomes have studied popula-
We found a positive correlation between the detec- tions that differed from ours [6,7,23,24]. The study
tion of P. gingivalis and C. rectus in subgingival plaque populations of Offenbacher et al. [6] and Jeffcoat et al.
samples and GI in all three groups (p < 0.05). We also [7] consisted of younger women and included a
observed positive correlations between the detection higher proportion of subjects from black ethnic
of F. nucleatum in subgingival plaque samples and groups. The population in a case-control study by
BOP (p < 0.05) and between the detection of Moore et al. [25] differed from those of previous case-
A. actinomycetemcomitans and CAL (p < 0.05). control studies with respect to demographic factors
Acta Odontol Scand Downloaded from informahealthcare.com by University of Pennsylvania on 06/21/13

No PTLBW occurred in group G or LP, whereas and periodontal disease severity, as it included an
16.6% (n = 4) of births in group GP were PTLBW. increased proportion of Caucasian subjects with
Periodontal bacteria were detected in the amniotic high socioeconomic status. The characteristics of
fluid samples of three women whose neonates were the study population in Moore et al. may account
PTLBW. The microbiological profiles of the subgin- for the finding of no association between the severity
gival plaque and amniotic fluid samples from the of periodontal disease and pregnancy outcome due to
women who gave birth to PTLBW neonates are given those factors. Similarly, all of the participants in our
in Table IV. By the end of the study, four subjects had study were Caucasian and about half were university
given birth to PTLBW neonates. Periodontal patho- graduates. These characteristics of our study popula-
gens were found in the amniotic fluid samples from tion may explain the low PTLBW ratio and low
three of these women. The amniotic fluid sample periodontal disease severity in the present study.
from the fourth woman was not positive for any of In a case-control study using a periodontal disease
For personal use only.

the tested bacteria and, to our knowledge, the fourth definition similar to ours, Nabet et al. [26] found a
woman had none of the evaluated risk factors. The significant association between generalized periodon-
women who delivered term and normal weight babies titis and pre-term birth induced due to pre-eclampsia.
had no bacteria in their amniotic fluid. Their results correlate with ours in that all of the
women who gave birth to PTLBW neonates were
Discussion in the GP group.
Our periodontal findings differ from those of some
Several studies [11,12] have demonstrated an associ- previously reported studies. In a case control study
ation between maternal periodontal disease and investigating the possible link between periodontal
PTLBW, but the underlying mechanisms have not infection and PTLBW, Buduneli et al. [9] evaluated
been elucidated yet [10,12]. Additional studies are the periodontal status of 181 women within 3 days
needed to validate this association and to determine post-partum and found no significant difference in
whether it is causal. The present study was under- periodontal status between the case (PTLBW) and
taken to investigate this association in a sample of
50 pregnant women who underwent amniocentesis.
Table IV. The presence of periodontal pathogens in subgingival
The oral health status of our study population was plaque (sp) and amniotic fluid (af) samples of patients who gave
determined through structured clinical examinations. birth to PTLBW neonates.
No radiographic evaluation was performed because Patient Patient Patient Patient
of the patients’ pregnancies; nevertheless, accurate 1 2 3 4
periodontal diagnoses were made.
Several factors such as smoking, alcohol or drug Bacterium sp af sp af sp af sp af
use, older or younger maternal age and marital status
A. a + - - - + - + -
are related to the risk for PTLBW [21,22]. The
T. forsythia + + + - + + + -
patients in our study did not use alcohol or drugs
and did not have systemic illnesses that could affect C.rectus + + + + + + - -
the periodontium or gestational duration or could T. denticola - - + - - - - -
cause bacterial vaginosis or multiple gestations. All P. intermedia + - + - + - - -
of our participants were 20–44-year-old, married, P. gingivalis + - + - + + + -
Caucasian women with a high education level.
E. corrodens - - - - + - - -
Although the patients in our study represented a
F. nucleatum + + + - + + + -
wide age range, those who gave birth to PTLBW
 558 E. Ercan et al.

control (full-term birth) groups, which had mean PDs As one of the most prevalent species found in
of 3.45 and 3.20 mm, respectively. These PDs are amniotic fluid, F. nucleatum has been implicated as
higher than those in our patients (1.76, 2.03 and the sole infectious agent in pre-term labor. We
2.27 mm for groups G, LP and GP, respectively). detected F. nucleatum in 8.3% of the amniotic fluid
In Offenbacher et al. [6], the group with the healthiest samples from women with GP and PTLBW neonates.
periodontal status had a mean PD of 2.87 mm and the Fusobacterium nucleatum has been shown to be hema-
case group had a mean PD of 3.17 mm. Similarly, togenously transmitted to the placenta and to cause
Davenport et al. [23] reported mean PDs of 3.85 and adverse pregnancy outcomes [8].
3.72 mm for the control and case groups, respectively. In an animal study, Lin et al. [10] subcutaneously
Thus, the PD values in the present study are lower inoculated subjects with P. gingivalis to induce fetal
than those reported previously. growth retardation. At the end of the study period, the
Periodontal disease severity may be a factor that bacteria were detected in the liver and uterus. Our
contributes to adverse pregnancy outcome. Rakato- study detected P. gingivalis in a single amniotic fluid
Alson et al. [5] demonstrated a significant relationship sample, which was taken from a woman who gave
between periodontal disease severity and adverse birth to a PTLBW neonate. This finding is consistent
Acta Odontol Scand Downloaded from informahealthcare.com by University of Pennsylvania on 06/21/13

pregnancy outcomes in a Madagascan population. with the results of Lin et al.

They found significantly higher PI and papillary blee- In another animal study, Arce et al. [30] showed
ding index scores in the pre-term birth group compared that C. rectus can translocate from a distant site of
with the full-term birth group. Holbrook et al. [27] infection to the placenta and can induce fetal growth
reported no link between low-grade periodontal disease restriction. Human placental tissues have not been
and pre-term birth in a healthy Caucasian population. In reported to harbor C. rectus, despite evidence of
our study population, periodontal disease severity was maternal infection and fetal exposure, based on fetal
low among Caucasian women with a high education IgG response. Buduneli et al. [9] evaluated the
level and good healthcare. possible link between periodontal infection and
Another factor that may affect the reported associ- PTLBW by means of clinical and microbiological
ation between periodontal status and adverse preg- data in post-partum women with low socio-economic
nancy outcomes is the definition of periodontitis. In level. Their findings indicate that, when subgingival
For personal use only.

the present study, the definition of periodontitis was bacteria were evaluated together, P. micros and
based on PD and CAL. However, Lieff et al. [28] C. rectus may have a role in increasing the risk for
determined that the traditionally used definitions PTLBW. We did not evaluate the presence of
were insufficient for clinical periodontal disease P. micros, but, to our knowledge, our detection of
assessment. The level of clinical periodontal disease C. rectus in 12.5% of the amniotic fluid samples from
that represents exposure to oral microbes sufficient to the GP group is the first report of its presence in the
impact pregnancy outcome is unknown. Radnai et al. human amniotic cavity.
[29] observed that BOP had the strongest association In the present study, bacteria detected in the amni-
with pre-term birth. They argued that bacteria and/ otic fluid were also found in the subgingival plaque.
or their products diffused more readily than normal This co-occurrence may explain the source of these
when vascular permeability increased in gingival bacteria. Microbes could enter the amniotic cavity
tissues. Our study results confirmed that patients through four possible mechanisms. The organisms
with periodontal bacteria in the amniotic fluid had could ascend to the uterus from the vagina and cervix.
generalized periodontitis and higher BOP scores Based on routine clinical examinations, an obstetri-
(‡90%; data not shown). The presence of subgingival cian determined that bacterial vaginosis was absent
bacteria in the amniotic fluid may occur in cases of from our study population. However, as no vaginal
high periodontal disease activity and high BOP scores. samples were obtained from the subjects, we cannot
We detected T. forsythia, C. rectus, P. gingivalis and rule out the possibility that there might have been sub-
F. nucleatum in the amniotic fluid of women with GP. clinical bacterial vaginosis. This is especially impor-
This finding differed from that of Dörtbudak et al. tant in the three cases in which infected amniotic fluid
[14], who assessed whether periodontitis predicted demonstrated C. rectus. This is a limitation of our
premature gestation through the examination of study.
amniotic fluid cytokines and periodontal pathogens. The second potential mechanism of microbe entry
They found no bacteria in the amniotic fluids is the hematogenous transmission of organisms origi-
studied. However, they observed the combined pre- nating elsewhere. Third, retrograde seeding from the
sence of elevated PGE2, IL-6 and IL-8 levels in the peritoneal cavity could occur through the Fallopian
amniotic fluid at 15–20 gestational weeks. The pre- tubes. Fourth, inoculation during invasive procedures
sent investigation was limited because it did not such as amniocentesis could introduce microbes into
examine cytokines or other biochemical markers. the amniotic cavity. We detected periodontal patho-
However, we did find periodontal pathogens in gens by PCR in both the subgingival plaque and
amniotic fluid samples. amniotic fluid samples and we believe that this
 The role of periodontitis in pregnancy 559

explains the source of the bacteria. Our study was restriction and resorption in mice. Infect Immun 2003;71:
5156–62.
limited because we did not use advanced molecular
[11] Coid CR, Fox H. Short review: camplobacters as placental
diagnostic techniques to detect sub-species of the pathogens. Placenta 1983;4:295–305.
bacteria. [12] Leon R, Silva N, Ovalle A, Chaparro A, Ahumada A,
Gajardo M, et al. Detection of P. gingivalis in the amniotic
Conclusions fluid in pregnant women with a diagnosis of threatened
premature labour. J Periodontol 2007;78:1249–55.
[13] Katz J, Chegini N, Shiverick KT, Lamont RJ. Localization of
In conclusion, periodontal pathogens from the oral
P. gingivalis in preterm delivery placenta. J Dent Res 2009;88:
cavity may reach the amniotic cavity and cause 575–8.
adverse pregnancy outcomes. This study sheds light [14] Dortbudak O, Eberhardt R, Ulm M, Persson GR. Periodon-
on the mechanism underlying the association between titis, a marker of risk in pregnancy for preterm birth. J Clin
periodontal infection and adverse pregnancy out- Periodontol 2005;32:45–52.
[15] Loe H, Sillness J. Periodontal disease and pregnancy. I.
comes. More studies with larger sample sizes are
Prevalence and severity. Acta Odontol Scand 1963;21:533–
needed to confirm this association. 51.
[16] Loe H. The gingival index, the plaque index and the retention
Acta Odontol Scand Downloaded from informahealthcare.com by University of Pennsylvania on 06/21/13

Acknowledgments index systems. J Periodontol 1967;38(Suppl):610–16.

[17] Ainamo J, Bay I. Problems and proposals for recording
The corresponding author thanks Dr Mevlut Celikoglu gingivitis and plaque. Int Dent J 1975;25:229–35.
[18] Armitage GC. Periodontal diagnosis and classification of
for his excellent support during revisions.
periodontal diseases. Periodontol 2000;200434:9–21.
[19] Lin D, Moss K, Beck JD, Hefti A, Offenbacher S. Persistently
high levels of periodontal pathogens associated with preterm
Declaration of interest: The authors report no pregnancy outcome. J Periodontol 2007;78:833–41.
conflicts of interest. This project was supported by [20] Slots J, Ashimoto A, Flynn MJ, Li G, Chen C. Detection of
Hacettepe University Scientific Research Center putative periodontal pathogens in subgingival specimens by
[Project no. 09D12201006(5089)]. 16s ribosomal DNA amplification with polymerase chain
reaction. Clin Infect Dis 1995;20:303–7.
[21] Hujoel PP, Lydon-Rochelle M, Robertson PB,
References del Aguila MA. Cessation of periodontal care during preg-
For personal use only.

nancy: effect on infant birth weight. Eur J Oral Sci 2006;114:

[1] Agueda A, Echeverria A, Manau C. Association between 2–7.
periodontitis in pregnancy and preterm or low birth weight: [22] Toygar HU, Seydaoglu G, Kurklu S, Guzeldemir E, Arpak N.
review of literature. Med Oral Pathol Oral Cir Bucal 2008;13: Periodontal health and adverse pregnancy outcome in
E609–15. 3576 Turkish women. J Periodontol 2007;78:2081–94.
[2] Moutsopoulos NM, Madianos PN. Low-grade inflammation [23] Davenport ES, Williams CECS, Sterne JAC, Murad S,
in chronic infectious diseases: paradigm of periodontal infec- Sivapathasundram V, Curtis MA. Maternal periodontal dis-
tions. Ann NY Acad Sci 2006;1088:251–64. ease and preterm low birthweight: case-control study. J Dent
[3] Hill GB. Preterm birth: association with genital and possibly Res 2002;81:313–18.
oral microflora. Ann Periodontol 1998;3:222–32. [24] Lopez NJ, Smith PC, Gutierrez J. Higher risk of preterm birth
[4] Mealey BL. Influence of periodontal infections on systemic and low birth weight in women with periodontal disease.
health. Periodontology 2000;199921:197–209. J Dent Res 2002;81:58–63.
[5] Rakoto-Alson S, Tenenbaum H, Davideu JL. Periodontal [25] Moore S, Randhawa M, Ide M. A case-control study to inves-
diseases, preterm births, and low birth weight: findings tigate an association between adverse pregnancy outcome and
from a homogeneous cohort of women in Madagascar. periodontal disease. J Clin Periodontol 2005;32:1–5.
J Periodontol 2010;81:205–13. [26] Nabet C, Lelong N, Colombier ML, Sixou M, Musst AM,
[6] Offenbacher S, Katz V, Fertik G, Collins J, Boyd D, Goffinet F, et al. Maternal periodontitis and the causes of
Maynor G, et al. Periodontal infection as a possible risk factor preterm birth: the case control Epipap study. J Clin Period-
for preterm low birth weight. J Periodontol 1996;67:1103–13. ontol 2010;37:37–45.
[7] Jeffcoat MK, Geurs NC, Reddy MS, Cliver SP, [27] Holbrook WP, Oskarsdottir A, Fridjonsson T, Einarsson H,
Goldenberg RL, Hauth JC. Periodontal infection and preterm Hauksson A, Geirsson RT. No link between low-grade peri-
birth: results of a prospective study. J Am Dent Assoc 2001; odontal disease and preterm birth: a pilot study in healthy
132:875–80. Caucasian population. Acta Odontol Scand 2004;62:177–9.
[8] Han YW, Redline RW, Li M, Yin L, Hill GB, McCormic TS. [28] Lieff S, Boggess KA, Murtha AP, Jared H, Madianos PN,
Fusobacterium nucleatum induces premature and term still- Moss K, et al. The oral conditions and pregnancy study:
births in pregnant mice: implication of oral bacteria in preterm periodontal status of a cohort of pregnant women.
birth. Infect Immun 2004;72:2272–9. J Periodontol 2004;75:116–26.
[9] Buduneli N, Baylas H, Buduneli E, Türkoglu O, Köse T, [29] Radnai M, Gorzo I, Nagy E, Urban E, Novak T, Pal A.
Dahlen G. Periodontal infections and preterm-low birth A possible association between preterm birth and early peri-
weight: a case-control study. J Clin Periodontol 2005;32: odontitis. A pilot study. J Clin Periodontol 2004;31:736–41.
174–81. [30] Arce RM, Diaz PI, Barros SP, Galloway P, Bobetsis Y,
[10] Lin D, Smith MA, Champagne C, Elter J, Beck J, Threadgill D, et al. Characterization of the invasive and
Offenbacher S. Porphyromonas gingivalis infection during inflammatory traits of oral Campylobacter rectus in a murine
pregnancy increases maternal tumor necrosis factor alpha, model of fetoplacental growth restriction and in trophoblast
suppresses maternal interleukin-10, and enhances fetal growth culture. J Reprod Immunol 2010;84:145–53.