Molecular and Cellular Probes 23 (2009) 259263

Contents lists available at ScienceDirect

Molecular and Cellular Probes

journal homepage: www.elsevier.com/locate/ymcpr

Culture-based PCR analysis of plaque samples of Japanese school children

to assess the presence of six common cariogenic bacteria and its association
with caries risk
Omar M.M. Rodis a, *, Seishi Matsumura a, Naoyuki Kariya b, Yoshihide Okazaki b,
Sagiri Ogata b, Daniel R. Reimann c
a
b
c

Department of Behavioral Pediatric Dentistry, Okayama University, Okayama City, Japan

Hospital of Medicine and Dentistry, Okayama University, Okayama City, Japan
Department of Prosthetic Dentistry, University Medical Center Hamburg-Eppendorf, Hamburg, Germany

a r t i c l e i n f o

a b s t r a c t

Article history:
Received 22 May 2009
Accepted 22 May 2009
Available online 30 June 2009

The aim of the study was to assess the presence of six common cariogenic bacteria from Cariostatinoculated plaque samples of Japanese elementary school children through PCR analysis and check its
associations with caries risk testing the validity of Cariostat as a caries risk assessment tool. This
epidemiological school-based study investigated plaque samples of 399 Japanese elementary school
children. Assessed using the Cariostat, 48.2% of the children had high caries risk. DNA detection of
Streptococcus mutans, Streptococcus sobrinus, Streptococcus salivarius, Lactobacillus casei, Lactobacillus
plantarum, Lactobacillus fermentum and both S. mutans and S. sobrinus was seen in 65.2%, 24.1%, 69.7%,
17.5%, 7.8%, 19.3%, and 17.3% of the participants, respectively. Except for S. salivarius, the presence of all
other investigated bacteria resulted in a statistically signicant increase among the proportion of subjects
with high caries risk. Caries risk assessed using Cariostat was signicantly inuenced by the presence of
cariogenic bacteria. Being a selective medium for cariogenic bacteria, the Cariostat can be a useful and
direct source of cariogenic bacterial DNA for PCR analysis while effectively assessing caries risk.
2009 Elsevier Ltd. All rights reserved.

Keywords:
Caries risk
Cariogenic bacteria
Cariostat
DNA
PCR

1. Introduction
The mixed dentition period occurs when the child is in the
elementary school years. It is characterized by bone growth and
development, tooth exfoliation and eruption, change in dietary
habits, early caries, frequent increase and decrease of caries activity
that occurs around the elementary school period, or when the child
is about 612 years of age. During this period, caries activity is
known to have different intensity peaks, which makes caries
assessment difcult and caries prevention important [1]. Since
early caries and tooth loss in itself can be a sufcient predictor of
caries attack in the future, caries risk assessment is therefore
important during this period [2].
Many methods to assess caries risk have been developed. Most
of the tests believed to predict future caries relate to the classic

* Correspondence to: Omar Marianito Maningo Rodis, Department of Behavioral

Pediatric Dentistry, Okayama University Graduate School of Medicine and Dentistry,
2-5-1 Shikata-cho, Okayama City 700-8525, Japan. Tel.: 81 (86) 235 6715; fax:
81 (86) 235 6719.
E-mail address: omarodis@md.okayama-u.ac.jp (O.M.M. Rodis).
0890-8508/$ see front matter 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.mcp.2009.05.001

factors of dental caries model: the host, cariogenic bacteria and

diet. This includes age, socio-economic factors, past caries experience and presence of a specic bacterium. In general, most of these
tests are better at selecting those who will not develop caries
(specicity) than they are at selecting those who will develop caries
(sensitivity). Recent developments in bacterial strain identication
at the molecular level have paved the way to more reliable results.
Studies have established the presence of Mutans Streptococci
(MS) and Lactobacilli (LB) species as the main etiologic factors for
dental caries [35]. The bacterial species most frequently isolated
from the mouth are Streptococcus mutans, Streptococcus sobrinus,
Streptococcus salivarius [6,7] Lactobacillus casei, Lactobacillus plantarum and Lactobacillus fermentum [8]. Microbiological methods
have been developed to predict future caries attack [911].
However, available test kits are expensive and have limitations of
identifying a specic bacterium only. Assessing the collective
activity of cariogenic bacteria from a special selective nutrient
medium is therefore needed. Among available caries activity test
kits is the Cariostat [12]. The test assesses acid-production of
microorganisms in a plaque sample inoculated in the ampoule. The
Cariostat method is a caries risk assessment test used in Japan in
clinical and epidemiological studies. It is based on color changes of

260

O.M.M. Rodis et al. / Molecular and Cellular Probes 23 (2009) 259263

the test solution brought about by a decrease in pH caused by acid

producing microorganisms from the patients plaque. After scoring,
the ampoule is shown to the patient with the intent of giving them
a simple visual aid regarding their caries risk. Okazaki et al.
reported that students in Japan were found to develop tooth decay
three years after being assessed as high risk by the Cariostat
method when they were 6 years old [13]. Other longitudinal studies
reported similar ndings, further validating the predictive ability of
the Cariostat [14,15]. Though the test has been shown to have
predictive ability, less is known about its association with the
presence of cariogenic bacteria through DNA detection. If the test
correctly assesses caries risk, cariogenic bacteria should be found
more often in subjects assigned as having a high caries risk than in
those with low risk. It was, therefore, necessary to investigate the
association between cariogenic bacteria and caries risk assessment
using Cariostat.
The study aims to assess the presence of six common cariogenic
bacteria from Cariostat-inoculated plaque samples of Japanese
elementary school children through PCR analysis, and to determine
the validity of the Cariostat test as a caries risk assessment tool.
2. Materials and methods
2.1. Study population
This study was part of a school health activity program, which
started in 1987, conducted in April 2004 with an initial list of 406
elementary school children of Kobe City, Japan. Their age ranged
from six to twelve years of age. There were seven participants who
had consented but were absent during the day of the health
examination, which resulted in a nal count of 399 participants
(219 boys and 180 girls). Plaque sampling was performed for each
grade level: from Grade 1 (n 72; 42 boys, 36 girls), Grade 2
(n 64; 42 boys, 22 girls), Grade 3 (n 73; 37 boys, 36 girls), Grade
4 (n 75; 41 boys, 34 girls), Grade 5 (n 65; 34 boys, 31 girls),
through Grade 6 (n 44; 23 boys, 21 girls). The students and
parents were made aware of the health activity program through
the schools newsletters while parents were requested to give
consent with regards to their childrens participation prior to the
scheduled examination. The privacy concerns of the participants
were addressed by the authors based on the schools ethical
guidelines regarding their students privacy rights. These guidelines have been followed since the beginning of the annual health
examination program. The program staff from the Department of
Behavioral Pediatric Dentistry of Okayama University is usually
composed of six to seven dentists. Dental status was not included in
the analysis due to the diverse state of tooth eruption, exfoliation
and caries during the mixed dentition period.
2.2. Plaque sampling and caries risk assessment
A sterile cotton swab was used to collect samples from the
buccal and labial surfaces of the upper teeth in a continuous side to
side sweeping stroke that is repeated 3 times for both quadrants.
The samples were then brought back to the university in a 4  C
container on the same day and incubated immediately. After 48 h
of incubation at 37  C, the samples were scored by two welltrained examiners who were calibrated prior to the scoring session.
Being a colorimetric test, scores are assigned according to color
changes after the 48-h incubation period. A score of 0 (blue; pH
6.1  0.3) and 1.0 (green; pH 5.4  0.3) is considered as low risk
while a score of 2.0 (yellow-green; pH 4.7  0.3) and 3.0 (yellow;
pH 4.0  0.3) is considered as high risk. The inter-examiner
agreement rate using the 4-scale Cariostat scoring system had
a kappa value of 0.94.

2.3. DNA extraction and PCR

After scoring, the samples were immediately prepared for
bacterial DNA extraction and stored at -20  C until ready for PCR
analysis. In the DNA extraction procedure, the Cariostat sample was
intermittently agitated by vortexing (Ecan Tube Mixer, TM-2000,
Asahi Techno Glass, Japan) for 10 s to dislodge bacteria sticking onto
the cotton swab. Bacterial pellet was produced by transferring 1 ml
of the Cariostat sample to a 1.5 ml mini centrifuge tube, centrifuged
at 4700  g (7500 rpm, Himac CF15R with T15AP21 rotor, Hitachi)
for 10 min. DNA extraction was done using Qiagens protocol for
DNA extraction of Gram-positive bacteria. The PCR (iCycler Dual
Block Thermal Cycler, Biorad) method was performed with a 20 ml
reaction volume of which a uniform DNA template of 2.0 ml was
used for all samples. To rule out the chance for false negative results,
analysis with 16S rRNA universal primers conrmed the presence of
bacteria in the plaque samples [16]. Bacterial 16S rRNA is known to
be highly conserved within the bacterial species and is considered
to be a standard in bacterial identication and detection.
DNA detection was done for the following bacterial strains:
S. mutans, S. sobrinus, S. salivarius, L. casei, L. plantarum, L. fermentum
and both S. mutans and S. sobrinus (henceforth referred to as
S. mutsob). For the mutans streptococcal strains, a 2-primer pair
was used to specically detect S. mutans and S. sobrinus at the same
time in one PCR reaction while another species-specic primer
was used to identify S. salivarius [1719]. Also, primers designed to
specically identify lactic acid bacteria including L. fermentum,
L. casei and L. plantarum, were used [20,21]. Reference strains
(S. mutans ATCC 25175, S. sobrinus ATCC 33478, S. salivarius ATCC
7073, L. casei ATCC 4646, L. plantarum ATCC 8014, and L. fermentum
ATCC 14931) of the six bacteria were also included in each procedure as positive controls while pure distilled water (Gibco UltraPURE Distilled Water, Grand Island, NY, USA) was used as negative
control. Amplied products were subjected to electrophoresis on
1% ethidium bromide-stained agarose gel in a Tris-acetate-EDTA
buffer. Products were checked at least two times under uniform
conditions in accordance with the presence or absence of DNA
bands and veried again randomly to conrm the results.
2.4. Data analyses
Cariostat scores and presence of bacteria were graphically presented as proportions for all participants stratied by grade level.
Differences in Cariostat scores, caries risk, and presence of bacteria
with respect to gender and grade level were tested for proportions
(caries risk, presence of oral bacteria) using chi2-test and for ranked
data (Cariostat scores) using Wilcoxon rank-sum test (two-group
comparison) and KruskallWallis rank test (more than two-group
comparison).
To analyze the association between presence of cariogenic
bacteria with Cariostat scores and with caries risk [low (Cariostat
scores 0 and 1.0) and high (Cariostat scores 2.0 and 3.0)], several
data analyses were performed. First, the functional relationships,
i.e., the linear or non-linear relationship between a predictor
(presence of cariogenic bacteria) and the outcome (Cariostat score
and caries risk) was explored. For approximate linear relationships
between presence of cariogenic bacteria and the categorized
measures (scores 0, 1.0, 2.0, and 3.0) of the Cariostat test, we
computed polychoric correlation coefcients. To assess the correlation between presence of cariogenic bacteria and caries risk,
a tetrachoric correlation coefcient was computed.
To compare the prevalence of cariogenic bacteria with caries risk
assessed using Cariostat, prevalence rate ratios (PRRs) were
computed. These measures are similar to risk ratios, for example,
a PRR of 1.4 for S. mutans would indicate that the prevalence of high

O.M.M. Rodis et al. / Molecular and Cellular Probes 23 (2009) 259263

261

All analyses were performed using the statistical software

package STATA (Stata Statistical Software: Release 9.2 StataCorp,
College Station, TX), with the probability of a type I error set at the
0.05 level.
3. Results and discussion
3.1. Distribution of caries risk and prevalence of cariogenic bacteria

Fig. 1. Proportion of children with a particular Cariostat score.

caries risk is 1.4 times more likely (equals a 40% increase) in

patients with the presence of S. mutans than in patients without S.
mutans. PRRs were computed using general linear models (GLM)
with binominal distribution and iterated, reweighted least-squares
(IRLS) optimization of the deviance without and with adjustment
for grade level and gender.
In the nal step of the analysis, the clinical usefulness of the
Cariostat test used as a diagnostic test to determine the presence of
cariogenic bacteria was assessed. The classication by PCR of
patients status regarding the presence of cariogenic bacteria was
considered the reference standard, while diagnostic test accuracy
measures such as sensitivity, specicity, and positive and negative
predictive values including 95% condence intervals were calculated. In addition, the positive and negative likelihood ratios
including their 95% condence intervals were calculated. A positive
likelihood ratio describes the ratio of the probability of the positive
test result among patients with the disease to the probability in
patients who do not have the disease, whereas the negative likelihood ratio describes the ratio of the probability of the negative
test result among patients with the disease to the probability in
patients who do not have the disease [22]. Likelihood ratio can be
used to estimate post-test odds by multiplying the pre-test odds by
the likelihood ratio [23]. Guidelines for the goodness of diagnostic
tests exist and were applied to judge Cariostat as a diagnostic test to
detect the presence of cariogenic bacteria [24]. According to these
recommendations, likelihood ratio values of more than 10 or less
than 0.1 signify convincing diagnostic evidence, values of 510 and
0.10.2 signify high diagnostic evidence, values of 25 and 0.20.5
signify weak diagnostic evidence and values between 2 and 0.5
signify hardly relevant diagnostic evidence.

The distribution of Cariostat scores and DNA detection according

to grade levels is presented in Figs. 1 and 2. High caries risk was
seen in 48.2% of the participants (Cariostat score 2.0 and 3.0). Initial
PCR analysis with universal primers conrmed the presence of
bacteria in all 399 plaque samples. DNA band detection of S. mutans,
S. sobrinus, S. salivarius, L. casei, L. plantarum, L. fermentum and
S. mutsob was seen in 65.2%, 24.1%, 69.7%, 17.5%, 7.8%, 19.3% and
17.3% of the participants, respectively. There was no signicant
difference in Cariostat scores (Wilcoxon rank sum: P > 0.05) and
in caries risk (chi2-test: P > 0.05) between genders. S. mutans was
signicantly more commonly observed among girls than boys
(72.2% vs. 59.4%, chi2-test: P < 0.01) whereas S. salivarius was
more prevalent among boys than girls (62.8% vs. 75.3%; chi2-test:
P < 0.01). Neither S. sobrinus, L. casei, L. plantarum, L. fermentum nor
S. mutsob differed signicantly with respect to gender (chi2-test: all
P > 0.05). There was neither a signicant effect of grade level on
Cariostat scores (KruskallWallis rank test: P > 0.05) nor on caries
risk (chi2-test: P > 0.05). The presence of S. sobrinus (chi2-test:
P < 0.001), S. salivarius (chi2-test: P < 0.001), L. plantarum (chi2test: P < 0.05) and S. mutsob (chi2-test: P < 0.001) differed signicantly among grade levels, but not the presence of S. mutans,
L. casei, and L. fermentum (chi2-test: all P > 0.05).
DNA detection of common oral bacteria known to cause dental
caries has been associated with a high score of Cariostat-assessed
caries risk in this sample of school children. The presence of these
bacteria early in childhood leads to caries, which may cause early
loss of teeth. To prevent caries in the permanent dentition, care
must be taken for the deciduous teeth to be caries-free until the
time of exfoliation. In the study, we found signicant differences in
the distribution of cariogenic bacteria among gender and grade
levels for some bacteria. However, there was no signicant difference in Cariostat scores or caries risk among grade levels or gender.
Among the most commonly detected cariogenic bacteria in the oral
cavity included in this study, S. salivarius, S. mutans and S. sobrinus
were found to be the most prevalent bacteria detected through
DNA in this group of children. Hamada and Slade reported that
S. salivarius has been known to inhabit the mouth 24 h after birth

Fig. 2. Proportion of detected cariogenic bacteria per grade level.

262

O.M.M. Rodis et al. / Molecular and Cellular Probes 23 (2009) 259263

Table 1
Proportions of subjects with high caries risk; prevalence rate ratios (PRRs) including their 95% condence interval (95% CI) and level of statistical signicance for the independent variables (presence of cariogenic bacteria) on the dependent variable (high caries risk) and adjusted for grade level and gender.
High caries risk [%]

PRR (95% CI)

P-value

PRR (95% CI) grade

and gender adjusted

P-value grade and

gender adjusted

S. mutans

No
Yes

35.3
55.0

1.56 (1.212.00)

<0.001

1.55 (1.212.00)

<0.001

S. sobrinus

No
Yes

44.9
58.3

1.30 (1.051.60)

<0.05

1.30 (1.061.60)

<0.05

S. salivarius

No
Yes

54.6
45.3

0.83 (0.681.02)

0.080

0.85 (0.691.05)

0.128

L. casei

No
Yes

44.7
64.3

1.44 (1.161.78)

<0.001

1.43 (1.161.77)

<0.001

L. plantarum

No
Yes

45.4
80.7

1.78 (1.452.19)

<0.001

1.73 (1.402.14)

<0.001

L. fermentum

No
Yes

37.3
93.5

2.51 (2.152.95)

<0.001

2.50 (2.142.91)

<0.001

S. mutans & S. sobrinus

No
Yes

45.2
62.3

1.38 (1.111.72)

<0.01

1.37 (1.101.70)

<0.01

[25]. It is considered to be one of the most abundant streptococcal

species in the human oral cavity and is also found on tooth surfaces,
especially during early plaque development and following repeated
ingestion of sucrose. However, there are no data to support
a causative role of S. salivarius in human caries, and data even
suggest an inverse relationship in the abundance of S. salivarius and
mutans streptococci [26]. The analyses, as with the abovementioned studies, also suggest no relationship between presence
of S. salivarius and increase in the proportion of high caries risk
subjects. S. sobrinus, along with S. mutans, has also been implicated
in caries initiation and there are reports suggesting a higher
incremental caries increase in children having both S. mutans and
S. sobrinus [2729]. The longitudinal study by Okada et al. indicated
that children harboring both S. mutans and S. sobrinus have
a signicantly higher incidence of dental caries one year after initial
examination than those positive for S. mutans alone [29].
3.2. Relationship between caries risk and presence
of cariogenic bacteria
The correlation coefcients for the relationship of Cariostat
scores with the presence of cariogenic bacteria were: for S. mutans
rpolychoric 0.28 (P > 0.05), S. sobrinus rpolychoric 0.22 (P > 0.05),
S. salivarius rpolychoric 0.15 (P > 0.05), L. casei rpolychoric 0.28
(P > 0.05), L. plantarum rpolychoric 0.34 (P > 0.05), L. fermentum
rpolychoric 0.70 (P < 0.05) and S. mutsob rpolychoric 0.25 (P > 0.05).
Correlation coefcients for the relationship of caries risk with the
presence of cariogenic bacteria were: for S. mutans rtetrachoric 0.30
(P < 0.001), S. sobrinus rtetrachoric 0.20 (P < 0.05), S. salivarius
rtetrachoric 0.14 (P > 0.05), L. casei rtetrachoric 0.27 (P < 0.01),

L. plantarum rtetrachoric 0.45 (P < 0.001), L. fermentum rtetrachoric

0.78 (P < 0.001) and S. mutsob rtetrachoric 0.24 (P < 0.05). These
ndings were supported by regression analyses (GLM) comparing
the prevalence of high caries risk in subjects with and without the
presence of various cariogenic bacteria (Table 1). Using the prevalence of the cariogenic bacteria as independent variables, the
presence of all other investigated cariogenic bacteria, except for S.
salivarius, resulted in a statistically signicant increase in the
proportion of subjects with high caries risk assessed using the
Cariostat test (Table 1). Additionally, regression analysis was
adjusted for grade level and gender. However, results didnt change
substantially (Table 1).
A highly signicant relationship was observed between the
presence of S. mutans, L. plantarum, L. casei and L. fermentum and
caries risk assessed using the Cariostat method. All four are highly
acidogenic and aciduric, which are oral bacterial properties that
hasten tooth demineralization. Another very interesting nding of
this study was the high association of L. fermentum with Cariostatassessed caries risk. However, there are still no studies of the direct
association of L. fermentum with dental caries in children. Thus,
further studies of L. fermentum must be done to assess its association with child caries.
3.3. Use of Cariostat as a diagnostic test for the presence
of cariogenic bacteria
Sensitivity of Cariostat as a diagnostic test to assess the presence
of cariogenic bacteria was dependent on the prevalence of the
investigated bacteria (Table 2), i.e., for bacteria that were less
prevalent (L. plantarum and L. fermentum), values for sensitivity

Table 2
Presence (Pres) of cariogenic bacteria; sensitivity (Sens), specicity (Spec), and positive (PPV) and negative predictive value (NPV) for caries risk assessment to diagnose the
presence of cariogenic bacteria; positive (LR) and negative likelihood ratio (LR); for all subjects.
Diagnostic test accuracy measures for Cariostat to diagnose the presence of specic bacterium

S. mutans
S. sobrinus
S. salivarius
L. casei
L. plantarum
L. fermentum
S. mutans & S. sobrinus

Pres (95% CI)

Sens (95% CI)

Spec (95% CI)

PPV (95% CI)

NPV (95% CI)

LR (95% CI)

LR (95% CI)

65.2
24.1
69.7
17.5
7.8
19.3
17.0

55.0
58.3
45.2
64.3
80.7
93.5
62.3

64.8
55.1
45.5
55.3
54.6
62.7
54.8

74.5
29.2
65.6
23.4
13.0
37.5
22.4

43.5
80.7
26.6
87.9
97.1
97.6
87.4

1.56
1.30
0.83
1.44
1.78
2.51
1.38

0.70
0.76
1.20
0.65
0.35
0.10
0.69

(60.369.8)
(20.028.6)
(64.974.2)
(13.921.6)
(5.310.9)
(15.423.5)
(14.021.4)

(48.761.2)
(47.868.3)
(39.451.4)
(51.975.4)
(62.592.6)
(85.597.9)
(49.873.7)

(56.272.7)
(49.360.8)
(36.454.8)
(49.860.8)
(49.459.8)
(57.268.0)
(49.360.3)

(67.780.5)
(22.836.1)
(58.472.3)
(17.630.1)
(8.618.6)
(30.644.8)
(16.729.0)

(36.650.5)
(74.685.8)
(20.733.1)
(82.792.0)
(93.898.9)
(94.599.2)
(82.191.6)

(1.212.00)
(1.051.60)
(0.681.02)
(1.161.78)
(1.452.18)
(2.152.93)
(1.111.72)

(0.580.83)
(0.580.98)
(0.961.50)
(0.460.90)
(0.170.73)
(0.040.24)
(0.500.95)

O.M.M. Rodis et al. / Molecular and Cellular Probes 23 (2009) 259263

were high (80.7% and 93.5%) whereas for bacteria that were more
prevalent (S. mutans and S. salivarius), values for sensitivity were
low (55.0% and 45.2%). Values for specicity were low for all
bacteria and ranged from 45.5% to 64.8%, i.e., only up to two thirds
of the subjects not having the specic bacterium were correctly
identied using the Cariostat test. The positive predictive values
ranged from 13.0% to 74.5% and the negative predictive values
ranged from 26.6% to 97.6%. High caries risk assessed using Cariostat increased the probability of actually having cariogenic bacteria
for almost all investigated bacteria (Table 2), e.g., for S. mutans,
a high caries risk increased the probability of having this bacterium
from 65.2% (prevalence of S. mutans, i.e., pre-test probability) to
74.5% (post-test probability or positive predictive value) or for
presence of S. mutsob, from 17.0% to 22.4%. This is equivalent to
a positive likelihood ratio (LR) of 1.56 and 1.38, respectively
(Table 2). The values for LR varied from 0.83 to 2.51 across the
different bacteria. Almost all of these values are considered hardly
relevant for a diagnostic test. Only for L. fermentum can the test be
considered as having weak diagnostic evidence. Except for
S. salivarius, the negative likelihood ratio (LR) for all bacteria was
less than 1 and varied from 0.10 to 0.76, i.e., the presence of bacteria
among subjects with low caries risk was lower in comparison with
the presence of bacteria among high caries risk subjects. Only the
LR of L. plantarum reached the value of 0.50, which is considered
to be the border for weak diagnostic evidence and the LR of
L. fermentum reached the value of 0.10, which is considered to be
the border for convincing diagnostic evidence.
Caries risk assessed using Cariostat was signicantly inuenced
by the presence of cariogenic bacteria commonly found inside the
oral cavity. It provides a collective representation of the entire
group of bacteria present in a sample and not the presence of just
one bacterium. The likelihood ratio analyses evaluated Cariostat as
a diagnostic test for the presence of a specic bacterium in the
Cariostat sample. Results showed almost all values were considered
hardly relevant for a diagnostic test. These results were expected
since the Cariostat was developed to assess the overall quality of
bacteria present in an individual. As with the factors in the caries
model, caries risk depends not only on one factor but a combination
of factors. Moreover, the Cariostat test was developed as a selective
medium not only for Mutans Streptococci but for the Lactobacilli
species as well. Although dental status is widely used as an indicator of caries risk in the permanent dentition, it is difcult for it to
be used as such during the mixed dentition period. In a review of
twenty-eight studies evaluating the effectiveness of routine dental
checks, there was no consistency across multiple studies in the
direction of effect of different dental check frequencies on measures
of caries in deciduous mixed or permanent dentition [30].
In conclusion, the study conveys the importance of caries
activity tests that can assess caries risk by taking into consideration
the overall quality of cariogenic bacteria present and not just the
presence of a specic bacterium. Although some studies have
shown the validity of Cariostat through clinical and epidemiological
studies, this study is the rst one that has checked the presence of
six commonly found oral bacteria through DNA detection from
Cariostat-inoculated plaque samples.
Acknowledgements
The authors wish to acknowledge the assistance of the principal,
teachers and students of the elementary school in Kobe, Prof.
Tsutomu Shimono, initiator of the school program, and Dr. Duan
Chunyan, Department of Stomatology, Dalian University, China.

263

References
[1] Schlagenhauf U, Rosendahl R. Clinical and microbiological caries-risk parameters at different stages of dental development. J Pedod 1990;14(3):1413.
[2] NIH Consensus Statement. Diagnosis and Management of Dental Caries
Throughout Life 2001;18(1):124.
[3] Narhi TO, Kurki N, Ainamo A. Saliva, salivary micro-organisms, and oral health
in the home-dwelling old elderly a ve-year longitudinal study. J Dent Res
1999;78(10):16406.
[4] Loesche WJ, Schork A, Terpenning MS, Chen YM, Stoll J. Factors which inuence levels of selected organisms in saliva of older individuals. J Clin Microbiol
1995;33(10):25507.
[5] Brambilla E, Twetman S, Felloni A. Salivary mutans streptococci and lactobacilli in 9- and 13-year-old Italian schoolchildren and the relation to oral health.
Clin Oral Investig 1999;3(1):710.
[6] Whiley RA, Beighton D. Current classication of oral Streptococci. Oral
Microbiol Immunol 1998;13:195216.
[7] Thurnheer T, Gmur R, Giertsen E, Guggenheim B. Automated uorescent in
situ hybridization for the specic detection and quantication of oral streptococci in dental plaque. J Microbiol Meth 2001;44:3947.
[8] Marchant S, Brailsford SR, Twomey AC, Roberts GJ, Beighton D. The predominant microora of nursing caries lesions. Caries Res 2001;35:397406.
[9] Kohler B, Pettterson BM, Bratthall D. Streptococcus mutans in plaque and
saliva and the development of caries. Scand J Dent Res 1981;89:1925.
[10] Jordan HV, Laraway R, Snirch R, Marmel M. A simplied diagnostic system for
cultural detection and enumeration of Streptococcus mutans. J Dent Res
1987;66:5761.
[11] Klock B, Krasse B. A comparison between different methods for prediction of
caries activity. Scand J Dent Res 1979;87:12939.
[12] Shimono T, Sobue S. A new colorimetric caries activity test. Dent Outlook
1974;43(6):82935.
[13] Okazaki Y, Ji Y, Oyuntsetseg B, Rodis O, Hori M, Matsumura S, Shimono T. Level
of caries activity and an estimate in the increase of permanent teeth caries:
a three-year follow up study in preschool senior children. Ped Dent J
2005;15(1):15.
[14] Nishimura M, Bhuiyan MM, Matsumura S, Shimono T. Assessment of the caries
activity test (Cariostat) based on the infection levels of mutans streptococci
and lactobacilli in 2 to 13-year-old childrens dental plaque. ASDC J Dent Child
1998;65(4):24851.
[15] Tsubouchi J, Yamamoto S, Shimono T, Domoto PK. A longitudinal assessment of
predictive value of a caries activity test in young children. J Dent Child
1995;34:347.
[16] Sato T, Hu JP, Ohki K, Yamaura M, Washio J, Matsuyama J, Takahashi N. Rapid
identication of cariogenic bacteria by 16S rRNA genes PCR-RFLP analysis.
Cariol Today 2001;2:813.
[17] Igarashi T, Yamamoto A, Goto N. Direct detection of the Streptococcus mutans
in human dental plaque by polymerase chain reaction. Oral Microbiol
Immunol 1996;11:2948.
[18] Igarashi T, Yamamoto A, Goto N. PCR for detection and identication of
Streptococcus sobrinus. J Med Microbiol 2000;49:106974.
[19] Igarashi T, Yano Y, Yamamoto A, Sasa R, Goto N. Identication of Streptococcus
salivarius by PCR and DNA probe. Lett Appl Microbiol 2001;32:3947.
[20] Chagnaud P, Machinis K, Coutte LA, Marecat A, Mercenier A. Rapid PCR-based
procedure to identify lactic acid bacteria: application to six common
Lactobacillus species. J Microbiol Meth 2001;44:13948.
[21] Dickson EM, Riggio MP, Macpherson L. A novel species-specic PCR assay for
identifying Lactobacillus fermentum. J Med Microbiol 2005;54:299303.
[22] McGee S. Simplifying likelihood ratios. J Gen Intern Med 2002;17:64750.
[23] Deeks JJ, Altman DG. Diagnostic tests 4: likelihood ratios. BMJ 2004;329:
1689.
[24] Jaeschke R, Guyatt GH, Sackett DL. Users guide to medical literature. III. How
to use an article about a diagnostic test. B. What are the results and will they
help me in caring for my patients? The Evidence-Based Medicine Working
Group. JAMA 1994;271(9):7037.
[25] Hamada S, Slade HD. Biology, immunology and cariogenicity of Streptococcus
mutans. Microbiol Rev 1980;44:33184.
[26] Emilson CG, Ravald N, Birkhed D. Effects of a 12-month prophylactic
programme on selected oral bacterial populations on root surfaces with active
and inactive carious lesions. Caries Res 1993;27(3):195200.
[27] Hirose H, Hirose K, Isogai E, Miura H, Ueda I. Close association between
Streptococcus sobrinus in the saliva of young children and smooth-surface
caries increment. Caries Res 1996;27:2927.
[28] Nie M, Fan M, Bian K. Transmission of mutans streptococci in adults within
a Chinese population. Caries Res 2002;36:1616.
[29] Okada M, Soda Y, Hayashi F, Doi T, Suzuki J, Miura K, Kozai K. Longitudinal
study of dental caries incidence associated with Streptococcus mutans and
Streptococcus sobrinus in pre-school children. J Med Microbiol 2005;54:
6615.
[30] Davenport CF, Elley KM, Fry-Smith A, Taylor-Weetman CL, Taylor RS. The
effectiveness of routine dental checks: a systematic review of the evidence
base. Brit Dent J 2003;195:8798.