Eur Arch Paediatr Dent
DOI 10.1007/s40368-016-0252-x
ORIGINAL SCIENTIFIC ARTICLE
Histochemical changes of occlusal surface enamel of permanent
teeth, where dental caries is questionable vs sound enamel surfaces
M. Michalaki1 • C. J. Oulis1 • N. Pandis2,3 • G. Eliades4
Received: 18 March 2016 / Accepted: 21 September 2016
 European Academy of Paediatric Dentistry 2016
Abstract
Aim This in vitro study was to classify questionable for
caries occlusal surfaces (QCOS) of permanent teeth
according to ICDAS codes 1, 2, and 3 and to compare them
in terms of enamel mineral composition with the areas of
sound tissue of the same tooth.
Methods Partially impacted human molars (60) extracted
for therapeutic reasons with QCOS were used in the study,
photographed via a polarised light microscope and classi-
fied according to the ICDAS II (into codes 1, 2, or 3). The
crowns were embedded in clear self-cured acrylic resin and
longitudinally sectioned at the levels of the characterised
lesions and studied by SEM/EDX, to assess enamel mineral
composition of the QCOS. Univariate and multivariate
random effect regressions were used for Ca (wt%), P
(wt%), and Ca/P (wt%).
Results The EDX analysis indicated changes in the Ca and P
contents that were more prominent in ICDAS-II code 3
lesions compared to codes 1 and 2 lesions. In these lesions,
Ca (wt%) and P (wt%) concentrations were significantly
decreased (p = 0.01) in comparison with sound areas. Ca
and P (wt%) contents were significantly lower (p = 0.02 and
& M. Michalaki
mmichal@dent.uoa.gr; mgmichalaki@yahoo.gr
1
Department of Paediatric Dentistry, School of Dentistry,
University of Athens, 2 Thivon Str., Goudi, 115 27 Athens,
Greece
2
Department of Orthodontics and Dentofacial Orthopedics,
School of Dental Medicine/Medical Faculty, University of
Bern, Bern, Switzerland
3
Corfu, Greece
4
Department of Biomaterials, School of Dentistry, University
of Athens, 2 ThivonStr, Goudi, 115 27 Athens, Greece
p = 0.01 respectively) for code 3 areas in comparison with
codes 1 and 2 areas. Significantly higher (p = 0.01) Ca
(wt%) and P (wt%) contents were found on sound areas
compared to the lesion areas.
Conclusions The enamel of occlusal surfaces of permanent
teeth with ICDAS 1, 2, and 3 lesions was found to have
different Ca/P compositions, necessitating further investi-
gation on whether these altered surfaces might behave
differently on etching preparation before fissure sealant
placement, compared to sound surfaces.
Keywords Incipient lesions
Questionable for caries
occlusal surfaces
ICDAS-II codes 1 and 2
Altered
enamel composition
Introduction
Dental pit and fissure sealants (FS) are considered to be the
most effective preventive measure for caries reduction on
occlusal surfaces of permanent and primary molars. Resin
FS application on sound occlusal surfaces was proposed
after the introduction of enamel etching by Buonocore
(1955).
In recent years, several studies have suggested the
therapeutic potential of FS on incipient caries and some
even on cavitated lesions. The studied parameters were the
bacterial viability and progression of lesions underneath FS
in comparison with unsealed tooth surfaces. The results of
some early studies verified that FS significantly reduced
bacterial counts within the lesions as well as the progress of
the lesion (Handelman et al. 1976; Mertz-Fairhurst et al.
1979), provided that the FS remained intact. Moreover, a
systematic review on effectiveness of FS in stabilising or
reducing bacteria levels in carious lesions showed that FS
123

reduced total bacteria counts, with an enhanced reduction
as a function of time after FS placement (Oong et al. 2008).
Based on the above findings, FS have shown maximum
effectiveness when applied in high caries risk patients
(Weintraub et al. 2001; Leskinen et al. 2008; Beauchamp
et al. 2008). These findings are in agreement with the study of
Heller et al. (1995), in which there was found a tenfold
increase on FS effectiveness when applied on surfaces with
incipient lesions and in high-risk patients compared to sound
surfaces of low-risk patients. Furthermore, the systematic
review by Griffin and co-workers (2008) and in the guide-
lines of the EAPD (Welbury et al. 2004), ADA (Beauchamp
et al. 2008), and AAPD (2008) have confirmed that sealing
incipient non-cavitated caries in permanent teeth is an
effective means of caries reduction.
However, when the effectiveness of FS was correlated to
the caries risk status of the patients, the results were con-
tradictory (Weintraub et al. 2001). Some studies have
shown that the higher the dft, the higher the FS failure
(Bravo et al. 1996), or that FS are not effective in high-risk
children (Tickle et al. 2007).
Furthermore, FS have been significantly associated with
their caries preventive effect, as long as FS remains intact
on sound tooth surfaces (Thylstrup and Poulsen 1978).
Twice as many of the teeth in the high-risk group of
children needed resealing and developed caries compared
with the low-risk group (Oulis and Berdousis 2009).
Therefore it seems the retention rate, and concomitantly,
the effectiveness of FS in high-risk patients who need them
the most is problematic, due to the possibility of having
more incipient caries on the occlusal surfaces (Bravo et al.
1996). Alternatively, the lower rates of retention resulting
in higher rates of resealing and restorations can be attrib-
uted to a higher degree of microleakage that FS present
when they are applied on incipient or questionable occlusal
surfaces (Michalaki et al. 2010).
Based on the above-mentioned contradictory results, the
purpose of the present in vitro study was to determine the
changes in enamel elemental composition of questionable
occlusal surfaces, classified according to International
Caries Detection and Assessment System (ICDAS-II)
codes 1 and 2 and cavitated code 3 (Pitts 2004) and to
compare them with sound surfaces used as control.
The null hypothesis was that enamel of questionable for
caries occlusal surfaces (QCOS) does not differ in ele-
mental composition from naturally sound enamel.
Methods
One hundred partially impacted permanent third molars,
with stains, hypocalcified walls, or incipient occlusal sur-
face lesions that had been extracted for therapeutic reasons
123
Eur Arch Paediatr Dent
were selected for the study. The teeth were cleaned with a
bristle brush attached to a low-speed handpiece and a non-
fluoride prophylaxis paste (Nupro prophylaxis paste with-
out fluoride, Densply, DeTrey GmbH, Germany) and then
stored in deionized water for 12–24 h. The 40 of 100 teeth
with sound surfaces, cavitated enamel lesions or extended
cavities on enamel and dentine, were rejected from the
study. The remaining 60 molars having non-cavitated,
initial enamel lesions and deep, stained, chalky pits, and
fissures were initially categorised according to ICDAS-II
criteria. The occlusal surfaces were air-polished with
sodium bicarbonate air-flow powder (Prophyflex 3,
KaVo,Biberach, Germany, 60 lm particle size) operated
for 10 s at 3.8 bar pressure from 3 mm distance and 90
angle.
Tooth classification according to ICDAS-II system
To proceed with the histochemical analysis, all the occlusal
surfaces of the teeth were photographed at 2.59 magnifi-
cation under a reflected light stereomicroscope (L80,
Leica Microsystems, Wetzlar, Germany) and classified
according to ICDAS II into codes 1, 2, and 3.
Three paediatric dentists, appropriately trained in
ICDAS-II classification system, examined and classified
the occlusal surfaces blind, to each other, by direct
inspection under a dental operating light, followed by
indirect examination of the stereomicroscopic images. The
examiners investigated different sites within pits and fis-
sures of each occlusal surface and classified them accord-
ing to ICDAS-II codes pending agreement of at least of two
examiners on the same score (Pitts 2004; Jablonski-
Momeni et al. 2008, 2009). The study was limited to
questionable for caries occlusal surfaces (QCOS) teeth with
ICDAS-II codes 1 and 2 and cavitated code 3.
To confirm the visual evaluation, the teeth were further
subjected to radiographic examination. A custom made
holder was used to standardise the cone-tooth angle (90)
and distance (21 cm), and an acrylic sheet of 0.15 mm
thickness was placed between the cone end and the teeth as
a soft tissue substitute. Radiographs were taken at a buc-
colingual direction employing an X-ray unit (Trophy GE
1000, General Electric Co, Milwaukee, WI, USA) at 70 kV
accelerating voltage, 8 mA current, and 0.16 s exposure
time. Phosphorous plates were used to capture the images,
which were then processed in an automatic unit. Radio-
graphic evaluation was made according to the criteria
established by Ekstrand and co-workers (1997).
Study of tooth sections
The tooth crowns were embedded in transparent self-cured
acrylic resin, and after 24 h storage at room temperature,
Eur Arch Paediatr Dent
they were longitudinally sectioned at the levels of the
characterised lesions by a hard tissue microtome (Isomet,
Buehler, Lake Bluf, IL, USA), ground with silicon carbide
papers (220–2000 grit size) under continuous water cooling
and polished with 0.25 lm diamond paste (DP Paste,
Struers, Ballerup, Denmark) on a grinding/polishing
machine (Ecomet III, Buehler). The entire tooth sections
were examined under the stereomicroscope (259 magni-
fication). Fissure details were further examined by reflected
polarised light microscopy (DM 4000B, Leica Microsys-
tems). The histological classification system of Ekstrand
and co-workers (1997) (ERK) was used to assess caries
severity at each section.
The section with the worst and more representative
(according to code, surface was characterised) lesions was
used from each tooth, which resulted in the 60 sections
suitable for SEM and EDX analysis and used for statistical
analysis. Section (18) were randomly selected from the 60
sections. All 60 sections, including the 18 selected for the
final analysis, were categorised in total agreement between
examiners on their clinical and histological classifications
from the beginning.
They were used as a representative sample of the
ICDAS-II codes 1, 2, and 3 and were further studied by
scanning electron microscopy and energy dispersive
Radiographic microanalysis (SEM/EDX). An SEM was
used (Quanta 200, FEI, Hilsboro, OR, USA) equipped with
a super ultra-thin Be window X-ray Li(Si) detector (Sap-
phire CDU, EDAX Int, Mahwah, NJ, USA). Sections were
sputter-coated with carbon and imaged with a solid-state
backscattered detector (SSD) operated in atomic number
contrast mode (CBE) under 20 kV acceleration voltage,
90 lA beam current, and 1909 nominal magnification.
Elemental analysis was performed utilising spot and area
EDX modes, under the same accelerating voltage, 108 lA
beam current, 200 s acquisition time, and 34% detector
dead time. The depth of analysis was estimated as 1 lm.
The quantitative analysis was performed in non-standard
mode using ZAF (atomic number, absorption, and fluo-
rescence) and carbon-coating corrections using the Genesis
v.5.2 software (EDAX, Int). Three different regions were
analysed per section as follows: Sound enamel region (SE),
superficial lesion at fissure orifice (SL), and in-depth fissure
lesion (DL).
Statistical analysis
ICDAS II, histological, and radiographical scores were
transferred to an Excel table. Cohen’s kappa test was used
to estimate inter- and intra-examiner reproducibility. Using
the classification 2–3 as a histological threshold, the ERK
histology was used to calculate the sensitivity and speci-
ficity of each ICDAS-II cutoff. The obtained sensitivity and
specificity values were used to carry out ROC analyses at
the ERK 2–3 thresholds for each examiner.
Univariable and multivariable random effect linear
regression models were used independently for estimations
of the Ca (wt%), P (wt%) contents, Ca/P (at%) ratios, and
fluorescence measurements. The Ca (wt%), P (wt%) and
Ca/P (at%) contents were used as dependent variables,
whereas the region and ICDAS-II codes were used as
predictors. Regression coefficients and 95% confidence
intervals were also estimated. All the analyses were per-
formed using the Stata 12.1 software (Stata Corp LP,
College Station, TX, USA).
Results
Figures 1, 2, and 3 show the sets of stereomicroscopic
occlusal surface images, buccolingual radiographic images,
longitudinal sectioning directions, and polarised light sec-
tion images from representative specimens of teeth classi-
fied as ICDAS-II codes 1, 2, and 3.
Table 1 demonstrates the kappa values arising from the
agreement of all examiners. The mean correlation for the
three examiners between ICDAS-II and histological scores,
as obtained by histological examination, is 79.3% (kappa
0.65) and is characterised as substantial. The agreement
rate between visual and radiographical scores was not
found to be as strong (39.7%, kappa 0.19).
In the present study, ERK histology using the classification
2–3 as a histological threshold was preferred, since they rep-
resent deeper caries in the cementoenamel junction, between
the outer and the mid-third of the cementum. For the ERK2
threshold, all histological scores 2–4 were classified as caries,
and for the ERK3 threshold, all histological scores 3 and 4
were classified as caries, while each ICDAS-II cutoff was used
to calculate sensitivity and specificity for all examiners.
Table 2 shows areas under the curve that are very high
(0.8257 at ERK2 diagnostic threshold and 0.9681 at ERK3
diagnostic threshold). At ERK2 diagnostic threshold, the
optimum sensitivity and specificity for each examiner was
obtained using ICDAS-II limits 1–2. This indicates that
each enamel change characterised by code 1 or higher was
classified as caries. At this diagnostic threshold, sensitivity
was high (100%) and specificity was low (47.06%). At
ERK3 diagnostic threshold, where enamel changes char-
acterised by codes higher than 2 were considered caries,
sensitivity (100%) and specificity were high (93.62%).
Representative CBE images of sectioned specimens with
the corresponding EDX spectra per region of interest and the
quantitative analysis data are illustrated in Figs. 4 and 5.
The results of Ca (wt%), P (wt%), and Ca/P (at%) content
per ICDAS-II codes and region (sound enamel-SE, superfi-
cial lesion-SL, and in-depth lesion-DL) are shown in
123

Fig. 1 Questionable surface
with ICADS II code 1:
a stereomicroscopic image of
the occlusal surface, b X-ray of
the crown, c sectioning
directions, and d section image
by reflected polarised light
microscopy
Table 3. For Ca (wt%), statistically significant lower values
were registered for ICDAS-II code 3 compared to code 1
after adjusting for area (b = -4.58, 95% CI -8.34, -0.83,
p: 0.02). In addition, Ca (wt%) was significantly lower on DL
compared to SE regions (b = 3.26, 95% CI 0.64, 5.88, p:
0.01). For P (wt%), statistically significant lower values were
recorded for ICDAS II code 3 compared to code 1
(b = -2.14, 95% CI -3.67, -0.60, p: 0.01). Furthermore, P
(wt%) was significantly higher on SE regions compared to
DL regions (b = 1.80, 95% CI 0.63, 2.98, p \ 0.01) and on
SL regions compared to DL regions (b = 1.65, 95% CI 0.35,
2.96, p: 0.01). Finally, for Ca/P (at%), statistically significant
differences were found for ICDAS-II 2 code compared to
code 1 (b = 0.08, 95% CI 0.15, 0.00, p = 0.05). All other
differences were not statistically significant. Notice that the
same superscripts in Table 3 indicate mean values with no
statistically significant differences (p [ 0.05) within each
region among ICDAS codes (letters) and among the regions
for the same ICDAS code (numbers).
Discussion
In the present study, the term ‘‘questionable for caries
occlusal surfaces—QCOS’’ corresponded to ICDAS-II
codes 1 and 2. Code 3 was added for the validation and
comparison of cavitated vs non-cavitated lesions.
123
Eur Arch Paediatr Dent
The sections with the worst lesion of QCOS were
selected as the most representative of ICDAS-II codes 1, 2,
and 3 surfaces, and their histochemical composition was
assessed and compared to the sites of sound enamel of the
same tooth (Jablonski-Momeni et al. 2008, 2009). The null
hypothesis of the study was that the surface enamel clas-
sified as ICDAS-II codes 1, 2, and 3 is not altered and in
terms of its histochemical composition does not differ from
that of a sound surface. Based on the findings of the study,
the enamel mineral composition of surfaces with ICDAS
codes 1, 2, and 3 was found to differ from that of sound.
Therefore the null hypothesis was rejected.
The results of the present study are in agreement with
the conclusions of previous studies, which have shown that
the altered enamel in vitro is more porous and contains
different amounts of minerals. Cochrane and co-workers
(2012) measured mineral content in white-spot enamel
lesion from the surface to the base of the lesion in vitro.
They estimated that the maximum mineral content in the
surface layer was 74–100% of that of sound enamel. Active
lesions exhibited a more porous layer than that of inactive.
Topoleceanu and co-workers (2013) have shown that
white- and brown-spot carious lesions exhibited a signifi-
cant loss of minerals. Specifically, there was a high loss of
Ca and P contents on white spots of enamel compared to
sound enamel wt% (51.21 in sound enamel samples to
28.37 wt% in white-spot enamel samples).
Eur Arch Paediatr Dent
Fig. 2 Questionable surface
with ICADS II code 2:
a stereomicroscopic image of
the occlusal surface, b X-ray of
the crown, c sectioning
directions, and d section image
by reflected polarised light
microscopy
It seems from the present study that occlusal surfaces,
classified as ICDAS-II codes 1, 2, and 3, vary considerably in
appearance and constitute an entirely different entity in terms
of content and tissue quality compared to sound surfaces. This
might be explained by the fact that during the eruption of the
posterior teeth, there are favourable conditions for plaque
accumulation on occlusal surfaces, favouring the initiation of
demineralization and development of incipient carious
lesions. During the de- and re-mineralisation cycles, inorganic
ions of calcium and phosphates are moving out, while other
organic materials (proteins) and stains are entering into the
lesions, altering the composition of the tissue. At the time of
examination, the lesion might progress to cavitation or might
be delayed and form a questionable or arrested lesion on the
occlusal surface, depending on the local conditions, the
presence of fluoride, and the individual caries risk.
Nevertheless, the lower concentration of Ca and P
means that the mineral composition of the enamel of these
surfaces is altered and cannot be considered as sound.
This altered composition of the surface might hinder
the demineralization capacity and acid penetration into
deeper layers of enamel and cause higher failure rates of
FS on these surfaces. Probably, this is the reason that
QCOS have demonstrated higher microleakage than
sound surfaces after FS application in vitro (Michalaki
et al. 2010).
In clinical practice, surfaces with ICDAS-II codes 1
and 2 as well as surfaces with microcavities (code 3) are
prepared and sealed as sound. Based on the results of
this study, the conventional etching of altered enamel
with 37% H3PO4, may not be adequate for proper
preparation of QCOS prior to FS application. Conse-
quently, FS retention on these surfaces decreases over
time.
This might necessitate further investigations on alter-
native methods of fissure preparation and etching for better
FS bonding on such substrates. Until then, QCOS should be
sealed with caution and re-examined closely.
123
 Eur Arch Paediatr Dent

Fig. 3 Questionable surface

with ICADS II code 3:
a stereomicroscopic image of
the occlusal surface, b X-ray of
the crown, c sectioning
directions, and d section image
by reflected polarised light
microscopy

Table 1 Relationship between visual, histological, and radiographic scores according to the consensus decision of all examiners
Agreement (%) Kappa values Standard error (SE) Prob [Z

Visual (ICDAS II) vs histological scores 79.3 0.65 0.09 [0.001

Visual (ICDAS II) vs radiographical scores 39.7 0.19 0.06 0.001

Table 2 Area under the ROC curve, optimum sensitivity, and specificity and corresponding ICDAS-II threshold used at each diagnostic
threshold (ERK 2–3)
ERK2 Diagnostic threshold ERK3 Diagnostic threshold
AUC (SE) 95% CI Opt. sens. (%) Opt. spec. ICDAS AUC (SE) 95% CI Opt. sens. Opt. spec. ICDAS
(%) cutoff (%) (%) cutoff

0.83 (0.04) 0.74, 0.92 100 47.06 1–2 0.97 (0.02) 0.93, 1.00 100 93.62 2–3
AUC area under curve, SE standard error, 95% CI 95% confidence interval, Opt. sens. optimal sensitivity, Opt. Spec. optimal specificity

123
Eur Arch Paediatr Dent

Fig. 4 Clinical and BE images (40009, scale 20 lm) of QCOS lesion. Phase identification and analysis of enamel minerals by spot
section, where enamel mineral composition has been estimated in and area scan EDX (SE and DL) for SE (b) and SL (c) areas
different areas (a). SL surface lesion, SE sound enamel, DL in-depth

123
 Eur Arch Paediatr Dent

Fig. 5 Clinical and BE images (40009, scale 20 lm) of QCOS lesion. Phase identification and analysis of enamel minerals by spot
section, where enamel mineral composition has been estimated in area scan EDX (SE and DL) for SE (b) and SL (c) areas
different areas (a). SL surface lesion, SE sound enamel, DL in-depth

123
Eur Arch Paediatr Dent

Table 3 Results of Ca (wt%), P

Sound enamel (SE) Superficial lesion (SL) Deep lesion (DL)
(wt%), and Ca/P (at%) per
ICDAS code and region ICDAS 1: Ca (wt%) 30.88 (2.19) a,1
30.29 (1.85) b,1
28.89 (4.82)c,1
examined (mean ± SD)
ICDAS 2: Ca (wt%) 29.56 (2.14)a,2 26.81 (4.03)b,2 27.08 (8.97)c,2
a,3 b,3
ICDAS 3: Ca (wt%) 26.83 (6.92) 30.69 (3.52) 21.69 (6.63)c,3
a,1 b,1
ICDAS 1: P (wt%) 16.79 (1.15) 16.46 (0.94) 15.57 (2.43)c,1
a,2 b,2
ICDAS 2: P (wt%) 16.23 (0.95) 15.45 (1.74) 14.91 (2.89)c,2
a,3 b,3
ICDAS 3: P (wt%) 14.78 (3.45) 17.21 (1.26) 12.03 (2.88)c,3
ICDAS 1: Ca/P (at%) 1.42 (0.033)a,1 1.42 (0.019)b,1 1.43 (0.036)c,1
a,2 b,2
ICDAS 2: Ca/P (at%) 1.402 (0.033) 1.34 (0.065) 1.33 (0.256)c,2
a,3 b,3
ICDAS 3: Ca/P (at%) 1.39 (0.054) 1.37 (0.059) 1.37 (0.190)c,3

Conclusions Cochrane NJ, Anderson P, Davis GR, et al. An X-ray microtomo-

• The Ca concentration is altered on questionable
occlusal surfaces. Specifically, Ca concentration is
lower on surfaces characterized by ICDAS code 3
compared to surfaces with code 1 and the difference is
statistically significant. Respectively, Ca concentration
is lower on surfaces with code 2 compared to surfaces
with code 1 but the difference is not significant.
• On surfaces with ICDAS codes 1, 2, and 3, the enamel
is different as far as the region is concerned. Concen-
tration of Ca and P is lower for DL region compared to
SE and SL regions.
• Therefore, mineral composition of questionable for
caries occlusal surfaces characterised by ICDAS codes
1, 2, and 3 is altered. Based on these findings, we may
have to consider a different approach in the preparation
procedure of these surfaces prior to FS application.
Compliance with ethical standards
Conflict of interest All of the authors declare they have no conflicts
of interest.
Ethical approval This article does not contain any studies with
human participants or animals performed by any of the authors.
References
AAPD American Academy of Pediatric Dentistry Clinical Affairs
Committee, Clinical Affairs Committee. Guideline on periodic-
ity of examination, preventive dental services, anticipatory
guidance counseling and oral treatment for infants, children
and adolescents. Reference Manual. 2008;29(7):102–7.
Beauchamp J, Caufield PW, Crall JJ, et al. Evidence-based clinical
recommendations for the use of pit-and-fissure sealants: a report
of the American Dental Association Council on Scientific
Affairs. J Am Dent Assoc. 2008;139(3):257–68.
Bravo M, Osorio E, Garcia-Anllo I, Llodra JC, Baca P. The influence
of dft index on sealant success: a 48-month survival analysis.
J Dent Res. 1996;75(2):768–74.
Buonocore MG. A simple method of increasing the adhesion of
acrylic filling materials to enamel surfaces. J Dent Res.
1955;34:849–53.
graphic study of natural white-spot enamel lesions. J Dent Res.
2012;91(2):185–91.
Ekstrand KR, Ricketts DN, Kidd EA. Reproducibility and accuracy of
three methods for assessment of demineralization depth on the
occlusal surface: an in vitro examination. Caries Res.
1997;31:224–31.
Griffin SO, Oong E, Kohn W, Vidakovic B, Gooch BF, CDC Dental
Sealant Systematic Review Work Group, et al. The effectiveness
of sealants in managing caries lesions. J Dent Res.
2008;87(2):169–74.
Handelman SL, Washburn F, Wopperer P. Two-year report of sealant effect
on bacteria in dental caries. J Am Dent Assoc. 1976;93(5):967–70.
Heller KE, Reed SG, Bruner FW, Eklund SA, Burt BA. Longitudinal
evaluation of sealing molars with and without incipient dental
caries in a public health program. J Public Health Dent.
1995;55(3):148–53.
Jablonski-Momeni A, Stachniss V, Ricketts DNJ, Heinzel-Guten-
brunner M, Pieper K. Reproducibility and accuracy of the
ICDAS-II for detection of occlusal caries in vitro. Caries Res.
2008;42(2):79–87.
Jablonski-Momeni A, Stachniss V, Ricketts DNJ, et al. Impact of
scoring single or multiple occlusal lesions on estimates of
diagnostic accuracy of the visual ICDAS-II system. Int J Dent.
2009;2009:798283. doi:10.1155/2009/798283
Leskinen K, Ekman A, Oulis C, et al. Comparison of the effectiveness
of fissure sealants in Finland, Sweden and Greece. Acta Odont
Scand. 2008;66:65–72.
Mertz-Fairhurst EJ, Schuster GS, Williams JE, Fairhurst CW. Clinical
progress of sealed and unsealed caries. Part I: depth changes and
bacterial counts. J Prosthet Dent. 1979;42(5):521–6.
Michalaki MG, Oulis CJ, Lagouvardos P. Microleakage of three
different sealants on sound and questionable occlusal surfaces of
permanent molars: an in vitro study. Eur Arch Paediatr Dent.
2010;11(1):26–31.
Oong EM, Griffin SO, Kohn WG, Gooch BF, Caufield PW. The effect
of dental sealants on bacteria levels in caries lesions. J Am Dent
Assoc. 2008;139:271–8.
Oulis CJ, Berdousis E. Fissure sealant retention and caries develop-
ment after resealing on first permanent molars of children with
low, moderate and high caries risk. Eur Arch Paediatr Dent.
2009;10(4):210–7.
Pitts N. ‘ICDAS’—an international system for caries detection and
assessment being developed to facilitate caries epidemiology,
research and appropriate clinical management. Commun Dent
Health. 2004;21:193–8.
Thylstrup A, Poulsen S. Retention and effectiveness of a chemically
polymerized pit and fissure sealant after 2 years. Eur J Oral
Sci.1978;86(1):21–4.
Tickle M, Yeung CA, Milsom KM, Blinkhorm AS. The prescription
and outcomes of fissure sealants applied to a group of high caries

123

risk children by general dental practitioners working in the North
West England. Commun Dent Health. 2007;24(3):135–9.
Topoliceanu C, Stoleriu S, Ghiorghe A, et al. Chemical changes of
enamel occlusal surfaces affected by incipient dental caries: an
EDX study. Rev Chim. 2013;64(11):1324–8.
123
Eur Arch Paediatr Dent
Weintraub JA, Stearns SC, Rozier RG, Huang CC. Treatment
outcomes and costs of dental sealants among children enrolled
in Medicaid. Am J Public Health. 2001;91(11):1877–81.
Welbury R, Raadal M, Lygidakis NA. EAPD guidelines for the use of
pit and fissure sealants. Eur J Paed Dent. 2004;5:179–84.