The performance of conventional and fluorescence-based methods for occlusal caries detection: an in vivo study with histologic validation.

Abstract

The authors conducted an in vivo study to determine clinical cutoffs for a laser fluorescence (LF) device, an LF pen and a fluorescence camera (FC), as well as to evaluate the clinical performance of these methods and conventional methods in detecting occlusal caries in permanent teeth by using the histologic gold standard for total validation of the sample.
One trained examiner assessed 105 occlusal surfaces by using the LF device, LF pen, FC, International Caries Detection and Assessment System (ICDAS) criteria and bitewing (BW) radiographic methods. After tooth extraction, the authors assessed the teeth histologically. They determined the optimal clinical cutoffs by means of receiver operating characteristic curve analysis.
The specificities and sensitivities for enamel and dentin caries detection versus only dentin caries detection thresholds were 0.60 and 0.93 and 0.77 and 0.52 (ICDAS), 1.00 and 0.29 and 0.97 and 0.44 (BW radiography), 1.00 and 0.85 and 0.77 and 0.81 (LF device), 0.80 and 0.89 and 0.71 and 0.85 (LF pen) and 0.80 and 0.74 and 0.49 and 0.85 (FC), respectively. The accuracy values were higher for ICDAS, the LF device and the LF pen than they were for BW radiography and the FC.
The clinical cutoffs for sound teeth, enamel carious lesions and dentin carious lesions were, respectively, 0 through 4, 5 through 27 and 28 through 99 (LF device); 0 through 4, 5 through 32 and 33 through 99 (LF pen); and 0 through 1.2, 1.3 and 1.4 through 5.0 (FC). The ICDAS, the LF device and the LF pen demonstrated good performance in helping detect occlusal caries in vivo. The ICDAS did not seem to perform as well at the D(3) threshold (histologic scores 3 and 4) as at the D(1) threshold (histologic scores 1-4). BW radiography and the FC had the lowest performances in helping detect lesions at the D(1) and D(3) thresholds, respectively.
Occlusal caries detection should be based primarily on visual inspection. Fluorescence-based methods may be used to provide a second opinion in clinical practice.

Full text

Page 1

2012;143;339-350
JADA
and Rita C.L. Cordeiro
Rodrigues, Lourdes Santos-Pinto, Adrian Lussi
Michele B. Diniz, Thalita Boldieri, Jonas A.
validation
detection : An in vivo study with histologic
cariesfluorescence-based methods for occlusal
The performance of conventional and
jada.ada.org ( this information is current as of April 2, 2012):
The following resources related to this article are available online at
http://jada.ada.org/content/143/4/339
in the online version of this article at:
including high-resolution figures, can be found
Updated information and services
http://jada.ada.org/content/143/4/339/#BIBL
, 5 of which can be accessed free:
45 articles
This article cites
http://www.ada.org/990.aspx
this article in whole or in part can be found at:
of this article or about permission to reproduce
reprints
Information about obtaining
prohibited without prior written permission of the American Dental Association.
Copyright © 2012 American Dental Association. All rights reserved. Reproduction or republication strictly
on April 2, 2012
jada.ada.org
Downloaded from

Page 2

C
subjective assessments of
color, translucency and den-
tal hardness,1,2as well as by
means of radiographic im -
ages. These methods, how-
ever, exhibit high specificity
and low sensitivity for
occlusal caries detection.3
The International Caries
Detection and Assessment
System (ICDAS) was devel-
oped as a way to reduce
visual inspection bias by pro-
viding six distinct carious
lesion severity stages,
ranging from initial changes
visible in the enamel to
frank cavitation in the
dentin.4-6The ICDAS is
based on the visual exami-
nation of clean and plaque-
free surfaces in both wet and
dry conditions. Investigators
have had good results from
using the ICDAS for in vitro
detection of occlusal caries in
permanent teeth.5,7-11
Because the detection of
dental caries is not easy, new
methods have been proposed
to improve caries detection
and to quantify early lesions.
linical caries detec-
tion commonly is per-
formed by means of
visual inspection and
Dr. Diniz is a professor, Department of Pediatric Dentistry, School of Dentistry, UNICSUL—Cruzeiro do Sul University, São Paulo; and researcher, Department
of Pediatric Dentistry, School of Dentistry of Araraquara, UNESP—Universidade Estadual Paulista, São Paulo. Address reprint requests to Dr. Diniz at Depart-
ment of Pediatric Dentistry, School of Dentistry of Cruzeiro do Sul University—UNICSUL, Rua Baltazar Lisboa, 148 apt. 501, 04110-060, São Paulo, Brazil,
e-mail mibdiniz@hotmail.com.
Dr. Boldieri is a doctoral student, Department of Pediatric Dentistry, School of Dentistry of Araraquara, UNESP—Universidade Estadual Paulista, São Paulo.
Dr. Rodrigues is a professor, Department of Pediatric Dentistry, School of Dentistry, Federal University of Rio Grande do Sul, Porto Alegre, Brazil.
Dr. Santos-Pinto is a professor, Department of Pediatric Dentistry, School of Dentistry of Araraquara, UNESP—Universidade Estadual Paulista, São Paulo.
Dr. Lussi is a professor, Department of Preventive, Restorative and Pediatric Dentistry, School of Dental Medicine, University of Bern, Switzerland.
Dr. Cordeiro is a professor, Department of Pediatric Dentistry, School of Dentistry of Araraquara, UNESP—Universidade Estadual Paulista, São Paulo.
The performance of conventional
and fluorescence-based methods
for occlusal caries detection
An in vivo study with histologic validation
Michele B. Diniz, DDS, MSc, PhD; Thalita Boldieri, DDS, MSc;
Jonas A. Rodrigues, DDS, MSc, PhD; Lourdes Santos-Pinto, DDS, MSc, PhD;
Adrian Lussi, DMD; Rita C.L. Cordeiro, DDS, MSc, PhD
A B S T R A C T
Background. The authors conducted an in vivo study to
determine clinical cutoffs for a laser fluorescence (LF)
device, an LF pen and a fluorescence camera (FC), as well
as to evaluate the clinical performance of these methods
and conventional methods in detecting occlusal caries in per-
manent teeth by using the histologic gold standard for total
validation of the sample.
Methods. One trained examiner assessed 105 occlusal surfaces by using the
LF device, LF pen, FC, International Caries Detection and Assessment System
(ICDAS) criteria and bitewing (BW) radiographic methods.After tooth extrac-
tion, the authors assessed the teeth histologically. They determined the optimal
clinical cutoffs by means of receiver operating characteristic curve analysis.
Results. The specificities and sensitivities for enamel and dentin caries
detection versus only dentin caries detection thresholds were 0.60 and 0.93
and 0.77 and 0.52 (ICDAS), 1.00 and 0.29 and 0.97 and 0.44 (BW radiography),
1.00 and 0.85 and 0.77 and 0.81 (LF device), 0.80 and 0.89 and 0.71 and 0.85
(LF pen) and 0.80 and 0.74 and 0.49 and 0.85 (FC), respectively. The accuracy
values were higher for ICDAS, the LF device and the LF pen than they were
for BW radiography and the FC.
Conclusions. The clinical cutoffs for sound teeth, enamel carious lesions and
dentin carious lesions were, respectively, 0 through 4, 5 through 27 and 28
through 99 (LF device); 0 through 4, 5 through 32 and 33 through 99 (LF pen);
and 0 through 1.2, 1.3 and 1.4 through 5.0 (FC). The ICDAS, the LF device and
the LF pen demonstrated good performance in helping detect occlusal caries in
vivo. The ICDAS did not seem to perform as well at the D3threshold (histologic
scores 3 and 4) as at the D1threshold (histologic scores 1-4). BW radiography
and the FC had the lowest performances in helping detect lesions at the D1and
D3thresholds, respectively.
Clinical Implications. Occlusal caries detection should be based primarily on
visual inspection. Fluorescence-based methods may be used to provide a second
opinion in clinical practice.
Key Words. Occlusal caries; caries detection; visual examination; radiog-
raphy; fluorescence; clinical study; permanent teeth.
JADA 2012;143(4):339-350.
J
A
D
A
C O N T I N U I N G
ARTICLE 1
E D U C A T I O N
✷
✷
®
JADA 143(4) http://jada.ada.org April 2012 339
S T O R Y
C O V E R
Copyright © 2012 American Dental Association. All rights reserved.
on April 2, 2012
jada.ada.org
Downloaded from

Page 3

340
JADA 143(4) http://jada.ada.org April 2012
Some of these methods are based on the fluores-
cence phenomenon because there is a difference
in fluorescence observed in sound teeth and in
demineralized dental tissues.3,12A laser fluores-
cence (LF) device (DIAGNOdent 2095, KaVo,
Biberach, Germany) and LF pen (DIAGNOdent
Pen 2190, KaVo) are able to capture, analyze
and quantify the fluorescence emitted from bac-
terial porphyrins and other chromophores when
the tooth surface is illuminated by diode lasers,
which have wavelengths of 655 nanometers.12-14
The greater fluorescence intensity is seen as a
higher reading.15,16Investigators in some studies
have reported good performance of the LF
device and the LF pen for occlusal caries detec-
tion in vitro8,15,16and in vivo.16-20However, other
investigators in some clinical studies21-24and a
systematic review of the literature3evaluated
the effectiveness of LF in helping detect carious
lesions. They observed a large number of false-
positive results, which limits the LF device’s use
as a principal diagnostic tool.
A fluorescence camera (FC) (VistaProof, Dürr
Dental, Bietigheim-Bissingen, Germany) was
developed to detect caries. This device emits
blue light at 405 nm, which stimulates por-
phyrins to emit red light that contains less
energy. The denser the bacteria colonization on
the tooth’s surface, the more intense the red
fluorescent signal is. Sound enamel, in contrast,
sends out a green fluorescent signal. These light
signals are recorded by the highly developed
optics and analyzed by the device’s software.
The software filters and quantifies the fluores-
cence emitted by the dental tissue and converts
the relationship between green and red fluores-
cent signals into numeric values ranging from 0
to 5, according to the pixel numbers in each
image. Higher values indicate more severe car-
ious lesions. The software also shows the red
fluorescent signal from carious lesions in dif-
ferent colors (blue, red, orange and yellow) that
are related to the numeric values, indicating the
carious lesion’s severity.25 Investigators in two
studies evaluated the in vitro performance of
this method for detecting occlusal caries in per-
manent teeth.8,26They found that use of an FC
had high reproducibility and good diagnostic
performance for detecting occlusal caries at
various stages of the disease process. The
results from another study showed that use of
an FC had high reliability in detecting caries on
occlusal and smooth surfaces in primary teeth.27
To our knowledge, no clinical study has been
performed to test the FC’s performance in
detecting caries.
The clinical performance of fluorescence-
based caries detection methods is related to the
cutoffs used in clinical practice, which enable
dental professionals to provide patients with the
appropriate treatment. The importance of cut-
offs means that a small change in the reading
could be the difference between performing an
operative intervention or not, which is a dan-
gerous aspect of these devices if they are used
throughout the decision-making process. Vari -
ous cutoffs have been suggested for the LF
device, not only by the manufacturer but also as
the result of in vitro8,13,15and in vivo17,20 studies.
For the LF pen, some in vitro cutoffs have been
proposed,8,13and, to our knowledge, only one
study has been published regarding the clinical
cutoffs for occlusal caries detection.19In the
latter study, the investigators validated the use
of the LF device in clinical conditions by exam-
ining each tooth on the basis of visual character-
istics and scoring it as a sound tooth, a tooth
with enamel carious lesion or a tooth with den -
tin carious lesion. They also examined each
tooth on the basis of radiographic findings and
scored it as no radiolucency (sound tooth),
radiolucency indicating enamel carious lesions
and radiolucency indicating dentin carious
lesions. They used the association of the visual
and radiographic findings to determine whether
an operative or a preventive intervention was
need ed. This partial validation of the sample
cannot be used to detect false-negative results.
To our knowledge, there are no clinical cutoffs
published for the FC. Consequently, more clin-
ical studies are needed to establish the most
appropriate cutoffs for all fluorescence-based
methods (the LF device, the LF pen and the FC).
Another important aspect in a clinical
research study is related to the gold standard
method used. Clinical examination of teeth indi-
cated for extraction owing to orthodontic or
other reasons followed by histologic validation
would be the optimal approach for solving the
ethical problem of validating sound surfaces and
enamel carious surfaces.16Two studies have
been conducted to evaluate the effectiveness of
the LF device in helping detect occlusal caries in
permanent teeth in clinical conditions in which
histologic validation was used as a gold
standard after tooth extraction.24,28
The aims of our in vivo study were to deter-
mine clinical cutoffs for the LF device, the LF
pen and the FC and to evaluate the clinical per-
formance of these methods and conventional
ABBREVIATION KEY. BW: Bitewing. FC: Fluorescence
camera. ICDAS: International Caries Detection and
Assessment System. LF: Laser fluorescence.
S T O R Y
C O V E R
Copyright © 2012 American Dental Association. All rights reserved.
on April 2, 2012
jada.ada.org
Downloaded from

Page 4

JADA 143(4) http://jada.ada.org April 2012 341
methods in detecting occlusal caries in perma-
nent teeth by using the histologic gold standard
for total validation of the sample.
METHODS
We conducted our study in accordance with the
Declaration of Helsinki,29and the ethics com-
mittee in research of the School of Dentistry of
Araraquara, UNESP—Universidade Estadual
Paulista, São Paulo (protocol 04/08) approved
the study. We explained the study’s aim, pro-
cedures, safety precautions and benefits to the
participants. We obtained written informed con-
sent from all volunteer participants before the
start of the study.
Participant selection. We recruited pa-
tients aged 18 through 35 years at the Depart-
ment of Oral and Maxillofacial Surgery of the
School of Dentistry of Araraquara. We included
in our study 88 patients who each had at least
one posterior tooth scheduled for extraction
because of periodontal disease or for orthodontic
reasons, independent of its occlusal surface con-
dition, which ranged from having macroscopi-
cally intact occlusal surfaces to different stages
of noncavitated and cavitated carious lesions.
We excluded patients who had severe diseases
or syndromes and those who received prescrip-
tion medications.
Among the participants, we selected 105 pos-
terior permanent teeth—40 premolars (83 per-
cent first premolars and 17 percent second pre-
molars) and 65 molars (5 percent first and
second molars and 95 percent third molars)
scheduled for extraction. We did not include
teeth with carious lesions of the approximal,
buccal or lingual surfaces; extensive coronal
destruction; restorations; pit-and-fissure
sealants; hypoplasia; orthodontic bands; or third
molars in eruption.
We cleaned the occlusal surfaces carefully by
using a low-speed handpiece with a rotating
brush and water,30since remnants of prophy-
lactic paste can influence the performance of
fluorescence-based methods significantly.31An
independently trained examiner (T.B.) who was
not involved in the clinical examinations ran-
domly selected one occlusal site on each tooth as
the test site, because having multiple investiga-
tion sites on one tooth does not result in statisti-
cally independent data.7We recorded these sites
and identified them on a simple drawing of the
occlusal surface to guide lesion location. We
obtained images of the occlusal surfaces with a
digital camera (EOS Rebel XSi, Canon, Tokyo)
with a 100-millimeter lens.
One examiner (M.B.D.) who had previous
experience using ICDAS criteria and conven-
tional and fluorescence-based caries detection
methods conducted the clinical examinations in
all sites before the patients’ teeth were ex-
tracted. A member of the ICDAS coordinating
committee calibrated the examiner’s technique
when the examiner was on a research trip to the
United States. The calibration consisted of a dis-
cussion based on articles regarding ICDAS and
practical sessions involving extracted teeth with
different ICDAS caries codes specifically selected
for this purpose. The examiner also participated
in the online ICDAS e-learning program devel-
oped by the ICDAS Foundation, which consists
of a 90-minute course regarding the use of the
coding system.32The course is divided into the
following categories: introduction, ICDAS exami-
nation protocol, ICDAS caries codes, information
regarding the application of the coding system, a
decision tree, instructions regarding how to col-
lect data for recording codes and interactive
quizzes. A chairside assistant handled all of the
information collection.
Visual examination. The examiner visually
examined patients while they were reclined in a
dental chair and with the aid of a light reflector,
air-water spray and size 5 dental mirror. First,
the examiner analyzed the teeth when wet and
then dry. The examiner then used a ball-ended
explorer without applying pressure to confirm
cavitation only. The chairside assistant recorded
caries status according to the examiner’s assess-
ments made on the basis of the ICDAS criteria:
d0: sound tooth surface (no evidence of caries
after prolonged air drying [five seconds]);
d1: first visual change in the enamel (opacity
or discoloration [white or brown] is visible at the
entrance to the pit or fissure after prolonged air
drying, which is not or hardly seen on a wet
tooth surface);
d2: distinct visual change in the enamel
(opacity or discoloration distinctly visible at the
entrance to the pit or fissure when wet, and the
lesion still is visible when dry);
d3: localized enamel breakdown due to caries
with no visible dentin or underlying shadow
(opacity or discoloration wider than the natural
fissure or fossa when wet and after prolonged
air drying);
d4: underlying dark shadow from the dentin;
d5: distinct cavity with visible dentin (visual evi-
dence of demineralization and dentin exposure);
d6: extensive distinct cavity with visible dentin
and more than one-half of the tooth surface
involved.7
Radiographic examination. We obtained
bitewing (BW) radiographs by using radio-
S T O R Y
C O V E R
Copyright © 2012 American Dental Association. All rights reserved.
on April 2, 2012
jada.ada.org
Downloaded from

Page 5

342
JADA 143(4) http://jada.ada.org April 2012
graphic equipment, a BW radiograph positioner
and ultraspeed radiographic film at 70 peak
kilovolts and 10 milliamperes with an exposure
time of 0.6 seconds. We developed all radio-
graphs under standardized conditions by means
of an automatic radiographic film developer.
The examiner used a radiograph viewer and a
radiographic film magnifier (magnification ×2 )
in a darkroom to determine the presence or
absence of radiolucency. We recorded caries as
follows:
d0: no radiolucency;
d1: radiolucency in the enamel;
d2: radiolucency in the outer one-half of the
dentin;
d3: radiolucency in the inner one-half of the
dentin.8
LF device and LF pen assessments. After
the examiner concluded the visual examination,
we examined each occlusal surface by using the
LF device and LF pen. We obtained the LF
device and LF pen measurements by using a
fiber-optic conical tip (tip A) specifically de-
signed for occlusal surfaces and a cylindrical
sapphire-fiber tip, respectively, according to the
manufacturer’s instructions.
Before taking each measurement, we cali-
brated the devices by using a ceramic standard,
and we recorded the fluorescence of a sound
spot on the cuspal area of the buccal surface of
each tooth to provide a baseline value for that
tooth. We assessed the sites under cotton roll
isolation and after briefly air-drying them with
a three-in-one syringe. We positioned the tip of
each LF device and LF pen perpendicular to the
test site and rotated it along its long axis
according to the manufacturer’s instructions.
Both devices showed integer values ranging
from 0 to 99. We recorded the highest reading
shown on the device and subtracted the baseline
value from this value.8To avoid giving the ex-
aminer an idea of the score of the carious lesion,
we did not use the audio option on either device
because the intensity of the noise emitted could
indicate the severity of the lesion. We handed
the chairside assistant the LF pen to be read, as
the examiner needed to be masked from the flu-
orescence measurement.
FC assessments. We performed the FC
measurements with the tooth under cotton roll
isolation and after brief air-drying with a three-
in-one syringe in a dark environment. We
selected the type of spacer tip was according to
the patient’s mouth opening. The thinner and
thicker spacers provide distances of about 0.5
centimeters and 1.0 cm, respectively, between
the tip and the tooth.27After capturing the
images, we analyzed them with the FC-specific
software (DBSWIN, Version 5.3, Dürr Dental),
which translates the red and green rate of fluo-
rescence to numbers that correspond to the
lesion severity.8We recorded the values, which
can vary from 0 to 5, for further analysis.
According to the manufacturer’s scale, 0.0 to 0.9
indicates sound enamel, 1.0 to 1.5 indicates
beginning enamel caries, 1.5 to 2.0 indicates
deep enamel caries, 2.0 to 2.5 indicates dentin
caries, and 2.5 to 5.0 indicates deep dentin
caries.26
Tooth extraction. With patients under local
anesthesia, oral and maxillofacial surgery resi-
dents carefully extracted the teeth under strict
asepsis by using forceps and elevators. We then
immediately froze each tooth at −20°C.
Histologic validation. First, we defrosted
the teeth for three hours, we removed the cal-
culus and debris by using a scaler, and then we
cleaned the teeth for 15 seconds with water and
a toothbrush. After that, we hemisectioned the
teeth perpendicular to the occlusal surface in a
buccal-to-lingual direction near the center of the
test site by using a water-cooled diamond blade
0.5-mm thick. Then we ground the hemisection
with the deepest demineralization by using sil-
icon carbide paper with decreasing grits of 400,
600, 1,200 and 2,000. We constantly checked the
progression of the grinding process under a
stereomicroscope with a magnification of ×10
until the periphery of the site was reached. We
then used a brush to color the hemisection with
saturated rhodamine B solution and immedi-
ately rinsed the solution off in tap water to
remove dye remnants. We examined the speci-
mens histologically according to how far the
rhodamine B penetrated into either the enamel
or both the enamel and dentin tissues. We
assessed the sites for caries extension (histologic
score; magnification ×10) as follows: 0, caries-
free; 1, caries extending as much as halfway
through the enamel; 2, caries extending into the
inner one-half of the enamel; 3, caries in the
outer one-half of the dentin; and 4, deep dentin
caries involving the inner one-half of the
dentin.15Two experienced examiners (M.B.D.,
R.C.L.C.) evaluated each tooth section inde-
pendently. If they disagreed about a section,
they re-examined that section and discussed
their findings until they reached a consensus.
Statistical analysis. We used statistical
software (MedCalc Software for Windows, Ver-
sion 9.3.0.0, Mariakerke, Belgium) to analyze
the data; the level of significance was P < .05.
We performed the analysis after we dichoto-
mized the histologic scores into four thresholds:
S T O R Y
C O V E R
Copyright © 2012 American Dental Association. All rights reserved.
on April 2, 2012
jada.ada.org
Downloaded from

Page 6

JADA 143(4) http://jada.ada.org April 2012 343
D1(histologic scores 1-4), D2(histologic scores 2-
4), D3(histologic scores 3 and 4) and D4(histo-
logic score 4). Using these scores as the refer-
ence standard, we plotted receiver operating
characteristic (ROC) curves for the LF device,
LF pen and FC readings at the D1, D2, D3and D4
thresholds. We determined the optimal clinical
cutoffs for the LF device, the LF pen and the FC
by the cutoff at which the sum of sensitivity and
specificity values was maximal (highest accu-
racy value) in the ROC curve analysis at each
threshold. The area under the ROC curve (Az)
value reflects the diagnostic performance of a
method, and an Azvalue near 1 indicates excel-
lent accuracy. Because the cutoffs at the D1and
D2thresholds and at the D3and D4thresholds
were the same, we presented the optimal cutoffs
in terms of sound teeth (histologic score 0),
enamel carious lesions (histologic scores 1 and
2) and dentin carious lesions (histologic scores
3 and 4).
We assessed the LF device, LF pen and FC
performances by using the cutoffs proposed by
the manufacturers, as well as the optimal cut-
offs we determined by means of the ROC curve
analysis. The cutoffs the manufacturer proposed
for the LF device were 0 through 10, sound
teeth; 11 through 30, enamel carious lesions;
and more than 30, dentin carious lesions. The
cutoffs the manufacturer proposed for the LF
pen were 0 through 13, sound teeth; 14 through
29, enamel carious lesions; and more than 29,
dentin carious lesions. The cutoffs the manufac-
turer proposed for the FC were 0 through 0.9,
sound teeth; 1 through 2, enamel carious
lesions; and more than 2, dentin carious lesions.
We calculated sensitivity, specificity, accuracy
and Azvalues at the D1(histologic scores 1-4;
enamel and dentin caries) and D3(histologic
scores 3-4; only dentin caries) thresholds. We
also calculated the percentage of en-amel car-
ious lesions that were detected correctly by
using all methods and different cutoffs. We used
the McNemar test (P < .05) to compare perform-
ance among the methods. The diagnostic per-
formance is the ability of a method to help
detect a carious lesion at a given scoring
threshold. We used a nonparametric statistical
test to assess the differences among the Az
values. We used a Spearman rank correlation
coefficient to evaluate the agreement of the dif-
ferent methods with histologic scores. We con-
sidered correlation coefficient values greater
than 0.90 excellent correlation, greater than
0.75 good correlation, greater than 0.50 mod-
erate correlation, greater than 0.25 fair correla-
tion and less than 0.25 poor correlation.
RESULTS
We included in our study 105 permanent teeth
(42 maxillary molars, 23 mandibular molars, 24
maxillary premolars and 16 mandibular premo-
lars). From the 105 occlusal test sites we as-
sessed, histologic examination results showed
that five sites (5 percent) had a score of 0, 14 (13
percent) had a score of 1, 59 (56 percent) had a
score of 2, 16 (15 percent) had a score of 3, and
11 (11 percent) had a score of 4. Of the 105
occlusal test sites, use of the ICDAS criteria
showed that 10 had a score of 0, 16 had a score
of 1, 47 had a score of 2, 24 had a score of 3, six
had a score of 4, one had a score of 5, and one
had a score of 6.
Table 1 shows cross-tabulation for the
ICDAS, BW radiography, LF device, LF pen and
FC methods with the corresponding histologic
results. For the five sound sites, both the BW
radiography and LF device methods correctly
indicated five (100 percent), whereas the LF pen
and FC methods indicated four (80 percent) and
the ICDAS method indicated three (60 percent).
Considering enamel caries in the histologic
analysis (histologic scores 1 and 2, total 73), the
ICDAS correctly indicated 48 enamel carious
lesions (66 percent), whereas the LF device cor-
rectly indicated 40 (55 percent), and the LF pen
correctly identified 39 (53 percent). In contrast,
the FC correctly indicated 10 (14 percent) and
BW radiography correctly indicated 11 (15 per-
cent). Considering dentin caries in the histologic
analysis (histologic scores 3 and 4, total 27), the
LF device correctly indicated 22 (82 percent),
and the LF pen and the FC each indicated 23
(85 percent) (data not shown). The ICDAS cor-
rectly indicated only 14 (52 percent), and BW
radiography correctly indicated 12 (44 percent).
Table 2 (page 345) shows the optimal clinical
cutoffs for the ICDAS, BW radiography, LF
device, LF pen and FC methods. It also provides
a summary of the values used to identify sound
teeth, enamel carious lesions and dentin carious
lesions for each method according to the histo-
logic scores. We determined that teeth had car-
ious lesions if the score was higher than zero for
the ICDAS and BW radiography, higher than 4
for the LF device and the LF pen and higher
than 1.2 for the FC.
Table 3 (page 346) shows the specificity, sen-
sitivity, accuracy, Azvalue, positive likelihood
ratio (LR+) and negative likelihood ratio (LR−)
for the ICDAS, BW radiography, the LF device,
the LF pen and the FC at the D1and D3thresh-
olds. We used the optimal cutoffs determined for
the LF device, the LF pen and the FC to calcu-
late these values. The accuracy values were
S T O R Y
C O V E R
Copyright © 2012 American Dental Association. All rights reserved.
on April 2, 2012
jada.ada.org
Downloaded from

Page 7

344
JADA 143(4) http://jada.ada.org April 2012
higher for the ICDAS, LF device and LF pen
methods than for the BW radiography and FC
methods. The ICDAS had high sensitivity and
moderate specificity at the D1threshold, and at
the D3threshold, the ICDAS demonstrated mod-
erate sensitivity and good specificity. BW radio-
graphy had the highest specificities and the
lowest sensitivities at the D1and D3thresholds.
We observed that the LF device and the LF pen
had high sensitivities and specificities for
enamel and dentin caries detection, whereas
they had high sensitivity and specificity values
for dentin caries detection. The FC had signifi-
cantly lower sensitivity and specificity for
enamel and dentin caries, respectively.
Table 4 (page 347)
shows the specificity,
sensitivity and accu-
racy for the LF device,
the LF pen and the
FC, according to the
cutoffs the manufac-
turers proposed and
the optimal cutoffs at
the D1and D3thresh-
olds. At the D1
threshold, use of the
optimal cutoffs for
both the LF device
and the LF pen signif-
icantly improved the
sensitivity and accu-
racy values compared
with use of the cutoffs
proposed by the man-
ufacturer. In contrast,
use of the optimal cut-
offs for the FC wors-
ened the sensitivity
and accuracy values
and improved speci-
ficity values. At the D3
threshold, there was
no significant change
in the performance of
the LF device and the
LF pen when we used
the optimal and the
manufacturer’s cut-
offs. However, use of
the optimal cutoffs for
the FC worsened
specificity and accu-
racy values and sig-
nificantly improved
the sensitivity values.
Table 5 (page 347)
shows the percentage of enamel carious lesions
detected correctly by means of all of the
methods. The cutoff recommended by the FC’s
manufacturer had the highest percentage, fol-
lowed by the ICDAS criteria and the optimal
cutoffs for the LF device and the LF pen. The
optimal cutoff for the FC had the lowest per-
centage, and BW radiography had the second
lowest percentage.
The Spearman rank correlation coefficients
were 0.48 for the ICDAS, 0.48 for BW radio-
graphy, 0.52 for the LF device, 0.55 for the LF
pen and 0.35 for FC, indicating poor to mod-
erate agreement of these methods with histo-
logic scores.
TABLE 1
Cross-tabulation for the methods with the
corresponding histologic score.
METHOD, HISTOLOGIC SCORESOUND
TEETH, NO.
ENAMEL
CARIES, NO.
DENTIN
CARIES, NO.
TOTAL
International Caries Detection
and Assessment System
0
1
2
3
4
Total
3
1
6
0
0
10
2
12
36
10
3
63
0
1
17
6
8
32
5
14
59
16
11
105
Bitewing Radiography
0
1
2
3
4
Total
5
13
47
9
2
76
0
1
10
3
1
15
0
0
2
4
8
14
5
14
59
16
11
105
Laser Fluorescence Device
0
1
2
3
4
Total
5
4
11
0
0
20
0
6
34
3
2
45
0
4
14
13
9
40
5
14
59
16
11
105
Laser Fluorescence Pen
0
1
2
3
4
Total
4
5
6
0
0
15
1
6
33
3
1
44
0
3
20
13
10
46
5
14
59
16
11
105
Fluorescence Camera
0
1
2
3
4
Total
4
6
18
2
0
30
0
1
9
2
0
12
1
7
32
12
11
63
5
14
59
16
11
105
S T O R Y
C O V E R
Copyright © 2012 American Dental Association. All rights reserved.
on April 2, 2012
jada.ada.org
Downloaded from

Page 8

JADA 143(4) http://jada.ada.org April 2012 345
DISCUSSION
We compared results
from the use of
fluorescence-based
methods to detect
occlusal caries in clinical
conditions with those of
visual and radiographic
examination methods.
Sound teeth and teeth
with enamel carious
lesions cannot be vali-
dated by means of fis-
sure eradication or
enameloplasty pro-
cedures in clinical conditions for ethical
reasons.16To our knowledge, only two studies
have been published regarding validated cutoffs
for the LF device in sound occlusal surfaces in
permanent teeth in clinical conditions.24,28Many
in vivo studies are limited to premolars or third
molars, for which extraction can be scheduled.
In our study, the sample consisted of 105 pos-
terior teeth (38 percent premolars and 62 per-
cent molars; 95 percent of the latter were third
molars) that were scheduled for extraction.
The successful performance of fluorescence-
based detection methods depends on the cutoffs
used. Manufacturers propose different cutoffs
than those obtained in in vitro and in vivo
studies, which may explain the variety of results
found in the literature, which can confound clini-
cians who are deciding the best treatment. An -
other complication is that the values obtained
with the LF device and the LF pen are not the
same. In one study, the fluorescence values
obtained by using the LF pen in vivo were sub-
stantially lower than those obtained by using the
LF device.18The data from our study showed that
the fluorescence values obtained with the LF pen
were higher than those from the LF device. The
results from other studies also showed higher flu-
orescence values for the LF pen.8,33
The LF device’s and LF pen’s optimal cutoffs
in our study are different from those published
in other clinical studies.17-20,34These differences
could be attributed to the clinical conditions,
sample size and validation method used. In our
study, we included all teeth indicated for extrac-
tion during the period of the study if they met
the inclusion criteria. This method made it
impossible to control the number of sound and
carious teeth. The small number of sound teeth
in our study was not intentional, since we
selected the test site randomly. The optimal cut-
offs (Table 2) were lower for sound teeth com-
pared with the cutoffs proposed by the manufac-
turer, which increased sensitivity with a non-
significant loss in specificity at the D1threshold.
Considering that there are differences between
the cutoffs, using those proposed by the manu-
facturer can affect the interpretation of the LF
device and LF pen measurements and, conse-
quently, lead to overestimation of carious lesion
detection and overtreatment. Considering only
their ability to help detect enamel carious
lesions, both the LF device and the LF pen
helped detect more lesions correctly with use of
the optimal cutoffs than with use of the cutoffs
proposed by the manufacturer. At the D3
threshold, because the optimal and manufac-
turer’s cutoffs for dentin caries were similar for
the LF device and the LF pen, there was no sig-
nificant difference in terms of specificity, sensi-
tivity and accuracy values. When we compared
the optimal cutoffs with those proposed by the
manufacturers, we found that the optimal cut-
offs provided a great balance between specificity
and sensitivity, which can help improve the per-
formances of the LF device and the LF pen.
We investigated the FC’s ability to detect
occlusal caries in vivo for the first time, to our
knowledge. Regarding the optimal cutoffs, our
results showed that the values presented by the
software analysis were different from the values
proposed by the manufacturer. The software
highlighted the carious lesions in different
colors and defined their extent on a scale from 0
to 5: 1.0 to 1.5 (blue) indicated beginning
enamel caries, 1.5 to 2.0 (red) indicated deep
enamel caries, 2.0 to 2.5 (orange) indicated
dentin caries, and 2.5 to 5.0 (yellow) indicated
deep dentin caries. According to our cutoffs, the
range between sound teeth and enamel caries
was narrowed with the cutoffs proposed by the
manufacturer, as was the range between enamel
caries and dentin caries. These cutoffs should be
interpreted carefully, as a 0.1 variation could
cause trouble when making a treatment deci-
TABLE 2
Optimal cutoffs.
HISTOLOGIC
SCORE
METHOD (OPTIMAL CUTOFF)
International
Caries
Detection
and
Assessment
System
Bitewing
Radiography
Laser
Fluorescence
Device
Laser
Fluorescence
Pen
Fluorescence
Camera
0
000-40-40.0-1.2
1, 2
1-215-275-321.3
3, 4
3-62-328-9933-991.4-5.0
S T O R Y
C O V E R
Copyright © 2012 American Dental Association. All rights reserved.
on April 2, 2012
jada.ada.org
Downloaded from

Page 9

346
JADA 143(4) http://jada.ada.org April 2012
sion. Our data are comparable with those from
an in vitro study.8Although the FC helped
detect 99 percent of the enamel carious lesions
correctly with the use of the cutoffs proposed by
the manufacturer, specificity and sensitivity
values differed with the use of the optimal cut-
offs, indicating that those cutoffs should be used
carefully in clinical practice.
In our study, use of ICDAS criteria resulted
in high sensitivity and moderate specificity at
the D1threshold, confirming its ability to help
detect occlusal carious lesions in permanent
teeth, especially early clinically visible enamel
changes due to carious demineralization. This
result also has been found in laboratory
studies.5,7,9,10The use of ICDAS criteria showed
high accuracy and high Azvalues. Considering
only the ability of the ICDAS in helping detect
enamel carious lesions, it helped detect 66 per-
cent of those lesions correctly. The results indi-
cate that the ICDAS method can be used for
detecting occlusal caries. For visual inspection
using ICDAS criteria, we first analyzed the sur-
faces of the teeth when they were wet and then
after we carefully dried them with pressurized
air, which clearly showed the opaque appear-
ance of demineralization. At the D3threshold,
the ICDAS had moderate sensitivity and good
specificity. Investigators in an in vitro study
also reported a higher specificity than sensi-
tivity for the detection of dentin caries.7In con-
trast, investigators in two other studies
reported good sensitivity for in vitro dentin
caries detection.8,9These differences could be
explained by the clinical conditions of the visual
examination and the small number of dentin
lesions in our study, which probably indicates
the difficulty of distinguishing hidden caries (a
dentinal occlusal lesion with clinical signs of
caries activity in an apparently intact occlusal
surface) in permanent teeth. The correlation
between the ICDAS criteria and the histologic
scores in our study was not strong, as it has
been in other studies.7,9In another study, how-
ever, this correlation was stronger.10To our
knowledge, no data are available regarding the
clinical validity of using ICDAS criteria to
detect occlusal caries.
Radiographic examinations have been used
widely in dentistry to supplement visual inspec-
tion for caries.35,36Occlusal surface radiographs
are two-dimensional images of a three-
dimensional tooth, making difficult the detec-
tion of mineral loss in the outer part of the
TABLE 3
Specificity, sensitivity, accuracy, area under the receiver operating
characteristic curve (Az), positive likelihood ratio (LR+) and negative
likelihood ratio (LR− −), according to method for clinical detection
of occlusal caries at the D1and D3thresholds.*
METHODSPECIFICITYSENSITIVITYACCURACYAZ
LR+
LR− −
D1†
D3‡
D1
D3
D1
D3
D1
D3
D1
D3
D1
D3
International
Caries
Detection and
Assessment
System
0.60a
0.77a
0.93a
0.52a
0.91a
0.70a
0.86a
0.75a,b
2.332.250.120.63
Bitewing
Radiography
1.00a
0.97b
0.29b
0.44a
0.32b
0.84b
0.64b
0.74a,b
1.0011.560.710.86
Laser
Fluorescence
Device
1.00a
0.77a
0.85c
0.81b
0.86a
0.78a
0.94c
0.84a
4.403.530.150.24
Laser
Fluorescence
Pen
0.80a
0.71a
0.89a,c
0.85b
0.89a
0.74a
0.95c
0.79a,b
4.352.890.160.21
Fluorescence
Camera
0.80a
0.49c
0.74d
0.85b
0.74c
0.58c
0.79a
0.72b
1.191.660.250.33
* Significant differences are represented by different superscript letters in the same column (McNemar test, P < .05 for specificity, sensitivity
and accuracy; nonparametric statistical test for Az).
† Histologic score: 0 = sound, 1-4 = carious; International Caries Detection and Assessment System: 0 = sound, 1-6 = carious; bitewing
radiography: 0 = sound, 1-3 = carious; the laser fluorescence device/laser fluorescence pen/fluorescence camera: 0 = sound,
1-2 = carious.
‡ Histologic score: 0-2 = sound, 3-4 = carious; International Caries Detection and Assessment System: 0-2 = sound, 3-6 = carious; bitewing
radiography: 0-1 = sound, 2-3 = carious; the laser fluorescence device/laser fluorescence pen/fluorescence camera: 0-1 = sound, 2 = carious.
S T O R Y
C O V E R
Copyright © 2012 American Dental Association. All rights reserved.
on April 2, 2012
jada.ada.org
Downloaded from

Page 10

JADA 143(4) http://jada.ada.org April 2012 347
enamel that is hidden by
sound tooth structure37
or even the borderline
between enamel and
dentin.35However, the
results of one study
showed that radio-
graphy is a more sensi-
tive diagnostic method
than is clinical inspec-
tion for helping detect
both approximal and
occlusal lesions in
dentin, for estimating
lesion depth and for
monitoring lesion
behavior.38Radiography
may reveal a high per-
centage of hidden
caries.39-41At the D1and
D3thresholds, BW radi-
ography had the highest
specificity, along with
the LF device, and the
lowest sensitivity of all
the methods we investi-
gated. In agreement
with our study results,
some investigators also found higher
specificity and lower sensitivity.24,42
However, investigators in another
study reported moderate sensitivity
and high specificity at the D3
threshold.17In our study, BW radio-
graphy helped detect only 15 percent
of enamel carious lesions correctly,
which confirms its inability to help
detect incipient lesions. In addition,
BW radiography had the lowest
accuracy for enamel and dentin car-
ious lesion detection and the highest
accuracy in helping detect only
dentin caries. The Spearman rank
correlation coefficient was 0.48 for
BW radiography (the same as for the
ICDAS). Some investigators, how-
ever, have reported a lower correla-
tion for BW radiography and for
visual examination than for histo-
logic analysis.43
The LF device has demonstrated
superior sensitivity but inferior specificity than
for visual examination.3In contrast, we
observed that the LF device showed good speci-
ficity and accuracy. The LF device showed high
sensitivity and specificity for enamel and dentin
caries detection. Comparing our results with
those of other authors,44we observed that the
sensitivity level we obtained was higher than
the level obtained in their study. Concerning
dentin caries detection, the LF device showed
high sensitivity and specificity values. Similar
results were described in another clinical
TABLE 4
Specificity, sensitivity and accuracy for each
method’s optimal cutoffs and for the cutoffs
proposed by the manufacturers.*
METHOD, CUTOFFSPECIFICITY SENSITIVITYACCURACY
D1† D3‡D1
D3
D1
D3
Laser
Fluorescence
Device
Optimal
1.00a
0.77a
0.85a
0.81a
0.86a
0.78a
Manufacturer
1.00a
0.81a
0.69b
0.74a
0.70b
0.79a
Laser
Fluorescence Pen
Optimal
0.80a
0.71a
0.89a
0.85a
0.89a
0.74a
Manufacturer
1.00a
0.68a
0.67b
0.85a
0.69b
0.72a
Fluorescence
Camera
Optimal
0.80a
0.49a
0.74a
0.85a
0.74a
0.58a
Manufacturer
0.00b
0.99b
1.00b
0.11b
0.95b
0.76b
* Significant differences are represented by different superscript letters in the same column (McNemar
test, P < .05).
† Histologic score: 0 = sound, 1-4 = carious; laser fluorescence device/laser fluorescence pen/fluorescence
camera: 0 = sound, 1-2 = carious.
‡ Histologic score: 0-2 = sound, 3-4 = carious; laser fluorescence device/laser fluorescence pen/fluores-
cence camera: 0-1 = sound, 2 = carious.
TABLE 5
Percentage of enamel carious lesions
detected correctly by means of all of the
methods and different cutoffs.
METHOD, CUTOFFENAMEL LESIONS CORRECTLY DETECTED,
NO. (%) (N = 73)
International Caries
Detection and
Assessment System
48 (66)
Bitewing Radiography
11 (15)
Laser Fluorescence Device
Optimal
40 (55)
Manufacturer
28 (38)
Laser Fluorescence Pen
Optimal
39 (53)
Manufacturer
17 (23)
Fluorescence Camera
Optimal
10 (14)
Manufacturer
72 (99)
S T O R Y
C O V E R
Copyright © 2012 American Dental Association. All rights reserved.
on April 2, 2012
jada.ada.org
Downloaded from

Page 11

348
JADA 143(4) http://jada.ada.org April 2012
study.42However, some study results showed
higher specificity values,17,20whereas the results
of other studies showed lower specificity
values.18,24The correlation of the LF device
measurements with the histologic scores was
better than for the ICDAS and BW radiography
in our study and another.43
The LF pen performed the same as the LF
device; there was no significant difference
between them. At the D1and D3thresholds, the
LF pen had high sensitivity and specificity
values. Other investigators also reported high
specificity values.19The results from other
studies, however, showed lower sensitivity
values for dentin caries detection in vivo.18,19
This finding may be due to the different cutoffs
used for performance calculation. The LF pen
also had high accuracy, especially in enamel and
dentin caries detection. These results are con-
sistent with those reported in an in vivo study.19
Although the LF pen had the highest correla-
tion with histologic scores, that correlation still
is a moderate one.
A technique that offers high specificity even
at the expense of a reduction in sensitivity
seems to be the best option.45The specificities of
1.00 (LF device) and 0.80 (LF pen) that we
observed in our study at the D1threshold means
that no or fewer sites were scored incorrectly as
carious, leading to the avoidance of false-
positive interpretations and overtreatment.
The fact that the use of LF methods results in
higher specificities must be interpreted care-
fully in clinical practice given the small number
of sound tooth sites observed in the histologic
analysis.
The FC showed good validity for helping
detect carious lesions. The sensitivity values
were high at the D1and D3thresholds. However,
the sensitivity for enamel caries and the speci-
ficity for dentin caries were substantially lower
when compared with those of the LF device and
the LF pen. For dentin caries, the sensitivity
was similar for all three methods. These results
could be expected, considering that the FC helps
detect increases in fluorescence by using the
same principle as do the LF device and the LF
pen.8,25Investigators conducted two in vitro
studies to evaluate the FC’s performance in
helping detect occlusal carious lesions in perma-
nent teeth.8,26In agreement with results from
one of these studies,8our results showed a weak
correlation of the FC scores with the histologic
scores. However, investigators in a recent study
found a moderate correlation for the FC and his-
tologic scores.26
Because Rodrigues and colleagues’8study was
the first to our knowledge in which the perform-
ance of the FC was evaluated in vitro, it is
important to discuss the FC’s advantages and
limitations to provide clinicians with a better
understanding of this method. In FC images,
large areas of dental surfaces can be inspected
within seconds, which is not possible with the
LF device or the LF pen, which use only a fiber-
optic probe to check fluorescence intensities at a
single point on the tooth surface. Compared
with in vitro experimental conditions, the use of
the FC in vivo has drawbacks, including access to
lesions on occlusal surfaces located more posteri-
orly in patients with limited mouth opening,
moisture in the oral environment and light-source
angulations. Some factors may influence the
quality of the FC image recording caries status,
such as the presence of plaque, calculus or
staining on the tooth surface; saliva; and the com-
plex light-scattering pattern generated on
occlusal surfaces, which can make the reconstruc-
tion of reliable fluorescence values difficult.
Ambient light, daylight or office light also may
influence FC image quality. Thus, images should
be captured in partial blackout conditions.
The fluorescence-based methods showed good
agreement with the histologic scores (80-100
percent) in helping detect sound occlusal sites.
However, this ability is not representative of the
whole population because there were only five
sound teeth in our study. Both the LF device
and the LF pen showed moderate agreement in
helping detect enamel carious lesions (55 and 53
percent, respectively), whereas the FC showed
poor agreement (14 percent). Discolored sites
tend to be overscored with the LF device,
resulting in a high rate of false-positive
results.24In our study, there was a tendency for
the LF device, the LF pen and the FC to lead to
underestimation or overestimation of enamel
carious lesions. This finding must be considered
carefully when one is using only fluorescence-
based readings to make treatment decisions in
clinical practice, because there will be a higher
risk of providing undertreatment or overtreat-
ment. Determining what type of intervention or
restorative treatments to provide depends on a
range of other variables, such as the patient’s
history and fluoride and dietary statuses, as
well as caries activity assessment and the
status of the tooth surface. Although fluores-
cence-based methods seem to be suitable for
detecting occlusal caries in a preselected popu-
lation of specific areas on occlusal surfaces in
permanent teeth, they should be used to provide
a second opinion in clinical practice.
Each method had its own LR+ and LR−,
S T O R Y
C O V E R
Copyright © 2012 American Dental Association. All rights reserved.
on April 2, 2012
jada.ada.org
Downloaded from

Page 12

JADA 143(4) http://jada.ada.org April 2012 349
which provided the odds of caries increase or
decrease in patients when a test result was posi-
tive or negative, respectively. At the D1thresh-
old, the LF device and the LF pen, followed by
the ICDAS, had higher LR+ values and lower
LR− values, indicating that their results were
associated with both presence and absence of
caries. At the D1threshold, BW radiography had
the lowest LR+ value (1.00) and the highest LR−
value (0.71), indicating that this method cannot
help distinguish healthy from carious teeth cor-
rectly. On the other hand, at the D3threshold,
BW radiography had the highest LR+ value
(11.56) and the FC had the lowest (1.66).
The Azvalues showed that the LF device and
the LF pen, followed by the ICDAS, were the
most accurate methods to use for occlusal caries
detection. At the D1threshold, the LF device
and the LF pen had the highest Azvalues com-
pared with the other methods. In contrast, at
the D3threshold, we found significant differ-
ences between the LF device, the LF pen or both
and FC Azvalues. The results from some in vivo
studies also have shown higher Azvalues for the
LF device16,20and for the LF pen.19
The assessment of dental caries is associated
with considerable interexaminer and intraex-
aminer variability, and it should be the first
item to be considered when evaluating caries
detection methods.46In our study, we did not
evaluate the intraexaminer and interexaminer
reproducibility of the conventional and
fluorescence-based methods because of the short
amount of time available to perform the clinical
examinations before the teeth were extracted.
We also analyzed more than one tooth in some
patients, which was another difficulty regarding
time for assessments. Investigators in other
clinical studies also did not assess intraexam-
iner and interexaminer reproducibility.18,20,47
To date, the number of clinical studies in
which investigators evaluate and validate
methods of carious lesion defects is limited.
Therefore, clinical studies should be conducted
to evaluate the performance of fluorescence-
based methods of caries detection, especially the
LF pen and the FC, and of the ICDAS criteria
for both detection of caries and caries activity
assessments.
CONCLUSIONS
We found that the clinical cutoffs for sound
teeth and enamel and dentin caries were,
respectively, 0 through 4, 5 through 27 and 28
through 99 (LF device); 0 through 4, 5 through
32 and 33 through 99 (LF pen); and 0 through
1.2, 1.3 and 1.4 through 5.0 (FC). However, the
cutoffs should not be considered exact threshold
measurements. We also found that ICDAS, the
LF device and the LF pen demonstrated good
performance in helping detect occlusal caries in
vivo. ICDAS did not seem to perform as well at
the D3threshold as it did at the D1threshold.
BW radiography and the FC had the lowest per-
formances in helping detect lesions at the D1
and D3thresholds. ■
Disclosure. None of the authors reported any disclosures.
The authors thank Coordenação de Aperfeiçoamento de Pessoal de
Nível Superior (CAPES) and Fundação de Amparo à Pesquisa do
Estado de São Paulo (FAPESP 2009/00218-0) for their financial
support.
This study was submitted to the School of Dentistry of Araraquara,
UNESP—Universidade Estadual Paulista, São Paulo, as a require-
ment for the a doctoral degree in pediatric dentistry for Dr. Diniz.
The authors thank the Department of Oral and Maxillofacial Surgery
of the School of Dentistry of Araraquara, UNESP—Universidade
Estadual Paulista, São Paulo, especially Isabela Cristine Manzolli
Rodrigues and the residents for their clinical assistance.
1. Ekstrand KR, Ricketts DN, Kidd EA. Reproducibility and accu-
racy of three methods for assessment of demineralization depth of
the occlusal surface: an in vitro examination. Caries Res 1997;
31(3):224-231.
2. Nyvad B, Machiulskiene V, Baelum V. Reliability of a new caries
diagnostic system differentiating between active and inactive caries
lesions. Caries Res 1999;33(4):252-260.
3. Bader JD, Shugars DA. A systematic review of the performance
of a laser fluorescence device for detecting caries. JADA 2004;
135(10):1414-1426.
4. Zandoná AF, Zero D. Diagnostic tools for early caries detection
(published correction appears in JADA 2007;138[3]:298). JADA
2006;137(12):1675-1684.
5. Ekstrand KR, Martignon S, Ricketts DJ, Qvist V. Detection and
activity assessment of primary coronal caries lesions: a methodologic
study. Oper Dent 2007;32(3):225-235.
6. Ismail AI, Sohn W, Tellez M, et al. The International Caries
Detection and Assessment System (ICDAS): an integrated system for
measuring dental caries. Community Dent Oral Epidemiol 2007;
35(3):170-178.
7. Jablonski-Momeni A, Stachniss V, Ricketts DN, 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.
8. Rodrigues JA, Hug I, Diniz MB, Lussi A. Performance of fluores-
cence methods, radiographic examination and ICDAS II on occlusal
surfaces in vitro. Caries Res 2008;42(4):297-304.
9. Diniz MB, Rodrigues JA, Hug I, Cordeiro Rde C, Lussi A. Repro-
ducibility and accuracy of the ICDAS-II for occlusal caries detection.
Community Dent Oral Epidemiol 2009;37(5):399-404.
10. Zandona AG, Al-Shiha S, Eggertsson H, Eckert G. Student
versus faculty performance using a new visual criteria for the detec-
tion of caries on occlusal surfaces: an in vitro examination with histo-
logical validation. Oper Dent 2009;34(5):598-604.
11. Jablonski-Momeni A, Ricketts DN, Weber K, et al. Effect of dif-
ferent time intervals between examinations on the reproducibility of
ICDAS-II for occlusal caries. Caries Res 2010;44(3):267-271.
12. Hibst R, Paulus R, Lussi A. Detection of occlusal caries by laser
fluorescence: basic and clinical investigations. Med Laser Appl
2001;16(3):205-213.
13. Lussi A, Hellwig E. Performance of a new laser fluorescence
device for the detection of occlusal caries in vitro. J Dent 2006;
34(7):467-471.
14. Lussi A, Hack A, Hug I, Heckenberger H, Megert B, Stich H.
Detection of approximal caries with a new laser fluorescence device.
Caries Res 2006;40(2):97-103.
15. Lussi A, Imwinkelried S, Pitts N, Longbottom C, Reich E. Per-
formance and reproducibility of a laser fluorescence system for detec-
tion of occlusal caries in vitro. Caries Res 1999;33(4):261-266.
16. Heinrich-Weltzien R, Kühnisch J, Oehme T, Ziehe A, Stösser L,
S T O R Y
C O V E R
Copyright © 2012 American Dental Association. All rights reserved.
on April 2, 2012
jada.ada.org
Downloaded from

Page 13

350
JADA 143(4) http://jada.ada.org April 2012
García-Godoy F. Comparison of different DIAGNOdent cut-off limits
for in vivo detection of occlusal caries. Oper Dent 2003;28(6):672-680.
17. Lussi A, Megert B, Longbottom C, Reich E, Francescut P. Clin-
ical performance of a laser fluorescence device for detection of
occlusal caries lesions. Eur J Oral Sci 2001;109(1):14-19.
18. Krause F, Jepsen S, Braun A. Comparison of two laser fluores-
cence devices for the detection of occlusal caries in vivo. Eur J Oral
Sci 2007;115(4):252-256.
19. Huth KC, Neuhaus KW, Gygax M, et al. Clinical performance of
a new laser fluorescence device for detection of occlusal caries lesions
in permanent molars. J Dent 2008;36(12):1033-1040.
20. Diniz MB, Rodrigues JA, de Paula AB, Cordeiro Rde C. In vivo
evaluation of laser fluorescence performance using different cut-off
limits for occlusal caries detection (published correction appears in
Lasers Med Sci 2009;24[3]:461). Lasers Med Sci 2009;24(3):295-300.
21. Sheehy EC, Brailsford SR, Kidd EA, Beighton D, Zoitopoulos L.
Comparison between visual examination and a laser fluorescence
system for in vivo diagnosis of occlusal caries. Caries Res 2001;
35(6):421-426.
22. Anttonen V, Seppä L, Hausen H. Clinical study of the use of the
laser fluorescence device DIAGNOdent for detection of occlusal caries
in children. Caries Res 2003;37(1):17-23.
23. Akarsu S, Köprülü H. In vivo comparison of the efficacy of
DIAGNOdent by visual inspection and radiographic diagnostic tech-
niques in the diagnosis of occlusal caries. J Clin Dent 2006;17(3):
53-58.
24. Angnes V, Angnes G, Batisttella M, Grande RH, Loguercio AD,
Reis A. Clinical effectiveness of laser fluorescence, visual inspection
and radiography in the detection of occlusal caries. Caries Res
2005;39(6):490-495.
25. Thoms M. Detection of intraoral lesions using a fluorescence
camera. In: Rechmann P, Fried D, eds. Proceedings SPIE 6137:
Lasers in Dentistry XII. Bellingham, Wash.: The International
Society for Optical Engineering; 2006:1-7.
26. Jablonski-Momeni A, Schipper HM, Rosen SM, et al. Perform-
ance of a fluorescence camera for detection of occlusal caries in vitro.
Odontology 2011;99(1):55-61.
27. De Benedetto MS, Morais CC, Novaes TF, de Almeida
Rodrigues J, Braga MM, Mendes FM. Comparing the reliability of a
new fluorescence camera with conventional laser fluorescence
devices in detecting caries lesions in occlusal and smooth surfaces of
primary teeth. Lasers Med Sci 2011;26(2):157-162.
28. Reis A, Mendes FM, Angnes V, Angnes G, Grande RH,
Loguercio AD. Performance of methods of occlusal caries detection in
permanent teeth under clinical and laboratory conditions. J Dent
2006;34(2):89-96.
29. World Medical Association. WMA Declaration of Helsinki: Eth-
ical principles for medical research involving human subjects.
www.wma.net/en/30publications/10policies/b3/index.html. Accessed
Feb. 29, 2012.
30. Anttonen V, Seppä L, Hausen H. Clinical study on the effect of
professional cleaning of occlusal tooth surfaces on laser fluorescence
measurements. Caries Res 2005;39(4):280-283.
31. Diniz MB, Sciasci P, Rodrigues JA, Lussi A, Cordeiro RC. Influ-
ence of different professional prophylactic methods on fluorescence
measurements for detection of occlusal caries. Caries Res 2011;
45(3):264-268.
32. Diniz MB, Lima LM, Santos-Pinto L, Eckert GJ, Zandoná AG,
de Cássia Loiola Cordeiro R. Influence of the ICDAS e-learning pro-
gram for occlusal caries detection on dental students. J Dent Educ
2010;74(8):862-868.
33. Diniz MB, Rodrigues JA, Hug I, Cordeiro RC, Lussi A. The
influence of pit and fissure sealants on infrared fluorescence meas-
urements. Caries Res 2008;42(5):328-333.
34. Khalife MA, Boynton JR, Dennison JB, Yaman P, Hamilton JC.
In vivo evaluation of DIAGNOdent for the quantification of occlusal
dental caries. Oper Dent 2009;34(2):136-141.
35. Espelid I, Tveit AB, Fjelltveit A. Variations among dentists in
radiographic detection of occlusal caries. Caries Res 1994;28(3):
169-175.
36. Dove SB. Radiographic diagnosis of dental caries. J Dent Educ
2001;65(10):985-990.
37. Espelid I, Tveit AB. A comparison of radiographic occlusal and
approximal caries diagnoses made by 240 dentists. Acta Odontol
Scand 2001;59(5):285-289.
38. Wenzel A. Bitewing and digital bitewing radiography for detec-
tion of caries lesions. J Dent Res 2004;83(special number C):
C72-C75.
39. Hintze H. Screening with conventional and digital bite-wing
radiography compared to clinical examination alone for caries detec-
tion in low-risk children. Caries Res 1993;27(6):499-504.
40. Hintze H, Wenzel A, Jones C. In vitro comparison of D- and
E-speed film radiography, RVG, and visualix digital radiography for
the detection of enamel approximal and dentinal occlusal caries
lesions. Caries Res 1994;28(5):363-367.
41. Weerheijm KL. Occlusal ‘hidden caries’. Dent Update
1997;24(5):182-184.
42. Chu CH, Lo EC, You DS. Clinical diagnosis of fissure caries
with conventional and laser-induced fluorescence techniques. Lasers
Med Sci 2010;25(3):355-362.
43. Toraman Alkurt M, Peker I, Deniz Arisu H, Bala O,
Altunkaynak B. In vivo comparison of laser fluorescence measure-
ments with conventional methods for occlusal caries detection.
Lasers Med Sci 2008;23(3):307-312.
44. Barbería E, Maroto M, Arenas M, Silva CC. A clinical study of
caries diagnosis with a laser fluorescence system. JADA 2008;
139(5):572-579.
45. Downer MC. Validation of methods used in dental caries diag-
nosis. Int Dent J 1989;39(4):241-246.
46. Kühnisch J, Bücher K, Henschel V, Hickel R. Reproducibility of
DIAGNOdent 2095 and DIAGNOdent Pen measurements: results
from an in vitro study on occlusal sites. Eur J Oral Sci 2007;
115(3):206-211.
47. Abalos C, Herrera M, Jiménez-Planas A, Llamas R. Perform-
ance of laser fluorescence for detection of occlusal dentinal caries
lesions in permanent molars: an in vivo study with total validation of
the sample. Caries Res 2009;43(2):137-141.
S T O R Y
C O V E R
Copyright © 2012 American Dental Association. All rights reserved.
on April 2, 2012
jada.ada.org
Downloaded from