on DD measurements was examined, after their
application to occlusal surfaces of molars and
premolars. While toothpaste did not affect at all
the DD values, seven of the polishing pastes have
an effect on the measurements with pumice being
the leading one. It seems that the intense auto-
fluorescence of certain polishing pastes may alter
the DD measurements, since their components
cannot be completely removed from the pits and
fissures of the occlusal surfaces of posterior teeth
even after brushing and rinsing. 29 Also, Lussi et al. 30
in another in vitro study examined the influence
of various toothpastes and prophylaxis paste
remnants, as well as, powder remnants influencing
DD readings. The results of this experiment showed,
that only one toothpaste (Nupro mint/cherry
medium, Dentsply De Trey, USA) and one polishing
paste (Clinic, 3M, Bioggio, Switzerland) had a
statistically significant effect on the measurements
(p<0.01), after rinsing for 3-6 seconds. These
formulations contain sticky elements, which in
combination with the high porosity of the decayed
tissue, are not sufficiently removed and thereby
increasing the DD measurements. If the teeth are
not intensely rinsed with water-air combination for
at least ten seconds, an incorrect assessment may
occur. This is more significant for the long term
monitoring of lesions, rather than the detection of
lesion per se. 30
3.1.3. Effect of sample storage means in the DD
measurements
The different storage means of the samples used
in laboratory studies, such as chloramine solutions,
formalin and thymol affect the final measurements
of DD. 31 Kaul et al. 32 used 90 extracted molars in
groups of ten and stored the eight groups in eight
different solutions
and one of them in a frozen
o
state of -20 C for one year. It was shown that the
most reliable method for teeth storing was the
frozen state. According to this statement it has to
be noted that only a few in vitro studies so far have
used samples that were stored in a frozen state, a
fact that should have an impact upon the clinical
interpretation of the results.
3.1.4. Accuracy and repeatability of DD
The characteristics of accuracy and repeatability of
the measurements of the DD and DDPen devices
are well documented. Chu et al. 33 mention that
different COV values show different results. In an in
vivo study using COV by Lussi et al., 12 the sensitivity
(0.95) and specificity (0.11) differ considerably,
while for COV=40, the sensitivity (0.70) and
specificity (0.84) differ less. The authors propose
the combination of visual observation with the use
of DD for caries detection, as it offers better results
in terms of specificity and quite good results in
respect to sensitivity.
Jablonski-Momeni et al., 5 in an in vitro study,
examined 181 points of 100 posterior teeth
comparing the DD detection capability with that
of direct visual observation during ICDASII. The
Stomatology Edu Journal
repeatability of the measurements for the DD
between examiners was very high (0.957). Enamel
(D1) and middle dentin (D3) have a specificity of
D1:0.54, and D3:0.91 respectively, whereas the
sensitivity was D1: 0.91 and D3: 0.70. Therefore,
the ICDASII values were higher than those of DD.
The researchers conclude that combining ICDASII
and DD investigating methods provide better
diagnostic results.
The first combined in vivo/in vitro study for the DD
device was conducted by Reis et al. 35 who studied
the caries detection of 57 third molars, both
by direct visual observation and DD. The direct
visual observation showed almost double in vivo
and in vitro repeatability, both between different
examiners (IR) (0.559) and between measurements
of the same examiner (IA) (0.559) compared to
that of DD. This study shows higher sensitivity of
DD measurements than the visual method, which
is not an usual finding in laboratory studies. The
presence of pigments in pits and fissures of the
occlusal surfaces may explain the above finding.
The authors suggest using 19-20 COV for the
differentiation of healthy versus carious dentin.
They also proposed that the visual observation
using ICDASII is quite a reliable caries detection
system. 5
Also in the study of De Paula et al., 35 visual
observation gave higher precision values than
the DD. These findings are in agreement with the
results by Rodrigues et al. and Agnes et al. 36,16 The
combination of detection techniques e.g. visual
observation, radiography and DD seems though
to result to more accurate diagnosis of caries as
mentioned also elsewhere. 37 But it should be noted
that the actual clinical experience of the operator
can affect the objectivity of the detection, either by
visual observation or by using devices such as DD.
Specifically, in a laboratory study, 3 undergraduate
dental students, 3 general dentists and 3
academics were asked to evaluate 25 molars by
visual observation and by using DD.
The results showed a substantial variation. The
sensitivity of the measurements ranged from 0.188
to 0.769 and the specificity from 0.714 to 0.969.
The group of the academics recorded the highest
sensitivity of DD (0.667), while the group of the
general dentists the highest specificity (0.942).
A substantial variation of measurements occurred
in respect of sensitivity (0.755-0.953) and
specificity (0.755-0.953) of the visual observation,
with the students reaching the greatest sensitivity
(0.80). 38 Ideally, a detection technique for the
occlusal surfaces must have a very high sensitivity
for D3 and D4 and moderate high specificity for
detecting enamel caries.
The DD shows higher specificity for lesions at the
level of dentin and lower for enamel lesions, since
it is unable to identify the healthy tissues from t he
carious ones extending to the half of the enamel.
CARIES DETECTION WITH LASER FLUORESCENCE DEVICES. LIMITATIONS OF THEIR USE
49