The European Journal of Orthodontics Advance Access published October 4, 2016

European Journal of Orthodontics, 2016, 1–10

doi:10.1093/ejo/cjw060

Randomized controlled trial

Changes in white spot lesions following

post-orthodontic weekly application of 1.25
per cent fluoride gel over 6 months—a
randomized placebo-controlled clinical trial.
Part I: photographic data evaluation
Niko C. Bock1, Laura Seibold2, Christian Heumann3, Erhard Gnandt4,
Miriam Röder5 and Sabine Ruf1
1
Department of Orthodontics, University of Giessen, 2Private Practice, Hadamar, 3Department of Statistics, University
of Munich, 4Private Practice, Friedberg and 5Private Practice, Wächtersbach, Germany

Correspondence to: Niko C. Bock, Department of Orthodontics, University of Giessen, Schlangenzahl 14, 35392 Giessen,

Downloaded from by guest on October 17, 2016

Germany. E-mail: niko.c.bock@dentist.med.uni-giessen.de

Summary
Background:  White spot lesions (WSLs) are a frequent side-effect of multibracket appliance
treatment. The effect of local fluoridation on post-orthodontic WSL is however inconclusive.
Objective:  Assessment of WSL changes in response to weekly 1.25 per cent fluoride gel application
after multibracket appliance treatment.
Trial design:  Randomized, single-centre, double-blind, parallel-group, placebo-controlled study.
Participants:  Patients with not less than 1 WSL (modified score 1 or 2) on not less than 1 upper
front teeth after debonding.
Interventions:  Professional fluoride/placebo gel application during weeks 1–2; self-administered
home application (weeks 3–24).
Outcomes:  Photographic WSL assessment (dimension and luminance) of the upper front teeth (T0–T5).
Randomization:  Random assignment to test (n = 23) or placebo group (n = 23) using a sequentially
numbered list (random allocation sequence generated for 50 subjects in 25 blocks of 2 subjects
each).
Recruitment:  The clinical study duration lasted from March 2011 to September 2013.
Blinding:  Unblinding was performed after complete data evaluation.
Numbers analysed:  Intent-to-treat analysis set comprising 39 participants (test: n = 21, placebo: n = 18).
Outcome:  Dimensional WSL quantification showed limited reliability. Luminance improvement
(%) of WSL, however, was seen after 6 months (test/placebo: tooth 12, 24.8/18.0; tooth 11, 38.4/35.4;
tooth 21, 39.6/38.3; and tooth 22, 15.2/25.0). No statistically significant group difference existed.
Data suggest that WSLs are difficult to measure with respect to reliability and repeatability and
methods for monitoring WSLs in clinical trials require improvement/validation.
Harms:  Similar adverse events occurred in both groups; none was classified as possibly related
to the study product.
Limitations:  The number of dropouts was higher than expected and the socio-economic status
was not assessed. Furthermore, the unknown level of compliance during the home application
phase must be considered as limitation.

© The Author 2016. Published by Oxford University Press on behalf of the European Orthodontic Society. All rights reserved.
1
For permissions, please email: journals.permissions@oup.com
2 European Journal of Orthodontics, 2016
Conclusion:  Based on the results of this study, no difference could be detected with respect to the
development of WSL under post-orthodontic high-dose fluoride treatment.
Registration:  The study was registered with ClinicalTrials.gov (Identifier: NCT01329731).
Protocol:  The protocol wasn’t published before trial commencement.
Introduction
White spot lesions (WSLs) are one of the most undesired iatrogenic
side-effects of orthodontic treatments (Txs) involving multibracket
appliances (MBAs). These labial surface demineralizations have been
reported to occur in up to 96 per cent of individuals undergoing
MBA Tx (1–10). Thus, subjects undergoing MBA Tx are more sus-
ceptible to WSLs (11) and caries development (12) than untreated
subjects.
Initial enamel decalcification can be seen as early as 4 weeks after
the beginning of MBA Tx (13, 14). The incidence for WSL on upper
front teeth (UFT) during MBA Tx has been reported to reach 41
per cent of the patients (15). The lateral incisors seem to have about
twice the risk compared with the central incisors (16).
WSLs usually cease to develop after removal of the MBA because
the cariogenic challenge has been removed (17). In addition, inac-
tive incipient carious lesions may regress and become less promi-
nent (17–19). However, they tend to remain visible and may be of
aesthetic concern even 5 years after Tx (11), particularly if they are
extensive (17) or exhibit aggravation.
The following risk factors have been revealed: previous WSL and
poor oral hygiene before Tx (20) as well as male gender, young age
at the start of Tx, long Tx duration, missed appointments, and poor
oral hygiene during Tx (16). During the last decades, different meas-
ures besides the need for excellent oral hygiene (5, 21, 22) have been
introduced to prevent WSL development during MBA Tx. Especially
the oral application of fluoride products is generally accepted to
reduce the risk of WSL development (23–27). In addition, fluorides
have an antibacterial effect (28–30) which is especially true for high-
dose fluoride products (31).
With the aim to enhance enamel remineralization, the effect of
local fluoridation on WSLs has been investigated using different
types of fluorides and various modes of application (30, 32–34).
Despite these earlier studies, evidence is still lacking with respect to
both the fluoride concentration enabling optimal remineralization
(35) and the application mode (36) offering the best chance for WSL
remineralization.
Objective
The study objective was to monitor the effect of an intraoral high-con-
centration fluoride application on the development of WSLs after ortho-
dontic MBA Tx (based on photographical assessment). Null hypothesis
(H0): no difference between high-dose fluoride gel and placebo.
Trial design
Single-centre prospective randomized placebo-controlled double-
blind clinical trial.
Subjects and methods
Ethical issue
Before start, the study protocol was approved by the Ethics
Committee of the University of Giessen, Germany and the Federal
Institute for Drugs and Medical Devices of Germany. In addition, as
required by law, the local authorities were acquainted of the study,
which was performed in compliance with Good Clinical Practice
(GCP) and the Declaration of Helsinki.
Registration
The study was registered with ClinicalTrials.gov (Identifier:
NCT01329731).
Patient screening
Healthy MBA patients (11 years and older) from the Department of
Orthodontics, University of Giessen, Germany, who had received Tx
including the upper jaw for at least 1 year and were scheduled for
MBA removal were asked to participate.
Inclusion criteria
- Written informed consent;
Downloaded from by guest on October 17, 2016
- Not less than 1 WSL with a modified score 1 or 2 (4) on not less
than 1 UFT (in case of lateral incisor aplasia, the replacing canine
was taken into account), at debonding; WSL not present on pre-Tx
intraoral photographs;
- No restorative or prosthetic therapy planned on UFT; and 
- Agreement not to use any additional oral hygiene products for
the whole study duration than the ones received in the study
(allowed: oral irrigator, dental floss, and interdental brushes).
Exclusion criteria
- WSL with a modified score 3 (4) on 1 UFT or greater at debond-
ing;
- Ongoing oral or dental Tx except emergency Tx;
- Known hypersensitivity/allergy to study products and/or materials
used;
- Professional administration of highly concentrated fluoride prod-
ucts within 30 days prior to enrolment;
- Professional administration or home use of highly concentrated
fluoride products not related to the study during the whole study;
- Alterations of the enamel, e.g. hypoplasia, fluorosis;
- Chronic use of medication causing a dry mouth or known xeros-
tomia;
- Medication for central nervous system conditions;
- Any illness/condition potentially affecting the study outcome;
- Known pregnancy or breastfeeding during the course of the study;
and
- Participation in a clinical trial or receipt of an investigational
medication/Tx within the last 30 days
Randomization, blinding, and study products
The subjects were randomly assigned to either a test group (high-
dose fluoride gel) or a placebo group. Participants who discontinued
the study prior to termination of the first week (before T2) were
replaced (Figure 1).
N. C. Bock et al. 3

instructed to refrain from eating and drinking for 1 hour. After the
Eligible patients
third appointment (T2), the participants were instructed to apply the
n=210
gel by themselves at home, brushing it onto all teeth for 2 minutes
with their regular toothbrush once a week in the evening after regu-
lar oral hygiene. They were instructed to spit out the remaining foam
Declined to participate and rinse their mouth and refrain from eating and drinking until the
n=162
next morning. A written instruction was provided.
At each visit, the participants were asked to fill out a question-
naire concerning medication, dental or medical Tx, and the use of
Randomised additional fluoride products. In addition, they had been instructed to
n=48
Drop out during Drop out during bring the tube containing the study product to every appointment in
week 1 -> replaced week 1 -> replaced order to assess compliance by verifying the weight of the tube.
n=1 n=1
Photographs of the UFT were taken in a dark room without nat-
ural light, using a Nikon digital camera D300 (CMOS Sensor with
Test group Control group
(High-dose fluoride) (Placebo) 12.3 megapixel). A head holder was used to minimize head rotation
n=23 n=23 and/or inclination differences. The participants were instructed to
take their habitual occlusion, and the cheeks were retracted using
Follow-up a cheek retractor. The camera was fixed in a stand (Manfrotto
055XPRO B, Junior Geared head 410)  at a constant distance
between the lens and teeth. The lens was a Nikon AF-S (105/2.8G
Data sets Data sets
n=23 n=23 VR IF-ED Micro) and the flash a Nikon SB-29S TTL ring flash. No
polarized filter was used. The digital data were transferred to a PC
using Nikon Transfer 1.5.0 software. Digital RAW data were devel-
Missed visit T4 Missed visit T4 oped in Nikon Capture NX2 software. The images were adjusted
Data analysis
n=2 n=5
for constant brightness, chrominance, and contrast and were saved
as TIFF data files. In addition, a hardware-calibrated PC monitor

Downloaded from by guest on October 17, 2016

ITT- Analysis ITT- Analysis
(6500k) was used to ensure standardized settings.
n=21 n=18

Figure 1.  Flow chart of the study patients.
Enrolment and assignment were undertaken by blinded staff
of the Department of Orthodontics, XXX University of Giessen,
Germany, and the clinical duration of the study lasted from March
2011 (first recruitment) to September 2013 (completion).
Colgate-Palmolive Europe provided the necessary amount of
study products in sequentially numbered containers (blinded).
The random allocation sequence had been generated by Colgate-
Palmolive Europe, creating a list for 50 subjects randomized into
25 blocks of 2 subjects (the respective list remained with Colgate-
Palmolive Europe until all study parameters had been fully assessed).
In addition, toothbrushes and amine fluoride-containing toothpaste
as well as oral hygiene instructions (the participants were asked to
brush their teeth at least twice daily) and stopwatches were provided
to the participants.
Treatment protocol and process
Approximately 1 cm high-dose fluoride gel (1.25 per cent fluoride
including 1 per cent fluoride from sodium fluoride and 0.25 per cent
fluoride from olaflur/dectaflur; approximately 0.5 g gel or 6.25 mg
fluoride in line with usual recommendations) or placebo gel (same
formulation as high-dose fluoride gel except fluoride components
but additional parabens as preservatives; approximately 0.5 g gel)
was used per application. Overall, six visits were scheduled (Table 1).
At baseline and at the two following visits (T0 to T2), the opera-
tors applied the gel in a standardized way. Cotton rolls were placed
labial in the upper jaw, and the teeth were dried with compressed
air for 10 seconds. The gel was applied using a rotating rubber cup
by the operator who brushed the respective teeth for 20 seconds per
tooth. The participants were asked to rinse after 3 minutes and were
Study parameters
Primary outcome
WSL changes in dimension (DWL%) over the 6  months were to
be quantified using a computer-assisted analysis. Based on studies
by Mattousch et  al. (37) and Ferreira et  al. (38), we assumed the
extent of remineralization to double and lesion size to halve within
3 months (T4) of Tx (primary study parameter).
All measurements were performed twice by the same calibrated
and blinded investigator (LS) with a time interval of at least 2 weeks
between the measurements.
The WSL changes were to be assessed from digital intraoral
photographs obtained for each study participant at each assess-
ment point using the software Image Pro Plus (Version 7.0, Media
Cyberkinetics, Santa Clara, California, USA). The image analysis
software was set to calculate the area of the WSL in relation to the
total tooth surface. The outline of each of the four UFT was traced
by means of a freehand tool with the computer mouse. The soft-
ware calculated the pixel size to determine the total area of each
tooth. Afterwards, the software calculated the size of the WSL by
automatically detecting the brightest area of the tooth surface. To
exclude all areas that were not part of the WSL (such as reflections
being caused by the flash light), the borderlines of the WSL were
corrected manually.
Secondary outcome
As the method of WSL quantification proved to have limited reliabil-
ity, a second endpoint (WSL luminance) was introduced prior to data
unblinding. This endpoint was chosen according to the following
considerations: the severity/visibility of an enamel demineralization
is proportional to the degree of the mineral loss of the enamel which,
in turn, changes its optical properties (39–43). Thus, the severity of
the WSLs was to be quantified on the basis of their colour changes
4 European Journal of Orthodontics, 2016
Table 1.  Course of the study. The visits T0 to T5 and the respective measures that were undertaken are given.
Visit T0 T1
Study week 0 1 (+3 days, −1 day)
Study phase Baseline Professional treatment
Multibracket appliance removal X
Inclusion/exclusion criteria X
Demographics X
General health status X X
Medical history X
Concomitant therapy/treatment* X
Randomization X
Application of study product X X
Oral hygiene instructions X X
Adverse events X X
*Including therapies/treatments within 30 days prior to enrolment.
according to the Commission Internationale de l’Éclairage (CIE)
Lab system. This method has been shown to be an objective tool for
tooth colour determination (44, 45). The software used to detect the
changes in luminance was Adobe Photoshop CS5 Extended (Version
12.0x64), licensed for medical use.
For luminance assessment, all pictures were evaluated in real
pixel size (100 per cent). The outline of each tooth was traced by
means of a freehand tool using the computer mouse. The software
calculated the number of pixels and a respective RGB value as well
as mean, standard deviation, and central value. In addition, the
mean luminance was measured for each tooth’s total labial surface.
Afterwards, the luminance of the WSL was measured by identify-
ing its brightest area with the pipet tool. In addition, the luminance
of healthy enamel was measured at two healthy sites of the tooth
surface. The mean of these two values was used as reference for ‘nor-
mal’ tooth colour.
Sample size calculation
Sample size calculation was carried out by means of a two-sided
Wilcoxon–Mann–Whitney U-test at a significance level of α = 0.05.
In the group treated with the placebo gel, we assumed a distribu-
tion of DWL% as mentioned by Willmot et al. (46). In this in vitro
trial, the investigator observed a mean DWL% of 8.1 for 21 teeth
with WSLs. For ΔDWL% from baseline to 12 weeks (no fluoride
application), they observed a mean of 3.6 with an SD of 2.0 cor-
responding to a reduction of 44.4 per cent. For our study, due to
high-dose fluoride application we assumed a value of 5.4 for the
test group, which corresponds to a reduction of 66.7 per cent and
was regarded as the minimal clinically relevant effect on which the
sample size calculation was based.
A sample size of 21 patients per group yields 80 per cent power
to detect the clinically relevant effect for a two-sided unpaired t-test.
To account both for the relative asymptotic efficiency between the
two-sided unpaired t-test and Wilcoxon–Mann–Whitney U-test and
dropout, an additional 10 per cent of patients were randomized per
group, i.e. 46 patients in total.
Statistical analysis
The primary efficacy variable was the difference in DWL% from T0
to T4 (baseline to assessment at 12 weeks): ΔDWL% = DWL% (T4)
− DWL% (T0). This was calculated for both study products, and
mean data were compared statistically.
T2 T3 T4 T5
2 (±3 days) 6 (±3 days) 12 (±3 days) 24 (±3 days)
Home treatment Final visit
X X X X
X X X X
Because of the small sample size, the statistical hypotheses were
assessed by the Wilcoxon–Mann–Whitney U-test. The primary effi-
cacy analysis (T0–T4) was performed for the intent-to-treat (ITT)
population (all participants who attended visits T0 and T4, irrespec-
tive of T1 to T3). The significance level was set to α = 0.05. There
was no adjustment for multiplicity of tests.
In addition, the reliability for the measurements of both study
parameters was assessed using an intrarater reliability coefficient
(47) based on the two replicated measurements (Figure 4).
Downloaded from by guest on October 17, 2016
Results
Patient flow and adverse events
A total of 48 patients were included in the study. Two participants
missed their appointment during the first week of the trial (T1)
and were replaced after dropping out. The ITT analysis set was
comprised of 39 participants (7 dropouts due to missed appoint-
ment T4); test group: 9 male and 12 female and placebo group:
8 male and 10 female. Several similar adverse events (most com-
mon: common cold with cough and fever, headache, rhinitis and
sore throat) occurred frequently in both groups during the trial,
but none of them was classified as possibly related to the study
products.
Baseline findings
The two groups were comparable with respect to age, gender, and
other baseline characteristics (Table 2). The sample consisted of 20
males (42 per cent) and 28 females (58 per cent) with a mean age
of 15.3  years. The general health status at T0 was predominantly
unremarkable; 2 of the 48 patients suffered either from rheumatism
or Marfan syndrome. Both diseases are not known to affect WSL or
any related parameter.
Primary outcome: WSL dimension (DWL%)
Many WSLs displayed severely diffuse margins that hampered the
accurate determination of lesion borders thus, reliably and validly
tracing the WSLs proved to be difficult or even impossible. These dif-
fuse margins were most probably due to the large size of the images
that were evaluated at (100 per cent = real pixel size). As the first and
second measurements of the relative WSL areas deviated by as much
as 70 per cent, the method of quantifying WSL changes in dimen-
sion (DWL%) using a computer-assisted analysis proved to be unre-
liable (after study completion). Thus, interpreting the results, the
N. C. Bock et al. 5

rather large method error for this variable needs to be kept in mind
(Supplementary Tables 1 and 2).
The baseline values of WSL dimension (T0) were similar in both
groups, showing no statistically significant difference. The same was
true for the values from T1 to T5. Looking at the changes seen after 12
weeks (T0–T4) and 24 weeks (T0–T5) of Tx, greater reductions in WSL
dimension were seen for the lateral incisors in the placebo group (11.1 to
21.8 per cent) compared with the test group (−4.6 to 10.8 per cent). The
central incisors, however, showed greater effects in the test group (47.7
to 48.6 per cent) compared with the placebo group (23.1 to 41.1 per
cent). Nevertheless, no statistically significant group difference was seen.
Secondary outcome: WSL luminance
The baseline (T0) luminance values were similar in both groups
showing no statistically significant group difference (Supplementary
Tables 3 and 4, Figures 2 and 3). For the lateral incisors, the mean
values were 12.1 ± 5.1 (test 12) and 10.5 ± 5.4 (test 22) and 10.6 ± 3.7
(placebo 12) and 12.8 ± 4.5 (placebo 22), respectively. The values for
the central incisors were 8.6 ± 3.9 (test 11) and 9.6 ± 3.9 (test 21) and
7.9 ± 3.3 (placebo 11) and 8.1 ± 3.4 (placebo 21), respectively.
For the final luminance values, no statistically significant group
difference existed neither for the endpoint T4 nor for T5. The lumi-
nance values ranged between 8.8 ± 3.5 and 9.6 ± 3.9 for the lateral
incisors and between 5.7 ± 3.0 and 6.2 ± 3.1 for the central incisors
in both groups, respectively.
Looking at the changes seen after 12 weeks (T0–T4) and 24
regime). Nevertheless, 39 participants qualified for the ITT popula-
tion, which corresponds to a dropout rate of 15 per cent. The ITT
population was defined as all participants who did not miss visit
T4 (primary efficacy analysis: T0–T4). The gender distribution was
28 females versus 20 males which is in concordance with the total
patient sample of the department. While females generally do exhibit
a greater interest in oral health (48) and orthodontic Tx (49), it
remains inconclusive whether girls have a higher risk for WSL devel-
opment during MBA Tx or boys (4, 16, 50). The socio-economic
status was not considered in this trial; due to the randomization,
however, it should be similar in both groups.
The WSLs showed a degree 1 or 2 at the start of Tx. These inclu-
sion criteria were chosen to ensure the existence of a continuous
enamel surface that offers the possibility for remineralization (51) in
contrast to lesions exhibiting cavitation, which are non-reversible and
demand invasive Tx (52, 53). In addition, in concordance with previ-
ous studies (30, 34), intraoral photographs from before orthodontic
Tx were examined to ensure no WSL had been present pre-Tx.
The participants were instructed to use only the oral hygiene prod-
ucts handed out by the study centre, as additional fluoride application
might have influenced the study results. However, it was impossible to
ensure that this directive was followed. While a respective question-
naire was filled out at each visit, a certain bias cannot be ruled out.
The same is true for the home application of the study product. While
all participants received oral and written instructions on usage and
were asked to bring the product to each visit for weight assessment,

Downloaded from by guest on October 17, 2016

weeks (T0–T5) of Tx, slightly greater reductions in WSL dimension
were seen in the test group (21.9 and 37.4 per cent) compared with
the placebo group (20.0 and 29.4 per cent) Nevertheless, no statisti-
cally significant group difference was seen.
Discussion
The 48 included subjects were patients of the Department of
Orthodontics at the University of XXX with a mean age of
15.3 years at the time of inclusion (after at least 12 months of MBA
Tx) and—due to strict inclusion/exclusion criteria—similar dental
characteristics, making them a representative sample.
Unfortunately, the number of dropouts was distinctly higher than
expected (mainly due to minor sicknesses and the rather tight study
no full control was possible. So, alike in previous studies, compliance
must be considered as a possible confounding factor (54, 55).
Several methods for the detection, quantification, and monitor-
ing of WSLs are described in literature. While histologic analysis and
transverse microradiography are the gold standards, these variables
are not suitable for clinical WSL assessment.
Several in vitro and in vivo studies (37, 56–60) employed quan-
titative light-induced fluorescence (QLF). Although the results of
recent QLF studies are promising (61–63), the validity of clinical QLF
use remains questionable. Inconsistent handling by the investigator,
remaining surface moisture, and the presence of plaque or tartar are
considered possible sources of error (56, 57). In addition, WSLs after
removal of orthodontic treatment, with their typical location and
shape, were not assessed in most studies. Because these WSLs tend to

Table 2.  Baseline data of the 48 randomized patients. Sex and ethnic origin for the test and placebo groups are given as well as age, WSL
dimension (DWL%) and WSL luminance (difference between the brightest spot and healthy enamel) for the teeth 12–22, which includes the
mean value (Mean) and standard deviation (SD) for these variables. WSL = white spot lesions.

Test group (high-dose fluoride) Placebo group Group difference

Sex 15 ♀, 9 ♂ 13 ♀, 11 ♂

Ethnic origin 23 Caucasian, 1 Black 24 Caucasian, 0 Black

Mean SD Mean SD P

Age (years) 15.0 2.3 15.5 1.6 0.476

WSL size (DWL%) Tooth 12 25.5 12.1 31.2 14.3 0.321
Tooth 11 12.4 11.9 13.4 12.8 0.856
Tooth 21 12.1 11.3 14.6 15.2 0.749
Tooth 22 29.5 13.1 36.3 15.3 0.174
WSL luminance* Tooth 12 12.1 5.1 10.6 3.7 0.335
Tooth 11 8.6 3.9 7.9 3.3 0.594
Tooth 21 9.6 3.9 8.1 3.4 0.345
Tooth 22 10.5 5.4 12.8 4.5 0.209

*Difference between the brightest spot and healthy enamel.

6 European Journal of Orthodontics, 2016

Downloaded from by guest on October 17, 2016

Figure 2.  WSL luminance (difference between the brightest spot and healthy enamel). Boxplot diagrams for the teeth 12–22 in the test and placebo groups at
T0, T4, and T5. WSL = white spot lesions.

be large, constant illumination with optimal intensity is difficult (56,
57, 61). Nevertheless, the application of QLF changes as additional
parameter might have revealed further worthwhile findings.
Computer-based analysis of WSLs on photographs has been
employed for in vivo quantification frequently (46, 56, 57, 64–66).
However, it is difficult to standardize the photographs since light
conditions, distance between the lens and tooth, and the angle of
the camera–tooth surface have to be kept constant (34, 46, 56, 57).
Therefore, a head/neck holder, constant distance between cam-
era and object and constant light conditions in a windowless room
were used in the present investigation. However, minimal rotations
of the head as well as artefacts due to lightning could not be com-
pletely eliminated. No polarized filter—which might have reduced the
reflections—was used.
N. C. Bock et al. 7

Figure 3.  Changes in WSL luminance (difference between the brightest spot and healthy enamel). Boxplot diagrams for the teeth 12–22 in the test and placebo
groups during 12 weeks (T0–T4) and 24 weeks (T0–T5) of observation. WSL = white spot lesions.

Downloaded from by guest on October 17, 2016

The evaluation of intraoral photographs has been reported to be
at least as valid as a clinical examination in case of constant condi-
tions (16, 46, 64, 67, 68) and to have a high reproducibility (16, 69).
This, however, is not in concordance with the results of the present
study where a difference of up to 70per cent was seen between the
first and second surface measurements. So, according to the present
investigation, the method of surface measurement on digital high-reso-
lution photographs was non-reliable (Figure 4). A reason might be the
increased resolution of current digital photographs compared to ana-
logue photographs resulting in a very detailed image where even the
slightest enamel anomaly is visible. In addition, magnification might
play a role as the photographs were evaluated on a computer screen
and assessed at 100 per cent pixel size in the present investigation.
As an alternative to measuring the size of WSLs, a WSL severity
evaluation according to the CIELab system, measuring the brightness
in terms of luminance was introduced (prior data unblinding) because Figure 4.  Boxplot diagrams showing the intrarater reliability coefficient (47)
of both study parameters’ measurement.
a WSL size does not necessarily correlate with its severity (42). As WSL
visibility depends not only on the WSL itself but also on the surrounding
tooth colour (43), the luminance of the WSL relative to the luminance reduction as a sign for the improvement of the WSLs in both groups
of the surrounding enamel was measured in the present investigation. without a statistically significant group difference.
Furthermore, as the tooth colour varies within a tooth surface (41, 70), An in vitro investigation by Ballard et  al. (72) evaluating the
the luminance was measured at two different tooth sites (gingivally and changes of artificial WSL on extracted premolars showed distinct
incisally). This method proved to be more reliable than the one for the luminance reductions for all test groups compared with a control
primary study parameter WSL dimension (Figure 4). group. In the test groups, different products (among others one with
While the literature offers some studies where WSL were evalu- a similar fluoride contents as in the present study—1.1 per cent) were
ated on digital photographs (16, 20, 25, 30, 71), the study designs used over a 28-day period. These products were applied twice daily
varied considerably in terms of materials and methods, which does which could explain the difference in effect size. Unfortunately, this
not allow for a direct comparison of the results. Willmot (34) ana- protocol cannot be adapted clinically due to potential toxic effects
lysed the application of a low-dose fluoride mouth rinse (50 ppm) (52). In any case, as the luminance changes were only assessed for
compared with a placebo. The WSL changes were assessed using the WSL and not for the surrounding enamel, no direct comparison
scanned intraoral analogue photographs (which probably were of to the present data can be performed.
much lower quality than those used in the present study). However, Knösel et  al. (73) also assessed the luminance changes of teeth
no significant group difference was seen regarding WSL size reduc- with WSLs after MBA Tx. While the performed Tx (resin infil-
tion. This is in line with the present results showing a clear luminance tration) is not comparable to the present study, the values of the
8 European Journal of Orthodontics, 2016
untreated control group are. The respective mean WSL luminance
value at baseline was 73.12 ± 3.81, the value of the surrounding
enamel was 72.46 ± 2.64; therefore, as only a very small difference in
luminance existed, the WSLs must have been very mild. This might
also explain why almost no luminance reduction was observed after
an observation period of 6 months (WSL 72.78 ± 4.16 and enamel
72.42 ± 2.51).
Thinking about WSL development after MBA Tx in general,
the balance between protective and pathologic factors will influ-
ence the respective direction of changes—regression or progression
(74). In both groups of the present investigation, improvement of
the WSLs was seen except for 12 of the 66 affected teeth in the test
and 12 of the 54 affected teeth in the placebo group. It was not
surprising to find WSL reduction in the placebo group as well, as
this effect has been reported before (37, 69, 75, 76). After removal
of the MBA appliance, WSL development generally stops and initial
lesions can become inactive in the absence of cariogenic factors/good
oral hygiene (17, 75). Even saliva itself is able to remineralize initial
lesions (77, 78). In a long-term observation, Shungin et al. (69) found
that the majority of natural WSL changes take place during the first
and second year after MBA removal. In contrast, Mattousch et al.
(37) observed natural WSL improvement during the first 6 months
after MBA removal only.
While the use of low-dose (not greater than 0.025 per cent) fluo-
ride products (30, 34) and the use of a high-dose product (5.0 per
cent) for 8 weeks (79) did not evoke an effect compared with the
untreated controls, a positive effect was seen when applying a prod-
uct with a high fluoride concentration (5.0 per cent) for 6 months
(32). Nevertheless the use of high-dose fluoride products for WSL
Tx is still controversial. Some studies reported an effect of hyper-
mineralization but without achieving an aesthetic improvement (21,
80). Nevertheless, fluoridated enamel and also lesions will be more
resistant to acid attacks and therefore prevent a progression in terms
of cavity development (81, 82).
Limitations
Several limitations have to be considered when evaluating the results
of the present randomized controlled trial. First of all, the number of
dropouts was much higher than expected. However, due to randomi-
zation both the test and the placebo groups were affected. Another
limitation regarding the study population is the socio-economic sta-
tus, which was not assessed. But again, due to randomization, the
respective structure should be similar in both groups.
Another limitation certainly is the compliance. Due to practicability
reasons, it was impossible to assess whether the participants followed
the instructions on oral hygiene and fluoride/study product application
during the second (home use) phase. While efforts were made to control
this uncertainty (questionnaire and assessment of tube weight upon each
visit), compliance must be considered as a possible confounding factor.
Generalization
The results of the present study can be generalized for patient groups
with a similar mean age (approximately 15 years), inclusion/exclu-
sion criteria [especially with respect to the initial WSL with a modi-
fied score 1 or 2 (4)] and treatment protocol (especially in terms of
the study products’ fluoride concentration).
However, as the present investigation did not result in conclusive
evidence regarding the Tx of WSL which developed during ortho-
dontic MBA Tx, several open questions remain and the data clearly
indicate a further need for improvement and validation of the meth-
ods for monitoring WSLs in clinical trials.
Conclusion
In the present study, no significant positive effect of high-dose fluoride
(1.25 per cent) Tx on post-orthodontic WSL development (based on
a photographical assessment of WSL luminance) could be detected
compared with the placebo Tx. However, as the trial statistic was
designed to show superiority of high-dose fluoride Tx, non-significant
intergroup findings do not prove that treatment groups are equivalent.
Supplementary material
Supplementary material is available at European Journal of
Orthodontics online.
Funding
The study was funded by Colgate-Palmolive Europe; this comprised
all fees for authorities, insurances, independent monitoring, statisti-
cal planning, and evaluation as well as the supply with study prod-
ucts, participants’ compensation, and financial compensation for the
use of the study centre’s facilities/personnel paid to the university.
Conflict of interest
The authors declare that there is no conflict of interest regarding this research.
This includes no personal company or financial conflict.
Downloaded from by guest on October 17, 2016
References
1. Årtun, J. and Brobakken, B.O. (1986) Prevalence of carious white spots
after orthodontic treatment with multibonded appliances. European Jour-
nal of Orthodontics, 8, 229–223.
2. Fornell, A.C., Sköld-Larsson, K., Hallgren, A., Bergstrand, F. and Twetman
S. (2002) Effect of a hydrophobic tooth coating on gingival health, mutans
streptococci, and enamel demineralization in adolescents with fixed ortho-
dontic appliances. Acta Odontologica Scandinavica, 60, 37–41.
3. Geiger, A.M., Gorelick, L., Gwinnett, A.J. and Griswold, P.G. (1988) The
effect of a fluoride program on white spot formation during orthodontic
treatment. American Journal of Orthodontics and Dentofacial Orthope-
dics, 93, 29–37.
4. Gorelick, L., Geiger, A.M. and Gwinnett A.J. (1982) Incidence of white
spot formation after bonding and banding. American Journal of Ortho-
dontics, 81, 93–98.
5. Lovrov, S., Hertrich, K. and Hirschfelder, U. (2007) Enamel demineralization
during fixed orthodontic treatment - incidence and correlation to various
oral-hygiene parameters. Journal of Orofacial Orthopedics, 68, 353–363.
6. Mitchell, L. (1992) Decalcification during orthodontic treatment with fixed
appliances—an overview. British Journal of Orthodontics, 19, 199–205.
7. Mizrahi, E. (1982) Enamel demineralization following orthodontic treat-
ment. American Journal of Orthodontics, 82, 62–67.
8. Pancherz, H. and Mühlich, D.P. (1997) Entwicklung von Karies bei kiefer-
orthopädischer Behandlung mit festsitzenden Apparaturen—Ein Vergleich
von Zähnen mit und ohne Kariesvorschädigungen. Kieferorthopädie, 11,
139–144.
9. Wenderoth, C.J., Weinstein, M. and Borislow, A.J. (1999) Effectiveness of
a fluoride-releasing sealant in reducing decalcification during orthodontic
treatment. American Journal of Orthodontics and Dentofacial Orthope-
dics, 116, 629–634.
10. Tufekci, E., Dixon, J.S., Gunsolley, J.C. and Lindauer, S.J. (2011) Preva-
lence of white spot lesions during orthodontic treatment with fixed appli-
ances. The Angle Orthodontist, 81, 206–210.
11. Øgaard, B. (1989) Prevalence of white spot lesions in 19-year-olds: a study on
untreated and orthodontically treated persons 5 years after treatment. Ameri-
can Journal of Orthodontics and Dentofacial Orthopedics, 96, 423–427.
12. Hadler-Olsen, S., Sandvik, K., El-Agroudi, M.A. and Ogaard, B. (2012)
The incidence of caries and white spot lesions in orthodontically treated
N. C. Bock et al.
adolescents with a comprehensive caries prophylactic regimen—a pro-
spective study. European Journal of Orthodontics, 34, 633–639.
13. Øgaard, B., Rølla, G. and Arends, J. (1988) Orthodontic appliances and
enamel demineralization. Part 1. Lesion development. American Journal of
Orthodontics and Dentofacial Orthopedics, 94, 68–73.
14. O’Reilly, M.M. and Featherstone, J.D. (1987) Demineralization and rem-
ineralisation around orthodontic appliances: an in vivo study. American
Journal of Orthodontics and Dentofacial Orthopedics, 92, 33–40.
15. Enaia, M., Bock, N. and Ruf, S. (2011) White spot lesions during multi-
bracket appliance treatment – A challenge for clinical excellence. Ameri-
can Journal of Orthodontics and Dentofacial Orthopedics, 140, e17–e24.
16. Chapman, J.A., Roberts, W.E., Eckert, G.J., Kula, K.S. and González-
Cabezas, C. (2010) Risk factors for incidence and severity of white spot
lesions during treatment with fixed orthodontic appliances. American
Journal of Orthodontics and Dentofacial Orthopedics, 138, 188–194.
17. Årtun, J. and Thylstrup, A. (1986) Clinical scanning electron microscopic
study of surface changes of incipient caries lesions after debonding. Scan-
dinavian Journal of Dental Research, 94, 193–201.
18. Backer Dirks, O. (1966) The clinical testing of agents for the prevention of
dental caries. Advancd Fluorine Research, 4, 1–2.
19. von der Fehr, F.R., Löe, H. and Theilade, E. (1970) Experimental caries in
man. Caries Research, 4, 131–148.
20. Julien, K.C., Buschang, P.H. and Campbell, P.M. (2013) Prevalence of
white spot lesion formation during orthodontic treatment. The Angle
Orthodontist, 83, 641–647.
21. Bishara, S.E. and Ostby, A.W. (2008) White spot lesions: formation, pre-
vention, and treatment. Seminars in Orthodontics, 14, 174–182.
22. O’Reilly, M.T., De Jesús Viñas, J. and Hatch, J.P. (2013) Effectiveness of a
sealant compared with no sealant in preventing enamel demineralization in
patients with fixed orthodontic appliances: a prospective clinical trial. Ameri-
can Journal of Orthodontics and Dentofacial Orthopedics, 143, 837–844.
23. Knösel, M., Forslund, L., Jung, K. and Ziebolz, D. (2012) Efficacy of differ-
ent strategies in protecting enamel against demineralization during fixed
orthodontic treatment. Journal of Orofacial Orthopedics, 73, 194–203.
24. Splieth, C.H., Treuner, A., Gedrange, T. and Berndt, C. (2012) Caries-pre-
ventive and remineralizing effect of fluoride gel in orthodontic patients
after 2 years. Clinical Oral Investigations, 16, 1395–1399.
25. Stecksén-Blicks, C., Renfors, G., Oscarson, N.D., Bergstrand, F. and Twet-
man, S. (2007) Caries-preventive effectiveness of a fluoride varnish: a ran-
domized controlled trial in adolescents with fixed orthodontic appliances.
Caries Research, 41: 455–459
26. Sudjalim, T.R., Woods, M.G. and Manton, D.J. (2006) Prevention of white
spot lesions in orthodontic practice: a contemporary review. Australian
Dental Journal, 51, 284–289.
27. Vivaldi-Rodrigues, G., Demito, C.F., Bowman, S.J. and Ramos, A.L.
(2006) The effectiveness of a fluoride varnish in preventing the develop-
ment of white spot lesions. World Journal of Orthodontics, 7, 138–144.
28. Bansal, G.S., Newman, H.N. and Wilson, M. (1990) The survival of sub-
gingival plaque bacteria in an amine fluoride-containing gel. Journal of
Clinical Periodontology, 17, 414–418.
29. Van Loveren, C. (1990) The antimicrobial action of fluoride and its role in
caries inhibition. Journal of Dental Research, 69, 676–681.
30. Akin, M. and Basciftci, F. (2012) Can white spot lesions be treated effec-
tively? The Angle Orthodontist, 82, 770–775.
31. Oosterwaal, P.J.M., Mikx, F.H.M., van den Brink, M.E. and Renggli, H.H.
(1989) Bactericidal concentrations of chlorhexidine-digluconate, amine flu-
oride gel and stannous fluoride gel for subgingival bacteria tested in serum
at short contact times. Journal of Periodontal Research, 24:155–160.
32. Du, M., Cheng, N., Tai, B., Jiang, H., Li, J. and Bian, Z. (2012) Rand-
omized controlled trial on fluoride varnish application for treatment of
white spot lesion after fixed orthodontic treatment. Clinical Oral Investi-
gations, 16, 463–468.
33. Lagerweij, M.D. and ten Cate, J.M. (2002) Remineralisation of enamel
lesions with daily applications of a high-concentration fluoride gel and a
fluoridated toothpaste: an in situ study. Caries Research, 36: 270–274.
34. Willmot, D.R. (2004) White lesions after orthodontic treatment: does low
fluoride make a difference? Journal of Orthodontics, 31, 235–242.
9
35. Hellwig, E. and Lussi, A. (2001) What is the optimum fluoride concentra-
tion needed for the remineralisation process? Caries Research, 35, Suppl
1, 57–59.
36. Chen, H., Liu, X., Dai, J., Jiang, Z., Guo, T. and Ding Y. (2013) Effect of
remineralizing agents on white spot lesions after orthodontic treatment:
a systematic review. American Journal of Orthodontics and Dentofacial
Orthopedics, 143, 376–382.
37. Mattousch, T.J., van der Veen, M.H. and Zentner, A. (2007) Caries lesions
after orthodontic treatment followed by quantitative light-induced fluores-
cence: a 2-year follow-up. European Journal of Orthodontics, 29, 294–298.
38. Ferreira, J.M., Silva, M.F., Oliveira, A.F. and Sampaio, F.C. (2008) Evalu-
ation of different methods for monitoring incipient carious lesions in
smooth surfaces under fluoride varnish therapy. International Journal of
Paediatric Dentistry, 18, 300–305.
39. Gozalo-Diaz, D.J., Lindsey, D., Johnston, W.M. and Wee, A.G. (2007)
Measurement of color for craniofacial structures using a 45/0-degree opti-
cal configuration. Journal of Prosthetic Dentistry, 97, 45–53.
40. Joiner, A. (2004) Tooth colour: a review of the literature. Journal of Den-
tistry, 32, Suppl 1, 3–12.
41. O’Brien, W.J., Hemmendinger, H., Boenke, K.M., Linger, J.B. and Groh,
C.L. (1997) Color distribution of three regions of extracted human teeth.
Dental Materials, 13, 179–185.
42. Torlakovic, L., Olsen, I., Petzold, C., Tiainen, H. and Øgaard, B. (2012)
Clinical color intensity of white spot lesions might be a better predictor of
enamel demineralization depth than traditional clinical grading. American
Journal of Orthodontics and Dentofacial Orthopedics, 142, 191–198.
43. Watts, A. and Addy, M. (2001) Tooth discolouration and staining: a review
of the literature. British Dental Journal, 190, 309–316.
44. Ishikawa-Nagai, S., Ishibashi, K., Tsuruta, O. and Weber, H.P. (2005)
Downloaded from by guest on October 17, 2016
Reproducibility of tooth color gradation using a computer color-matching
technique applied to ceramic restorations. Journal of Prosthetic Dentistry,
93, 129–137.
45. Joiner, A., Hopkinson, I., Deng, Y. and Westland S. (2008) A review of
tooth colour and whiteness. Journal of Dentistry, 36, Suppl 1, 2–7.
46. Willmot, D.R., Benson, P.E., Pender, N. and Brook, A.H. (2000) Reproduc-
ibility of quantitative measurement of white enamel demineralisation by
image analysis. Caries Research, 34: 175–181.
47. Gwet, K.L. (2008) Intrarater Reliability. Wiley Encyclopedia of Clinical
Trials, New York, NY, pp. 1–13.
48. Östberg, A.L., Halling, A. and Lindblad, U. (1999) Gender differences in
knowledge, attitude, behaviour and perceived oral health among adoles-
cents. Acta Odontologica Scandinavica, 57, 231–236.
49. Wheeler, T.T., McGorray, S.P., Yurkiewicz, L., Keeling, S.D. and King,
G.J. (1994) Orthodontic treatment demand and need in third and fourth
grade schoolchildren. American Journal of Orthodontics and Dentofacial
Orthopedics, 106, 22–33.
50. Khalaf, K. (2014) Factors affecting the formation, severity and location
of white spot lesions during orthodontic treatment with fixed appliances.
Journal of Oral and Maxillofacial Research, 5, e4.
51. Chang, H.S., Walsh, L.J. and Freer, T.J. (1997) Enamel demineralization
during orthodontic treatment. Aetiology and prevention. Australian Den-
tal Journal, 42, 322–327.
52. Hellwig, E., Klimek, J. and Attin, T. (2007) Einführung in die Zahnerhal-
tung. Urban & Fischer, Munich, Germany.
53. Stahl, J. and Zandona, A.F. (2007) Rationale and protocol for the treatment of
non-cavitated smooth surface carious lesions. General Dentistry, 55, 105–111.
54. Beerens, M.W., van der Veen, M.H., van Beek, H. and ten cate, J.M. (2010)
Effects of casein phosphopeptide amorphous calcium fluoride phosphate
paste on white spot lesions and dental plaque after orthodontic treatment:
a 3-month follow-up. European Journal of Oral Sciences, 118, 610–617.
55. Geiger, A.M., Gorelick, L., Gwinnett, A.J. and Benson, B.J. (1992) Reducing
white spot lesions in orthodontic populations with fluoride rinsing. Ameri-
can Journal of Orthodontics and Dentofacial Orthopedics, 101, 403–407.
56. Benson, P.E., Pender, N. and Higham, S.M. (2003) Quantifying enamel
demineralization from teeth with orthodontic brackets—a comparison of
two methods. Part 1: repeatability and agreement. European Journal of
Orthodontics, 25, 149–158.
10
57. Benson, P.E., Pender, N. and Higham, S.M. (2003) Quantifying enamel
demineralization from teeth with orthodontic brackets—a comparison
of two methods. Part 2: validity. European Journal of Orthodontics, 25,
159–165.
58. Boersma, J.G., van der Veen, M.H., Lagerweij, M.D., Bokhout, B. and
Prahl-Andersen, B. (2005) Caries prevalence measured with QLF after
treatment with fixed orthodontic appliances: influencing factors. Caries
Research, 39: 41–47.
59. Pretty, I.A., Pender, N., Edgar, W.M. and Higham, S.M. (2003) The in vitro
detection of early enamel de- and re-mineralization adjacent to bonded
orthodontic cleats using quantitative light-induced fluorescence. European
Journal of Orthodontics, 25, 217–223.
60. Zantner, C., Martus, P. and Kielbassa, A.M. (2006) Clinical monitoring of
the effect of fluorides on long-existing white spot lesions. Acta Odonto-
logica Scandinavica, 64, 115–122.
61. Aljehani, A., Tranaeus, S., Forsberg, C.-M., Angmar-Månsson, B. and
Shi, X.-Q. (2004) In vitro quantification of white spot enamel lesions
adjacent to fixed orthodontic appliances using quantitative light-induced
fluorescence and DIAGNOdent. Acta Odontologica Scandinavica, 62,
313–318.
62. Al-Khateeb, S., ten Cate, J.M., Angmar-Månsson, B., de Josselin de Jong,
E., Sundstrom, G., Exterkate, R.A. and Oliveby A. (1997) Quantification
of formation and remineralisation of artificial enamel lesions with a new
portable fluorescence device. Advances in Dental Research, 11, 502–506.
63. Ando, M., van Der Veen, M.H., Schemehorn, B.R. and Stookey, G.K.
(2001) Comparative study to quantify demineralized enamel in decidu-
ous and permanent teeth using laser- and light-induced fluorescence tech-
niques. Caries Research, 35, 464–470.
64. Benson, P.E., Pender, N. and Higham, S.M. (2000) Enamel demineralisa-
tion by computerised image analysis of clinical photographs. Journal of
Dentistry, 28, 319–326.
65. Kanthathas, K., Willmot, D.R. and Benson, P.E. (2005) Differentiation of
developmental and post-orthodontic white lesions using image analysis.
European Journal of Orthodontics, 27, 167–172.
66. Livas, C., Kuijpers-Jagtman, A.M., Bronkhorst, E., Derks, A. and Katsaros,
C. (2008) Quantification of white spot lesions around orthodontic brack-
ets with image analysis. The Angle Orthodontist, 78, 585–590.
67. Benson, P.E., Pender, N., Higham, S.M. and Edgar, W.M. (1998) Morpho-
metric assessment of enamel demineralization from photographs. Journal
of Dentistry, 26, 669–677.
68. Cochran, J.A., Ketley, C.E., Saches, L., Mamai-Homata, E., Oila, A.M.,
Arnadottir, I.B., van Loveren, C., Whelton, H.P. and O’Mullane, D.M.
(2004) A standardized photographic method for evaluating enamel opaci-
ties including fluorosis. Community Dentistry and Oral Epidemiology, 32,
Suppl 1, 19–27.
69. Shungin, D., Olsson, A.I. and Persson, M. (2010) Orthodontic treatment-
related white spot lesions: a 14-year prospective quantitative follow-up,
European Journal of Orthodontics, 2016
including bonding material assessment. American Journal of Orthodontics
and Dentofacial Orthopedics, 138, 136.e1––136.e8.
70. Dozic, A., Kleverlaan, C.J., Aartman, I.H.A. and Feilzer, A.J. (2004) Rela-
tion in color of three regions of vital human incisors. Dental Materials, 20,
832–838.
71. Kim, S., Kim, E.Y., Jeoun, T.S. and Kim, J.W. (2011) The evaluation of
resin infiltration for masking labial enamel white spot lesions. Interna-
tional Journal of Paediatric Dentistry, 21, 241–248.
72. Ballard, R.W., Hagan, J.L., Phaup, A.N., Sarkar, N., Townsend, J.A. and
Armbruster, P.C. (2013) Evaluation of 3 commercially available materials
for resolution of white spot lesions. American Journal of Orthodontics
and Dentofacial Orthopedics, 143, S78–S84.
73. Knösel, M., Eckstein, A. and Helms, H.J. (2013) Durability of esthetic
improvement following Icon resin infiltration of multibracket-induced
white spot lesions compared with no therapy over 6  months: a single-
center, split-mouth, randomized clinical trial. American Journal of Ortho-
dontics and Dentofacial Orthopedics, 144, 86–96.
74. Featherstone, J.D. (2000) The science and practice of caries prevention.
Journal of American Dental Association, 131, 887–899.
75. Al-Khateeb, S., Forsberg, C.M., de Josselin de Jong, E. and Angmar-Mans-
son, B. (1998) A longitudinal laser fluorescence study of white spot lesions
in orthodontic patients. American Journal of Orthodontics and Dentofa-
cial Orthopedics, 113, 595–602.
76. van der Veen, M.H., Mattousch, T. and Boersma, J.G. (2007) Longitudinal
development of caries lesions after orthodontic treatment evaluated by
quantitative light-induced fluorescence. American Journal of Orthodontics
and Dentofacial Orthopedics, 131, 223–228.
77. Øgaard, B., Rølla, G., Arends, J. and ten Cate, J.M. (1988) Orthodontic
appliances and enamel demineralization. Part 2. Prevention and treatment
Downloaded from by guest on October 17, 2016
of lesions. American Journal of Orthodontics and Dentofacial Orthope-
dics, 94, 123–128.
78. Øgaard, B. and ten Bosch J.J. (1994) Regression of white spot enamel
lesions. A new optical method for quantitative longitudinal evaluation in
vivo. American Journal of Orthodontics and Dentofacial Orthopedics,
106, 238–242.
79. Huang, G.J., Roloff-Chiang, B., Mills, B.E., Shalchi, S., Spiekerman, C., Kor-
pak, A.M., Starrett, J.L., Greenlee, G.M., Drangsholt, R.J. and Matunas,
J.C. (2013) Effectiveness of MI Paste Plus and PreviDent fluoride varnish
for treatment of white spot lesions: a randomized controlled trial. American
Journal of Orthodontics and Dentofacial Orthopedics, 143, 31–41.
80. Guzman-Armstrong, S., Chalmers, J. and Warren, J.J. (2010) Ask us. White
spot lesions: prevention and treatment. American Journal of Orthodontics
and Dentofacial Orthopedics, 138, 690–696.
81. Featherstone, J.D. (1999) Prevention and reversal of dental caries: role of low
level fluoride. Community Dentistry and Oral Epidemiology, 27, 31–40.
82. Koulourides, T. and Cameron, B. (1980) Enamel remineralization as a factor
in the pathogenesis of dental caries. Journal of Oral Pathology, 9, 255–269.