European Journal of Orthodontics 35

1 of(2013)
8 361–368 © The Author 2012. Published by Oxford University Press on behalf of the European Orthodontic Society.
doi:10.1093/ejo/cjr138 All rights reserved. For permissions, please email: journals.permissions@oup.com
Advance Access publication 12 January 2012

Screening for salivary levels of deoxypyridinoline and bone-

specific alkaline phosphatase during orthodontic tooth
movement: a pilot study
Gloria A. Flórez-Moreno, Lina M. Marín-Restrepo, Diana M. Isaza-Guzmán and
Sergio I. Tobón-Arroyave
POPCAD Research Group, Laboratory of Immunodetection and Bioanalysis, Faculty of Dentistry, University of
Antioquia, Medellín, Colombia

Downloaded from https://academic.oup.com/ejo/article/35/3/361/450013 by guest on 28 February 2023

Correspondence to: Sergio Iván Tobón-Arroyave, POPCAD Research Group, Laboratory of Immunodetection and
Bioanalysis, Faculty of Dentistry, University of Antioquia, Calle 64 N° 52-59, Medellín, Colombia. E-mail: gflorez@
une.net.co

SUMMARY Deoxypyridinoline (DPD) and bone-specific alkaline phosphatase (BAP) have been regarded
as systemic determinants of bone remodelling. Owing this fact, this study aimed to determine whether
the variations in the salivary concentration of these two biomarkers as detected through a longitudinal
follow-up with four consecutive visits may be linked with the different phases of orthodontic tooth
movement (OTM). Twenty-two healthy subjects who required fixed appliance therapy not involving tooth
extractions/surgical procedures were selected. Unstimulated whole saliva samples were collected from
each patient prior to fitting the orthodontic appliances and 24–48 hours, 2 weeks, and 5 weeks after the
activation. Salivary DPD and BAP concentrations were determined by enzyme-linked immunosorbent
assay. The data were analysed using non-parametric statistics. There were no statistically significant
differences in salivary levels of biomarkers regarding demographic and clinical parameters. Overall,
although DPD values revealed an increasing nature after force application and BAP values showed a
descending trend, only the former showed statistically significant changes over time. Furthermore, post
hoc comparisons for DPD salivary levels revealed significant differences between every paired sampling
times, except for the pair baseline test/24–48 hours test. Synchronously, a moderate positive significant
correlation between both salivary biomarkers was observed at 2 weeks test. The findings indicate that
although salivary levels of DPD and BAP may act as indicators of increased bone remodelling, it appears
that DPD dominates the earlier phases of OTM, whereas BAP might serve as indicator of bone formation
as soon as the tooth movement stops.

Introduction
Orthodontic tooth movement (OTM) constitutes a highly
complex process dened as an adaptive biological response
to interference in the physiological equilibrium in the
dentofacial structures by an externally applied force (Proft,
2000). The host response to orthodontic force has been
described as an aseptical and transitory inammation
(Garlet et al., 2008) that mainly alters the vascularity and
blood ow of periodontal ligament (PDL), resulting in local
synthesis and release of different mediators involved in
alveolar bone remodelling (Krishnan and Davidovitch,
2006). These molecules can evoke many cellular responses
by various cell types in and around teeth, providing a
favourable microenvironment for bone resorption or
apposition (Davidovitch, 1991). Although these effects are
both physical and biochemical in nature and are frequently
intertwined and interdependent (Krishnan and Davidovitch,
2009), the evidence has shown large inter-individual
differences in both human research (Iwasaki et al., 2000,
2005) and animal experiments (van Leeuwen et al., 1999).
In other words, with standardized, constant, and equal
forces, the rate of OTM may vary substantially, while with
considerably different forces, the rates of OTM may be
almost the same among and even within individuals (Ren
et al., 2003).
It has been hypothesized that such inter-individual
variability is possibly related to several factors, among
them, genetic background (Iwasaki et al., 2006), variation
in the structure and cellular activity within the PDL and
alveolar bone (Ren et al., 2003), morphological and
biomechanical differences in teeth (von Böhl et al., 2004),
as well as differences in the expression of various bone
remodelling mediators (Iwasaki et al., 2001; Perinetti et
al., 2002; Ren et al., 2002; Ingman et al., 2005; Isik et al.,
2005; Iwasaki et al., 2005; Giannopoulou et al., 2008).
Notwithstanding, depending on force characteristics,
OTM follows a specic pattern in time which comprises
three phases (Burstone, 1962): an initial phase, which
2 of 8
362
lasts 24 hours to 2 days after force application and
represents a rapid displacement of the tooth in the PDL
space; a lag phase, which lasts 4–20 days, with relatively
low rates of tooth displacement or no displacement; and a
linear (postlag) phase during which the rate of OTM
gradually or suddenly increases. Each of these phases is
determined by biochemical, cellular, and tissue-specic
reactions involving the recruitment of osteoblast and
osteoclast precursors as well as, the extravasation and
chemotaxis of inammatory cells (Krishnan and
Davidovitch, 2006).
In order to improve the knowledge on this subject, it
has been proposed that multiple biomarkers, which acting
as systemic determinants of increased bone remodelling,
might be practical and reliable for determining when
a specific event predominates. In this sense, both
deoxypyridinoline (DPD), which appears to be the most
specic marker of systemic osteoclast activity and bone-
specic alkaline phosphatase (BAP), which is considered
as the key indicator of osteoblast function essential for
bone formation (Seibel, 2000; McGehee and Johnson,
2004), have shown promising results in the development
of strategies to understand the biology of OTM, specially
regarding to the processes involved in bone resorption
and apposition (Griffiths et al., 1998; Perinetti et al.,
2002; Isik et al., 2005; Asma et al., 2008). Therefore, if
it would be possible to monitor the predominant process
during the application of orthodontic forces based on
the assessment of the levels of both markers, the control
of orthodontic mechanotherapy could be based on
the individual tissular response, thus enhancing the
effectiveness of treatment.
Many of the human studies regarding the biology of
OTM have focused on the assessment of these and other
biomarkers in gingival crevicular uid (GCF; Iwasaki
et al., 2001; Perinetti et al., 2002; Ren et al., 2002; Lee et al.,
2004; Ingman et al., 2005; Isik et al., 2005; Iwasaki et al.,
2005; Giannopoulou et al., 2008); however, it is difcult to
draw rm conclusions because the number of studies
concerning variations in the levels of bone remodelling
biomarkers through the different phases of OTM is sparse
and has yielded contradictory results (Grifths et al., 1998;
Perinetti et al., 2002; Ingman et al., 2005; Isik et al., 2005;
Asma et al., 2008; Sprogar et al., 2010). Although the
clinical and radiographic follow-up examination remains
the basis for patient’s evaluation, analysis of saliva, a uid
that contains local and systemically derived markers,
may offer the basis for a phase-specic screening of
OTM. Since, currently, there are no reports on DPD or BAP
levels in saliva during orthodontic treatment with xed
appliances, this study aimed to determine whether the
variations in the concentration of these bone remodelling
biomarkers as detected through a longitudinal follow-up
with four consecutive visits may be linked with the different
phases of OTM.
G. A. FLÓREZ-MORENO ET AL.
Subjects and methods
Subject selection and inclusion/exclusion criteria
This descriptive prospective cohort study was conducted
at the Faculty of Dentistry, University of Antioquia in
Medellín, Colombia. The study conformed to the ethical
guidelines of the Helsinki Declaration and was evaluated
and approved by the Institutional Ethics Committee
for Human Studies (Technical Research Council-CIFO,
Code IORG 0002429). The study population comprised
all of the patients that sought treatment at the
Postgraduate Orthodontic Clinics of the University of
Antioquia from September 2009 to September 2011. All
Downloaded from https://academic.oup.com/ejo/article/35/3/361/450013 by guest on 28 February 2023
potential referred participants were examined at a
screening session by trained research associates (GAF-M
and LMM-R) for establishing their suitability for the
study.
Participants were privately interviewed to obtain medical
and demographic information and were given a clinical
screening for oral pathology and a periodontal examination.
The information that was gathered included the subject’s
gender, age, type of dentofacial pattern, and severity of
dental arch crowding. Eligibility criteria included subjects
having a minimum of 20 remaining teeth, need for xed
appliance therapy not involving tooth extractions/surgical
procedures, and good general and periodontal health
(probing depth values not exceeding 3 mm in the whole
dentition and no radiographic evidence of periodontal bone
loss). Furthermore, exclusion criteria included pregnancy
and lactation; patients who would not give informed
consent; previous history of habits, such as alcohol
consumption and/or tobacco use; ongoing orthodontic or
orthopaedic therapy; any systemic condition that could
affect the host’s periodontal status and bone metabolism
(e.g. osteoporosis, gastrointestinal diseases related to
nutrition and mineral metabolism, endocrine diseases,
immunological disorders, connective tissue diseases) or
that would require pre-medication for monitoring or
treatment procedures (e.g. heart conditions, joint
replacements, hormonal or bisphosphonate antiresorptive
therapies, and chronic therapy with heparin or
corticosteroids); use of antibiotics, corticosteroids, and/or
anti-inammatory drugs within the last 3 months; and
professional cleaning or periodontal treatment within the
last 6 months.
The patients were consecutively invited to participate in
the study if they did not meet any of the exclusion criteria.
Preceding the baseline sampling, all subjects received oral
hygiene instructions for the use of toothbrush, dental oss,
and interdental brush. All participants or the parents of
those under 18 years of age provided written informed
consent prior to their enrolment into the study. Recruitment
of the study was terminated at 22 patients because of slow
accrual. A considerably larger sample size was hoped for
but not achieved.
SALIVARYBIOMARKERS
SALIVARY BIOMARKERSAND
ANDORTHODONTIC
ORTHODONTICMOVEMENT
MOVEMENT 3 of
Saliva collection and clinical procedures
Once the diagnosis and treatment planning were established,
prior to tting the orthodontic appliances, baseline saliva
samples were collected from each patient. About 5 ml of
unstimulated whole saliva were collected into a 50 ml
sterile plastic centrifuge tube (Greiner Bio-one®,
Frickenhausen, Germany) before breakfast intake and any
dental hygiene procedure. No antiseptic mouth rinse was
used before collection. Immediately after collection, whole
saliva was clarified by centrifugation for 5 minutes at
800 × g (IEC® Centra CL2 Centrifuge; Thermo Electron
Corporation, Mildford, Massachussetts, USA). The
supernatants were collected and aliquoted into 500 µl
volumes and frozen at −75°C until processed.
All subjects were treated by using a Roth prescription
0.018 × 0.025 inch bracket slot appliance bonded to the
upper and/or lower teeth. A 0.014 inch nickel titanium
conventional archwire was then placed for initial levelling
and alignment stage. Thereafter, saliva samplings were
taken at 24–48 hours (initial phase), 2 weeks (lag phase),
and 5 weeks (post-lag phase) after force application. No
further appliance reactivations were performed throughout
the different sampling times.
Enzyme-linked immunosorbent assay for the quantitation of
DPD and BAP in whole saliva samples
Salivary levels of DPD and BAP were measured through
an enzyme-linked immunosorbent assay (ELISA)-based
capture assay by using the commercial kits MicroVue®
DPD and MicroVue® BAP (Quidel® Corporation, San
Diego, California, USA) following the manufacturer ’s
instructions. Claried saliva was used undiluted. For both
proteins, optical density was determined within 15 minutes
after 100 µl stop solution (0.5 N NaOH) was added, using
a microplate reader (ChroMate® 4300; Awareness
Technology, Palm City, Florida, USA) set to 405 nm. To
calculate the concentrations, a non-linear regression model
was performed using GraphPad Prism® version 5.04
(GraphPad Software, San Diego, California, USA). The
standard curves obtained were used to calculate the real
concentration of each protein in samples, standards, and
low/high internal controls. R2 values for the typical
standard curves were 0.9996 for DPD (4-parameter
calibration curve) and 0.9930 for BAP (quadratic
calibration curve). According to the manufacturer, the
assays used have a sensitivity of 1.1 nmol/l minimum
detectable dose for DPD and 0.7 U/l for BAP. In addition,
the monoclonal anti-DPD antibody has selective high
afnity for free DPD and negligible binding to DPD
peptides and free or peptide bound pyridinoline. Likewise,
the BAP antibody has selective high afnity for the BAP
isoform, low cross-reactivity to the liver form of alkaline
phosphatase, and negligible binding of intestinal and
placental isoenzymes.
3638
Statistical analysis
Data collected were analysed using Statistical Package for
the Social Sciences (SPSS) 15.0® (SPSS®, Chicago, Illinois,
USA). Reproducibility for ELISA assays was determined
through double evaluations for each specic test performed by
the same observer with ve samples selected randomly using
a computer-generated randomization code (Epidat 3.1®;
PAHO/WHO, Washington, DC, USA). The interval between
tests 1 and 2 was 7 days. For comparisons, the reliability
between the two series of data was assessed using the
intraclass correlation coefcient (ICC) test. A value greater
than 0.6 was considered high reproducibility, a value greater
Downloaded from https://academic.oup.com/ejo/article/35/3/361/450013 by guest on 28 February 2023
than 0.8 represented very high reproducibility, and a value
greater than 0.9 represented excellent reproducibility.
All parameters were tested for normal distribution using
the Shapiro–Wilk test for small sample size. Because the
data showed non-normal distributions, the variables were
analysed using non-parametric methods. Consequently,
analysis of categorical variables was based on Pearson’s
chi-square test. Likewise, for the analysis of immunoenzymatic
ndings against other demographic and clinical parameters,
the Kruskal–Wallis H- and Mann–Whitney U-tests were
used where appropriate. Also, Friedman test for repeated
measures was used in order to compare the immunoenzymatic
ndings over the sampling time periods. In signicant cases,
Wilcoxon signed-rank test for matched samples were
performed as post hoc tests. Finally, Spearman’s rank
correlation coefcient was used to describe the relationship
between the two salivary biomarkers at each sampling time.
All tests were two-sided and statistical signicance was
assumed at a P-value less than 0.05.
Results
Reproducibility of measurements and statistical power
calculation
Overall, reproducibility was excellent for both DPD and
BAP quantitative estimation (ICC = 0.995 and 0.999,
respectively; P < 0.001). The statistical power calculation
indicated that the sample size might allow the detection of
signicant differences in the salivary levels of these two
biomarkers across the sampling times with an α value of
0.05 per cent and 70–98 per cent power for BAP and DPD,
respectively.
Demographic and clinical prole of the study subjects
All the study subjects were compared to determine the
similarities between them and to assist in interpreting the
results. None of the participant subjects showed clinical
evidence of gingival/periodontal inammation prior to
tting the orthodontic appliances, and there were no clinical
signs of inammation after appliance activation. The study
included nine males ranging in age from 11 to 52 years
4 of 8
364
(median age 13 years) and 13 females ranging in age from
11 to 57 years (median age 17 years). A Class I relationship
was the most common dentofacial pattern (10 subjects).
This was followed by a Class II relationship (seven subjects)
and Class III pattern (ve subjects). According to gender
and dental arch crowding severity, 12 subjects were mild
(ve males and seven females) and 10 subjects were
moderate (four males and six females). Finally, 17 subjects
were treated with monomaxillary orthodontic appliances,
while ve subjects were treated with bimaxillary appliances.
There were no signicant differences between men and
women with respect to age (P = 0.126, Mann–Whitney
U-test), or the frequency of dentofacial pattern, dental arch
crowding, or the location of orthodontic appliance (all P >
0.05, chi-square test, data not shown).
Quantitative determination of DPD and BAP salivary levels
by ELISA
A total of 88 unstimulated whole saliva samples (four per
subject) were processed for each specic analyte. The
ELISA analysis revealed detectable salivary levels of DPD
in 48 samples (54.5 per cent), whereas BAP was above the
limit of detection (LOD) in 74 samples (84.1 per cent). It
was noticeable that for DPD, non-detected values occurred
among the same subjects in the baseline (16 samples), 24–
48 hours (14 samples), 2 weeks (7 samples), and 5 weeks (3
samples) tests. On the contrary, the results of BAP were
more consistent, with non-detected values only in six
samples at baseline, three samples at 24–48 hours, two
samples at 2 weeks, and three samples at 5 weeks tests. As
depicted in Table 1, the assessment of these bone remodelling
biomarkers at each time interval revealed no signicant
differences (P > 0.05, Kruskal–Wallis H- and Mann–
Whitney U-tests) regarding to demographic and clinical
parameters, thus allowing optimal comparability of the data
between subjects.
The results of the overall quantitative changes of salivary
concentrations of DPD and BAP at different sampling times
are presented in Table 2. It was noteworthy that while mean
ranks of DPD showed an increasing nature, BAP values
revealed a downward trend over time after activation visit.
However, the results of the Friedman test showed that
whereas salivary concentrations of BAP did not signicantly
differ over time (chi-square = 1.311; df = 3; P = 0.727),
there were statistically signicant quantitative changes in
the mean ranks of the DPD salivary levels through the
different sampling times (chi-square = 30.399; df = 3; P <
0.001). As also can be noted from Table 2, Spearman
correlation analysis showed a moderate positive signicant
correlation (rs = 0.499, P = 0.018) between salivary levels
of DPD and BAP at 2 weeks test. Furthermore, post hoc
comparisons for DPD salivary levels (Table 3) revealed that
there were signicant differences in the mean ranks of this
biomarker between every paired sampling times (all P <
G. A. FLÓREZ-MORENO ET AL.
0.05, Wilcoxon signed-rank test), except for the pair
baseline test/24–48 hours test (P = 0.514).
Discussion
The majority of the human studies concerning the biology
of OTM have focused on the analysis of different mediators
involved in alveolar bone remodelling after the use of
intrusive or extrusive forces of the same magnitude, which
not necessarily represent the complex three-dimensional
nature of OTM (Iwasaki et al., 2001; Perinetti et al., 2005;
Zhao et al., 2008). This is the rst study examining the
salivary concentration of bone remodelling biomarkers
Downloaded from https://academic.oup.com/ejo/article/35/3/361/450013 by guest on 28 February 2023
before and after use of orthodontic forces exerted by a
exible archwire model. Consequently, the prospective
design of the present study allowed maintaining the
correspondence with the normal clinical situation over time,
giving continuous but reducing active three-dimensional
archwire forces as tooth alignment occurred, like previously
described (Perinetti et al., 2004, 2005).
It has been recognized that both physiologic and
pathologic conditions can cause signicant changes in
systemic bone turnover (Seibel, 2000) and, therefore, can
have a profound effect on DPD and BAP results and on data
interpretation (Pellegrini et al., 2008). In order to control
and minimize these sources of variability, in this study,
bone marker concentrations were analysed regarding
demographic and clinical parameters at each time interval.
According to the present results, the participants were
relatively homogeneous, thus minimizing confounding by
gender, age strata, dentofacial pattern, dental arch crowding
severity, and location of orthodontic appliances.
Given that different mediators involved in alveolar bone
remodelling are continuously washed into saliva by GCF
(Frodge et al., 2008), whole-saliva samples may constitute
an easy alternative to individual gingival sulcular samples
for determining analytes of bone turnover that are present
within the periodontal environment, providing a sensitive
and inexpensive detection technique. In addition, although
it is difcult to compare the results of this study with those
of other investigations due to variations in the population,
experimental methodology, and mathematical treatment of
the data, it has been stated that the DPD and BAP levels
found in saliva appear to correspond well with their urinary
and serum concentrations (Johnson et al., 2002; McGehee
and Johnson, 2004; Pellegrini et al., 2008). In this study,
while salivary results of BAP were more consistent and
showed no differences in different times, only 54.5 per cent
of the ELISA analysis for DPD was detectable but
signicantly variable across the time. Although this nding
partially coincides with a previous report that failed to
detect DPD during OTM in GCF samples (Grifths et al.,
1998), taking into account that the majority of non-detected
values of the present study occurred mainly in the earlier
phases of OTM within the same patients, it should be
 SALIVARY
SALIVARYBIOMARKERS

Table 1 Quantitative determination of deoxypyridinoline (DPD) and bone-specic alkaline phosphatase (BAP) salivary concentrations by enzyme-linked immunosorbent assay at
each time interval according to demographic and clinical parameters.
BIOMARKERSAND

Parameter Number of Immunoenzymatic ndings*

cases
DPD (nmol/l) BAP (U/l)
ANDORTHODONTIC

Baseline test 24–48 hours test 2 weeks test 5 weeks test Baseline test 24–48 hours test 2 weeks test 5 weeks test

Gender
 Male 9 0.00 (0.00–3.44) 0.00 (0.00–5.81) 0.00 (0.00–5.09) 2.09 (0.00–7.80) 1.74 (0.00–4.02) 1.56 (0.81–2.97) 1.69 (0.00–16.36) 0.95 (0.00–2.88)
ORTHODONTICMOVEMENT
MOVEMENT

 Female 13 0.00 (0.00–3.20) 0.00 (0.00–3.46) 1.53 (0.00–5.56) 2.07 (1.12–5.12) 2.27 (0.00–8.16) 2.21 (0.00–6.85) 2.03 (0.00–6.57) 1.79 (0.00–7.49)
P-value** 0.695 0.948 0.051 0.695 0.522 0.300 0.548 0.110
Age strata (years)
 Under 18 13 0.00 (0.00–3.44) 0.00 (0.00–2.25) 1.26 (0.00–1.79) 2.06 (0.00–3.97) 2.36 (0.00–8.16) 2.21 (0.00–6.85) 1.79 (0.00–6.57) 1.37 (0.00–7.49)
 18 or more 9 0.00 (0.00–3.20) 1.17 (0.00–5.81) 1.49 (0.00–5.56) 2.77 (0.00–7.80) 1.18 (0.00–5.74) 1.32 (0.00–3.21) 2.03 (0.00–16.36) 1.41 (0.00–2.55)
P-value** 0.262 0.393 0.096 0.235 0.098 0.060 0.947 0.763
Dentofacial pattern
 Class I relationship 10 0.00 (0.00–1.78) 0.00 (0.00–5.81) 0.63 (0.00–1.53) 2.25 (0.00–7.80) 1.96 (0.00–8.16) 2.03 (0.00–6.35) 1.74 (0.00–3.92) 1.36 (0.00–2.55)
 Class II relationship 7 0.00 (0.00–3.44) 1.11 (0.00–1.27) 1.68 (0.00–5.56) 1.74 (1.12–3.09) 1.74 (0.00–7.85) 0.90 (0.00–4.36) 2.73 (0.00–16.36) 1.37 (0.00–5.22)
 Class III relationship 5 0.00 (0.00–1.24) 0.00 (0.00–3.46) 1.28 (0.00–4.01) 2.75 (0.00–4.97) 2.27 (0.00–4.02) 2.73 (1.56–6.85) 2.03 (1.13–6.57) 1.88 (0.95–7.49)
P-value*** 0.863 0.609 0.051 0.464 0.865 0.057 0.623 0.393
Dental arch crowding
 Mild 12 0.00 (0.00–3.44) 0.00 (0.00–5.81) 1.40 (0.00–5.09) 2.24 (1.12–7.80) 1.91 (0.00–8.16) 1.96 (0.00–6.85) 2.26 (0.00–16.36) 1.37 (0.00–7.49)
 Moderate 10 0.00 (0.00–3.20) 0.00 (0.00–3.46) 0.72 (0.00–5.56) 2.08 (0.00–5.12) 2.06 (0.00–7.85) 1.79 (0.00–4.36) 1.44 (0.00–3.92) 1.60 (0.00–3.17)
P-value** 0.872 0.923 0.821 0.628 0.641 0.895 0.176 0.895
Location of orthodontic appliance
 Monomaxillary 17 0.00 (0.00–3.44) 0.00 (0.00–5.81) 1.28 (0.00–5.09) 2.22 (0.00–7.80) 2.27 (0.00–8.16) 2.21 (0.00–6.85) 1.79 (0.00–16.36) 1.37 (0.00–7.49)
 Bimaxillary 1.12 (0.00–3.20) 0.00 (0.00–2.25) 1.44 (0.00–5.56) 1.75 (1.12–3.97) 0.70 (0.00–2.31) 1.17 (0.00–2.07) 2.40 (0.75–4.59) 1.79 (0.94–5.22)
P-value** 5 0.249 0.880 0.493 0.649 0.105 0.065 0.724 0.610

*Results are given as median (range).

**Mann–Whitney U-test.
***Kruskal–Wallis H-test.
5 of
3658

Downloaded from https://academic.oup.com/ejo/article/35/3/361/450013 by guest on 28 February 2023

6 of 8
366 G. A. FLÓREZ-MORENO ET AL.

Table 2 Quantitative changes and statistical correlation analysis of salivary concentrations of deoxypyridinoline (DPD) and bone-specic
alkaline phosphatase (BAP) at different sampling times. IQR, interquartile range; rs, Spearman’s rho correlation coefcient.

Trial phases Immunoenzymatic ndings Correlation analysis*

DPD (nmol/l) BAP (U/l) rs P-value

Median (IQR) Mean rank Median (IQR) Mean rank

Baseline test 0.00 (0.00–1.15) 1.84 1.96 (0.00–3.96) 2.66 −0.043 0.848
24–48 hours test 0.00 (0.00–1.20) 1.95 1.92 (1.10–3.03) 2.57 −0.089 0.693
2 weeks test 1.40 (0.00–1.71) 2.68 1.91 (1.16–3.03) 2.52 0.499 0.018
5 weeks test 2.08 (1.57–3.97) 2.99 1.39 (0.95–2.54) 2.25 0.122 0.589
Chi-square 30.399 1.311

Downloaded from https://academic.oup.com/ejo/article/35/3/361/450013 by guest on 28 February 2023

P-value** <0.001 0.727

*Spearman’s rank correlation test.

**Friedman test.

Table 3 Results of paired comparisons for deoxypyridinoline
(DPD) salivary levels across the sampling time.
Matched trial phases Z-value P-value*
Baseline test versus 24–48 hours test −0.652 0.514
Baseline test versus 2 weeks test −2.580 0.010
Baseline test versus 5 weeks test −3.340 0.001
24–48 hours test versus 2 weeks test −2.120 0.034
24–48 hours test versus 5 weeks test −3.783 <0.001
2-weeks test versus 5 weeks test −2.495 0.013
*Wilcoxon signed-rank test.
possible to assume that bone remodelling associated with
OTM in these cases may not generate DPD or it may be
conned within the tissues and not released (Grifths et al.,
1998). Alternatively, this might be caused by a high dilution
of DPD in saliva, which might derail the detection of low
levels of the biomarker. Even so, there is increasing evidence
suggesting that the biomarkers utilized in this study are
reliable predictors of bone remodelling when assayed from
whole saliva (McGehee and Johnson, 2004; Koka et al.,
2006; Pellegrini et al., 2008). Moreover, experimental data
have demonstrated that both total and salivary ow-adjusted
concentrations of these biomarkers are correlated,
suggesting that assessment of salivary ow rate is not
necessary for the accurate analysis of these analytes
(McGehee and Johnson, 2004).
The sampling times recorded in this study were selected
taking into account those three phases of OTM originally
described by Burstone (1962). The foremost ndings
reported herein were that while salivary DPD values
revealed a statistically signicant increasing nature at each
time interval after force application; salivary BAP levels
showed a non-signicant downward trend during time
intervals after activation visit. However, there are apparently
divergent data in the literature on the biological and clinical
signication of DPD and BAP levels concerning OTM.
Whereas some authors have demonstrated that DPD values
can signicantly decrease with force application (Isik et al.,
2005), others failed to detect DPD expression during
different stages of OTM in GCF samples (Grifths et al.,
1998). Also, different researchers have reported both
increasing (Perinetti et al., 2002; Batra et al., 2006) and
decreasing BAP levels (Isik et al., 2005; Asma et al., 2008)
as tooth movement occurs. Since bone remodelling
biomarkers are driven by a variety of mechanisms, a partial
explanation for the disagreement could be attributed to the
differences in the sampling methods/protocols, processing
methodology, sensitivity/specicity of the immunoassays,
as well as to the differences in the type of orthodontic
mechanotherapy, force levels, and sample size.
It has been thoroughly acknowledged that in the earlier
phases of tooth movement, bone resorption is greater than
bone apposition, but in a later phase, resorption and
apposition can become synchronous (Keeling et al., 1993;
Perinetti et al., 2002). In this study, salivary DPD levels
showed signicant differences between every paired
sampling times, except for the pair baseline test/24–48
hours test. It is quite probable that the increasing trend of
DPD values may be due to collagen breakdown resulting
from mechanical loading, ischemia, and hypoxia, which
appear immediately after force application and last
throughout most of the initial and lag phase of OTM
(Sprogar et al., 2010). Although the elevation observed at
24–48 hours was too low to reach statistical differences
with respect to baseline examination, it might be associated
with the initial shift of the teeth within the PDL space and
early bone resorption (Keeling et al. 1993) observed during
the initial phase of OTM. Additionally, in the lag phase (2
weeks test), when salivary DPD levels showed a signicant
increase regarding baseline and 24–48 hours tests, a
signicant correlation between DPD and BAP salivary
levels also could be noted. These ndings could be
SALIVARYBIOMARKERS
SALIVARY BIOMARKERSAND
ANDORTHODONTIC
ORTHODONTICMOVEMENT
MOVEMENT 7 of
consistent with the intense cellular activity of broblasts,
macrophages, osteoclasts, and osteoblasts, as well as the
reestablishment of cell and bre function, as previously
demonstrated (Krishnan and Davidovitch, 2006). Finally,
based on the data presented herein, the signicant augment
of salivary DPD values observed at 5 weeks test concurs
with the increased rate of tooth movement observed during
the linear (postlag) phase (Burstone, 1962; Krishnan and
Davidovitch, 2006). Though, this can represent an increase
in osteoblast and osteoclast number and activity (Ren et al.,
2005; Krishnan and Davidovitch, 2006) since bone
resorption is faster than bone formation (12 days versus 3
months), any increase in remodelling results in bone loss
and in a negative bone balance (Pellegrini et al., 2008).
Hence, although both DPD and BAP may act as determinants
of increased bone remodelling, it appears that DPD
dominates these phases of OTM, whereas BAP might serve
as indicator of the formation of new bone layers as soon as
the tooth movement stops. Altogether, these salivary
ndings might be a reection of the actual enzymatic prole
of GCF and consequently of the biologic activity within the
periodontal environment during OTM.
Several limitations were associated with this study.
Firstly, the small sample size due to the strict guidelines for
recruiting patients might have inuenced the results as the
immunoenzymatic ndings in these subjects might not
represent landmark data to evaluate the association between
bone remodelling biomarkers and the different phases of
OTM. An analysis with more patients would have greater
statistical power and precision. Secondly, factors such as
stress distribution in the PDL, three-dimensional quantication
of tooth movement, magnitude of the forces, and frictional
losses in the xed appliance were not standardized.
Nevertheless, this study design could accurately reproduce
the clinical effects of three-dimensional archwire forces that
are of direct clinical relevance, as reported in other studies
(Perinetti et al., 2002, 2004, 2005). Thirdly, the LOD of
ELISA kits was relatively high, so that lower concentrations
of the biomarkers could not be precisely measured. Hence,
the improvements in sensitivity of immunoassays might
justify not only further investigation of these bone markers
but also the use of other research modalities in molecular
biology that should help verify whether salivary levels of
DPD and BAP correlate with their expression at cellular level
during OTM.
Conclusions
The ndings when considered within the limitations of this
study indicate that although salivary levels of DPD and
BAP may act as indicators of increased bone remodelling, it
appears that DPD dominates the earlier phases of OTM,
whereas BAP might serve as indicator of bone formation as
soon as the tooth movement stops. Hence, these analytes
may serve in a panel of salivary functional biomarkers that
3678
could assist the screening of orthodontic treatment in the
clinical practice.
Funding
Research Development Committee of the University of
Antioquia (CODI-Code 8700-1618/2009)
References
Asma A A A, Rohaya M A W, Hisham S 2008 Crevicular alkaline
phosphatase activity during orthodontic tooth movement: canine
retraction stage. Journal of Medical Sciences 8: 228–233
Batra P, Kharbanda O, Duggal R, Singh N, Parkash H 2006 Alkaline
Downloaded from https://academic.oup.com/ejo/article/35/3/361/450013 by guest on 28 February 2023
phosphatase activity in gingival crevicular uid during canine retraction.
Orthodontics & Craniofacial Research 9: 44–51
Burstone C J 1962 The biomechanics of tooth movement. In: Kraus B S,
Riedel R A (eds). Vistas in orthodontics. Lea & Febiger, Philadelphia,
pp. 197–213
Davidovitch Z 1991 Tooth movement. Critical Reviews in Oral Biology
and Medicine 2: 411–450
Frodge B D, Ebersole J L, Kryscio R J, Thomas M V, Miller C S 2008
Bone remodeling biomarkers of periodontal disease in saliva. Journal of
Periodontology 79: 1913–1919
Garlet T P, Coelho U, Repeke C E, Silva J S, Cunha Fde Q, Garlet G P
2008 Differential expression of osteoblast and osteoclast
chemmoatractants in compression and tension sides during orthodontic
movement. Cytokine 42: 330–335
Giannopoulou C, Mombelli A, Tsinidou K, Vasdekis V, Kamma J 2008
Detection of gingival crevicular uid cytokines in children and
adolescents with and without xed orthodontic appliances. Acta
Odontologica Scandinavica 66: 169–173
Grifths G S, Moulson A M, Petrie A, James I T 1998 Evaluation of
osteocalcin and pyridinium crosslinks of bone collagen as markers of
bone turnover in gingival crevicular uid during different stages of
orthodontic treatment. Journal of Clinical Periodontology 25: 492–498
Ingman T, Apajalahti S, Mäntylä P, Savolainen P, Sorsa T 2005 Matrix
metalloproteinase-1 and -8 in gingival crevicular uid during orthodontic
tooth movement: a pilot study during 1 month of follow-up after xed
appliance activation. European Journal of Orthodontics 27: 202–207
Isik F, Sayinsu K, Arun T, Ünlüçerçi Y 2005 Bone marker levels in gingival
crevicular uid during orthodontic intrusive tooth movement: a
preliminary study. Journal of Contemporary Dental Practice 6: 27–35
Iwasaki L R, Haack J E, Nickel J C, Morton J 2000 Human tooth movement
in response to continuous stress of low magnitude. American Journal of
Orthodontics and Dentofacial Orthopedics 117: 175–183
Iwasaki L R, Haack J E, Nickel J C, Reinhardt R A, Petro T M 2001 Human
interleukin-1β and interleukin-1 receptor antagonist secretion and
velocity of tooth movement. Archives of Oral Biology 46: 185–189
Iwasaki L R et al. 2005 Tooth movement and cytokines in gingival
crevicular uid and whole blood in growing and adult subjects.
American Journal of Orthodontics and Dentofacial Orthopedics 128:
483–491
Iwasaki L R, Gibson C S, Crouch L D, Marx D B, Pandey J P, Nickel J C
2006 Speed of tooth movement is related to stress and IL-1 gene
polymorphisms. American Journal of Orthodontics and Dentofacial
Orthopedics 130: 698.e1–e9
Johnson R B et al. 2002 Effect of estrogen deciency on skeletal and
alveolar bone density in sheep. Journal of Periodontology 73: 383–391
Keeling S, King G, Valdez M 1993 Serum and alveolar bone phosphatase
changes reect remodeling during orthodontic tooth movement. American
Journal of Orthodontics and Dentofacial Orthopedics 103: 320–326
Koka S, Forde M D, Khosla S 2006 Systemic assessments utilizing saliva:
part 2 osteoporosis and use of saliva to measure bone turnover.
International Journal of Prosthodontics 19: 53–60
8 of 8
368
Krishnan V, Davidovitch Z 2006 Cellular, molecular, and tissue-level
reactions to orthodontic force. American Journal of Orthodontics and
Dentofacial Orthopedics 129: 469.e1–e32
Krishnan V, Davidovitch Z 2009 On a path to unfolding the biological
mechanisms of orthodontic tooth movement. Journal of Dental Research
88: 597–608
Lee K J, Park Y C, Yu H S, Choi S H, Yoo Y J 2004 Effects of continuous
and interrupted orthodontic force on interleukin-1beta and prostaglandin
E2 production in gingival crevicular uid. American Journal of
Orthodontics and Dentofacial Orthopedics 125: 168–177
McGehee J W Jr, Johnson R B 2004 Biomarkers of bone turnover can be
assayed from human saliva. Journals of Gerontology, Series A:
Biological Sciences and Medical Sciences 59: 196–200
Pellegrini G G, Gonzales C M, Somoza J C, Friedman S M, Zeni S N 2008
Correlation between salivary and serum markers of bone turnover in
osteopenic rats. Journal of Periodontology 79: 158–165
Perinetti G et al. 2002 Alkaline phosphatase activity in gingival crevicular
uid during human orthodontic tooth movement. American Journal of
Orthodontics and Dentofacial Orthopedics 122: 548–556
Perinetti G, Varvara G, Festa F, Esposito P 2004 Aspartate aminotransferase
activity in pulp of orthodontically treated teeth. American Journal of
Orthodontics and Dentofacial Orthopedics 125: 88–92
Perinetti G, Varvara G, Salini L, Tetè S 2005 Alkaline phosphatase activity
in dental pulp of orthodontically treated teeth. American Journal of
Orthodontics and Dentofacial Orthopedics 128: 492–496
Proft W R 2000 Biomechanics and mechanics. In: Proft W R (ed.)
Contemporary orthodontics. Mosby, St Louis, pp. 296–361
G. A. FLÓREZ-MORENO ET AL.
Ren Y, Maltha J C, Von Den Hoff J W, Kuijpers-Jagtman A M, Ding Z 2002
Cytokine levels in crevicular uid are less responsive to orthodontic
forces in adults than in juveniles. Journal of Clinical Periodontology 29:
400–405
Ren Y, Maltha J C, Kuijpers-Jagtman A M 2003 Optimum force magnitude
for orthodontic tooth movement: a systematic literature review. Angle
Orthodontist 73: 86–92
Ren Y, Kuijpers-Jagtman AM, Maltha JC 2005 Immunohistochemical
evaluation of osteoclast recruitment during experimental tooth movement
in young and adult rats. Archives of Oral Biology 50: 1032–1039
Seibel M J 2000 Molecular markers of bone turnover: biochemical, technical
and analytical aspects. Osteoporosis International 11(Suppl 6): S18–S29
Sprogar S, Meh A, Vaupotic T, Drevensek G, Drevensek M 2010
Expression levels of endothelin-1, endothelin-2, and endothelin-3 vary
during the initial, lag, and late phase of orthodontic tooth movement in
Downloaded from https://academic.oup.com/ejo/article/35/3/361/450013 by guest on 28 February 2023
rats. European Journal of Orthodontics 32: 324–328
van Leeuwen E J, Maltha J C, Kuijpers-Jagtman A M 1999 Tooth
movement with light continuous and discontinuous forces in beagle
dogs. European Journal of Oral Sciences 107: 468–474
von Böhl M, Maltha J C, Von Den Hoff J W, Kuijpers-Jagtman A M 2004
Focal hyalinization during experimental tooth movement in beagle
dogs. American Journal of Orthodontics and Dentofacial Orthopedics
125: 615–623
Zhao Z, Fan Y, Bai D, Wang J, Li Y 2008 The adaptive response
of periodontal ligament to orthodontic force loading—a combined
biomechanical and biological study. Clinical Biomechanics 23(Suppl
1): S59–S66