Bone. 12, pp. 40149 (1991) Printed in the USA. All rights reserved.
Copyright
8756-3282191 $3.00 + .OO 0 1991 Pergamon Press plc
Histomorphometric Study of Alveolar Bone Turnover in Orthodontic Tooth Movement G. J. KING,’
S. D. KEELING’
and T. J. WRONSKI’
‘Department of Orthodontics, University of Florida, College of Dentistry, Gainesville, FL, U.S.A. ‘Department of Physiological Sciences, University of Florida, College of Veterinary Medicine, Gainesville, FL, U.S.A. Address for correspondence and reprints: Gregory J. King, Box J-444, JHMHC,
Abstract Selected histomorphometric parameters were measured in alveolar bone adjacent to rat molars treated with a 40gram tipping force designed to tip the molar mesially and with a sham procedure. Undecalcified parasagittal sections of teeth and surrounding tissues were prepared for static and dynamic hlstomorphometry at 1, 3, 5, 7, 10, and 14 days. Inltlal tooth displacement was assessed by measuring differences between groups in the widths of the day-l perlodontal ligaments (PDL) at various vertical locations and correlating these using linear regression analysis. All parameters were measured in the alveolar bone adjacent to four quadrants around the teeth (mesial-distal; occlusalapical). Means and standard errors for each parameter ln each group were calculated and compared for time- and group-specific differences using ANOVA and pairwlse comparisons with Scheffe’s multiple comparison tests. In shamtreated rats, bone resorption predominated on the distal alveolar surface, but signlficant surface-related differences between mesial and distal surfaces ln bone formation were not demonstrated. Time-specific effects in bone resorption were not evident on either surface in the shams. These findings suggested that molar distal drift in the sham rat is facilitated by resorption in the distal alveolar bone. PDL width changes in orthodontically-treated rats were greatest in the me&l occhtsal half of the root and decayed linearly toward the apex indicating that the greater initial tooth displacement occurred in the occhtsal half of the root surface. Histomorphometric parameters of alveolar bone turnover were also seen to be greater in these locations, suggesting that these processes were sensitive to the increased tooth displacement. In orthodontically-treated rata, the total paradental alveolar bone and both pressure (mesial) and tension (diital) surfaces experienced a wave of resorption that lasted for four days, was preceded by an activation period of two days, and was followed by a wave of formation that persisted until the end of the experiment (14 days). The relative balance between bone formation and resorption waves was location-specific, with formation predominant on the tension side and resorption predomlnant on the pressure. These fhulings are consistent with the conclusion that pressure and tension initiate similar bone turnover events, and that bone deposition or removal is controlled by the degree
Gainesville,
FI 32610, U.S.A.
of balance between the relative amounts of bone formation and resorption that occur at these sites. Key Words: Bone turnover-Tooth Formation -Orthodontic.
movement-Resorption-
Introduction The movement of teeth in the alveolar process requires bone turnover. An extensive literature exists demonstrating that orthodontic treatment has the capacity to alter dramatically a preexisting pattern of alveolar bone turnover, with bone resorption occurring at sites of pressure and formation in areas of tension (Reitan 1967; Storey 1973; Rygh 1972, 1976). However, there are no histomorphometric studies on how these bone turnover dynamics are expressed as functions of time in the total paradental alveolar bone or at locations adjacent to teeth with defined orthodontic displacements. Under routine conditions, orthodontic treatment results in no net long-term changes in the periodontal ligament (PDL), the amount of alveolar bone, or its nature (Sadowsky et al. 1981). This could occur if the amount of bone formation stimulated was simultaneously balanced by an equal amount of resorption. This level of balance has been shown to exist during distal drift of rat molars (Vignery et al. 1980). Alternatively, orthodontic stress may act to stimulate bone turnover events resulting in waves where the resorption and formation occur in tandem. Such a mechanism would predict an early transient bone loss and temporary widening of the PDL, followed by bone formation, which would return the tissue architecture to normal. This view is consistent with the observation that alveolar bone shows a transient osteopenia followed by recovery during orthodontic treatment (Bridges et al. 1988); that a wave of increased osteoclast numbers occurs during the fit few days of orthodontic treatment (Markostamou et al. 1973); that bone cells at sites of formation go through a synchronized remodeling cycle following the mechanical stimulus of removal of an opposing tooth (Tran Van et al. 1982); and that orthodontists report teeth are mobile and display a widened PDL on radiographs during active treatment.
G. J. King
402
et al.: Bone
This study investigated, histomorphometrically, the time course of changes in alveolar bone turnover at locations in the alveolar socket where the tooth displacement geometry was known during a precisely specified orthodontic treatment. Materials and Methods Seventy-two adult male Sprague-Dawley strain rats (89-94 days old) were purchased from Charles River Breeding Laboratories (Wilmington, MA). These were shipped by air freight and acclimatized for at least two weeks under experimental conditions, including housing in plastic cages, a diet of ground laboratory chow and distilled water ad libitum, and a standard 12-hour light/dark cycle. All animal manipulations, including sacrifices, were done at the same time of day. Weights were recorded and anesthesia was obtained using intra-muscular injections of ketamine (87 mg/kg) and xylazine (13 mg/kg). Modified orthodontic cleats were bonded bilaterally to the occlusal surfaces of all maxillary first molars as previously described (King et al. 1991), and the mandibular first and second molars were extracted. The animals were then allowed to recover for one week by monitoring wound healing and weight gain. For appliance activations, the animals were positioned in a head restrainer and orthodontic springs were placed as previously described (King et al. 1991). Briefly, one end of a length of closed coil spring was ligated to the molar cleat while the other was attached to a suspended weight. The anterior end of the coil was then bonded with autocuring methacrylate to the acid-etched lateral surface of the maxillary incisors, followed by removal of the weight and excess coil spring. This method assured a precise initial orthodontic force designed to tip the maxillary first molars to the mesial. The rats were randomly divided into two groups of equal size. One group received the orthodontic force at a level of 40
tooth
movement
Table I. Parameters obtained by histomorphometric
measurements
Parameter
Formula
Abbreviation
Osteoclast number (#/surface lgth.) Osteoclast surface (%) Eroded surface (%) Osteoblast surface (%) Osteoid surface (%) Osteoid thickness (mcm) Mineralizing surface (%)
N.Oc.
N.Oc./B.Pm
Oc.Pm. E.Pm. Ob.Pm. O.Pm. O.Th. M.S.
Mineral apposition rate (mcm/day) Bone formation rate (mcm3/mcm2/day) Reversal surface (%)
M.A.R.
Oc.Pm./B.Pm. E.Pm./B.Pm. Ob.Pm./B.Pm. O.Pm./B.Pm. O.Th./N. [dL Pm. + (sL Pm/2)]/ B.Pm. 1r.L. Th/ 1r.L.t.
B.F.R.
(M.A.R.
Rv.Pm.
E.Pm.
Nomenclature according Additional abbreviations double labeled perimeter, inter-label thickness, 1r.L.t: Fig. 1. Camera lucida drawing of the tissue under investigation. AB - alveolar bone; PDL - periodontal ligament; TR - tooth root; P - pulp chamber; MO - mesio-occlusal quadrant; DO - Disto-occlusal quadrant; MA - mesio-apical quadrant; DA - disto-apical quadrant. Representative non-overlapping fields for histomorphometric analysis at 400 x magnification are displayed as squares in the right DO quadrant.
in orthodontic
X M.S.)/B.Pm.
-
Oc.Pm.
to Parfitt et al. 1987. as follows: B.Pm: bone perimeter, dL Pm: sL Pm: single labeled perimeter, Ir.L.Th: inter-label time.
while the other was sham-treated, receiving all procedures except spring placement. The spring was excluded to allow normal distal drift of the molars. Six animals from each group were sacrificed at each of six timepoints (1, 3, 5, 7, 10, and 14 days). Previous studies have indicated that these orthodontic parameters and time intervals adequately produce and monitor tooth movement kinetics, which exhibit characteristic changes including instantaneous movement, a delay period and late tooth movement (King et al. 1991). All rats were injected intraperitoneally with tetracycline derivatives (15 mg/kg) on two separate occasions. Oxytetracycline hydrochloride (Terramycin; Pfizer, Inc., Brooklyn, New York) was administered five days prior to sacrifice. After a two-day period with no tetracycline administration, all rats were injected again with demeclocycline (Declomycin, Lederle Laboratories, Pearl River, New York) two days before sacrifice. This regimen resulted in the deposition of a double tetracycline label at bone surfaces that were actively mineralizing throughout the injection period (Baron et al. 1983). Animals were killed by decapitation at each of the time intervals. Maxillae were dissected into halves and fixed in 70% ethanol for 24 hours, followed by dehydration in increasing concentrations of ethanol and embedding undecalcified in methyl metacrylate (Baron et al. 1983). The embedded samples were sectioned parallel to the long axis of the first molar teeth at 4 pm thicknesses with an A0 AutocutIJung 1150 Microtome and mounted on 1% gelatinized slides. Alternate sections were stained according to the Von Kossa method with a tetrachrome counterstain. All surface-based parameters were quantified on these sections using a semi-automatic image analyzer (Bioquant, R & M Biometrics, Nashville, TN) coupled to a photomicroscope equipped with epifluorescence (Nikon Labophot). All tetracycline-derived parameters were measured on similarly mounted unstained sections cut at 6 p,m thicknesses. Paradental and inter-radicular alveolar bone immediately adjacent to the periodontal ligament of disto-buccal and middle-buccal maxillary first molar roots was sampled as follows: Parasagittal sections were selected by choosing those demonstrating a radicular pulp to the apical third of the root. The tooth roots, pulp chambers, and alveolar bone surfaces were traced at low magnification using the camera lucida (Fig. 1). A line was drawn on this tracing bisecting the root longitudinally. This line was then divided by one drawn perpendicular to it at the vertical midpoint of the tooth socket. These two lines grams,
G. J. King et al.: Bone in orthodontic
403
tooth movement
Initial root displacement in the sagittal plane
R”2 I 0.898
-20: 0
-
. -6 * . I . 20 40
Difference
.
I . 60
.
I . 80
.
1 100
in mesial PDL width (pm)
Fig. 2. Linear regression analysis between PDL width differences (control minus treated) at 20 equally spaced points vertically along the root.
were used to define equal-sized quadrants around the root (i.e., mesio-occlusal, mesio-apical, disto-occlusal, and disto-apical). Multiple non-overlapping fields at 400 X were then sampled from each quadrant. The bone parameters that were measured are shown in Table I and are defined according to Pa&t et al. (1987). One section was measured for each hemimaxilla, thereby preventing the sampling of the same area twice. The means and standard errors of each parameter were calculated for each group, by location (i.e., total section, side, quadrant) and timepoint. An analysis of variance (ANOVA) was performed to examine differences among force groups at each timepoint and across times within each group. Pairwise (Scheffe, 1953) comparisons were performed between groups, locations, and days when ANOVA indicated that significant differences existed (p < 0.05). Since bone remodeling does not commence until after day 1, alterations in PDL width were considered to adequately represent the initial tooth displacement. This was assessed on the mesial PDL space by measuring the difference in width between control and treated sections from day 1 at each of 20 equally spaced locations vertically along the root from the crestal bone to the apex. A linear regression line was then used to depict the initial tooth displacement and its center of rotation.
RMlftS Initial tooth displacement There was a highly significant correlation (R2 = 0.90, p = 0.0001) between vertical location on the root and mean difference in PDL width between control and treatment groups at day 1 (Fig. 2). The greatest reduction occurred at the alveolar
crest (i.e., occlusal) and the least at the apex. This root displacement is consistent with the teeth being tipped toward the mesial with a center of rotation at or near the root apex. Using traditional orthodontic terminology, the mesial-occlusal quadrant is in “pressure” and the disto-occlusal is in “tension.” The observations obtained from total alveolar sections, the mesial occlusal quadrant, and the distal occlusal quadrant are reported below. The occlusal quadrants are displayed because they represented the sites of greater tooth displacement.
Resorptive parameters Data on the resorptive parameters are presented in Table II. For the total alveolar bone section, ANOVA indicated no differences across time in the control group. However, significant differences across time existed in the treatment group for osteoclast perimeter, osteoclast numbers, and erosion perimeter, where Scheffe comparisons indicated that day 5 represented a consistent peak. Thus, the alveolar bone around the orthodontically-treated teeth experienced a period, lasting from l-3 days, when there were no increases in these parameters. This was then followed by a wave of increase that reversed between 5 and 7 days. A similar trend also exists in reversal perimeter but this did not achieve statistical significance. When comparisons between quadrants (i.e., mesial-occlusal, distal-occlusal) are considered, the control group had consistently higher resorptive parameters on the distal alveolar bone surface than the mesial at each timepoint, but there were no significant differences across time at either location. In the orthodontically-treated group, ANOVA indicated that significant differences in osteoclast perimeter, osteoclast numbers, and erosion perimeter across time existed on the mesial-
0
2
4
6
8
10 12 14
-50
-40
-30
30
40
50
0
E
8
S
t
m
3
1
0
2
-30,1-50
-25 -1
-20
-15
-10
-5
0
5
10
3 g r
15
-a-
20
25
4
6
:
: :
8
, _--
-40
t
t 40
-30
-20
-30
1:
10 12 14
4’ 4’
13 ,.
Tension side: distal occlusal quadrant
1
-25-
0
2
-30 I-50
\ \, .
4
/' f
6
8
10 12 14
- -40
Pressure side: mesial occlusal quadrant
Fig.3. Bone turnover as result of orthodontic force, depicted by changes in osteoclast and osteoblast perimeter minus control values over time. Left panel - Total pamdental tissue. Differences across time existed in % osteoclast surface (p < .OOOl), manifested by a peak at days 3 and 5 (versus days I, 10, and 14; p < .05). Differences across time existed in % osteoblast surface @ < .OOOl), characterized by a peak at days 10 and 14 (versus days 3 and 5, p < .05). Middle panel - Tension side: No differences were detected in % osteoclast surface over time @ > .05). Differences across time existed in % osteoblast surface @ < .OOl), manifested by a peak at day 14 (versus days 1, 3, and 5: p < .05). Right panel - Pressure side: Differences across time were detected in % osteoclast surface @ < .OOl), manifested by a peak at day 5 (versus days I and 14; p < .05). Differences across time were also detected in % osteoblast surface @ < ,001). where a peak occurred at day 14 (versus days 3 and 5; p < .05).
-30
-25
-20
-15
20
25
30
(40 gram dafa adjusted tar moan conlrol valuoa)
Total paradental tissue
G. J. King et al.: Bone in orthodontic
tooth movement
405
Table II. Mean and standard error of the mean for resorptive parameters by group and day Control group
Day Oc.Pm 3 5 10 14
Mesial Occl. 0.0 4.7 3.6 0.0 1.8 1.0
2
* ‘5 i k NS
Distal Occl.
0.0 2.8 2.3 0.0 1.8 0.5
14.8 20.9 9.0 11.1 15.2 17.8
k 2.9 f 4.8 f 1.1 k 1.3 k 4.9 ~fr 3.4 NS
Treatment group Total
Mesial Occl.
Distal Occl. 11.2 11.3 12.5 3.8 3.9 1.6
8.1 7.5 6.5 6.8 1.3 10.7
k 2 k 2 k ‘NS
1.3 1.1 0.7 0.9 1.7 2.2
3.8 18.4 25.2 11.7 9.1 4.8 ***,
k 0.9 2 4.2 k 2.3 2 5.1 2 3.3 k 2.0 Note 1
2 2 2 2 k k NS
Between Group ANOVA’ Total
3.1 5.7 1.6 1.8 3.4 1.6
6.6 16.5 16.2 6.9 5.2 3.7 ***,
k 1.1 k 2.4 + 1.7 L 2.0 2 2.7 k 1.7 Note 2
Mesial
Distal
Occl
Occl.
Total
* *** *
NS NS NS *
NS *
NS NS
NS **
** NS NS *
**
0.3 1.5 1.9 0.2 0.5 0.9
2 2 t t ? k NS
0.3 1.0 0.8 0.2 0.5 0.4
5.4 7.3 4.4 5.3 5.3 6.4
f k t + + + NS
1.3 1.0 1.1 1.0 1.9 1.0
3.3 2.8 2.7 3.0 2.5 4.6
” 2 2 2 k k NS
0.8 0.5 0.4 0.6 0.6 0.9
1.7 6.6 12.4 3.5 3.1 1.9 ***,
2 0.4 2 1.6 2 1.4 k 1.3 2 1.1 k 1.1 Note 3
3.8 k 1.1 2.3 k 1.2 5.4 2 1.1 1.3 k 0.6 1.1 k 0.7 0.8 2 0.6 *, Note 0
2.1 4.9 6.7 2.1 1.9 1.4 ***,
k 0.3 + 0.6 2 0.6 + 0.6 k 0.8 ” 0.7 Note 3
* * *** *
NS *
NS *
NS **
NS NS
NS **
** NS NS *
+ k + 2 2 i NS
10.1 7.7 6.6 3.0 7.3 4.1
50.0 48.7 41.9 56.1 49.6 59.9
k -c + 2 k ?I NS
5.3 2.2 7.5 4.1 8.1 6.8
27.6 25.1 26.5 30.6 26.1 38.7
2 k 2 k k 2 NS
2.3 4.8 2.1 3.5 2.8 6.8
20.4 56.9 65.0 41.6 35.7 40.4 **,
+ 1.3 + 5.2 + 12.2 + 11.2 k 7.2 k 5.7 Note 4
54.9 k 7.8 44.2 i- 8.6 57.3 k 3.3 37.7 k 5.4 24.9 2 11.0 28.3 + 10.3 *, Note 0
31.8 51.2 60.5 43.0 34.8 35.0 **,
k 5.4 L 4.0 k 3.6 k 5.1 k 8.6 + 6.3 Note 5
NS *** *** * * *
NS NS NS *
14
10.1 12.1 16.2 4.9 8.7 19.0
NS *
NS ** *** NS NS NS
5 I 10 14
10.1 7.4 12.6 4.9 6.9 18.0
+ k + k k k NS
10.1 6.3 7.2 3.0 5.5 3.9
35.2 27.8 32.9 45.1 34.4 42.1
k 2 2 2 -t2 NS
4.5 6.8 6.6 4.3 6.6 4.2
19.6 17.5 20.0 23.9 18.8 27.9
2 * 2 2 2 2 NS
2.6 5.0 2.1 3.2 2.1 5.0
16.6 38.5 39.9 30.0 26.6 35.6
k k k f 2 k NS
43.1 32.9 44.8 34.0 21.1 26.7
25.2 34.6 44.3 36.1 29.6 31.3
+ k 5 k k t NS
NS * * * * *
NS NS NS NS NS NS
NS * ** NS NS NS
N.Oc. 5 7 10
14 E.Pm. 5 10
Rv.Pm
6.9 7.1 4.9 9.7 6.0 5.1
+ + k * 2 k NS
9.0 4.8 3.7 5.1 9.6 9.4
4.8 3.0 4.7 5.3 7.5 5.4
‘Levels of significance for ANOVA between groups at each day for each location: *p < 0.05; **p < 0.01; ***p < 0.001. Levels of significance for ANOVA among days within each group by location: *p < 0.05; **p < 0.01; ***p < 0.001. Notes on results of Scheffe comparisons between days where p < 0.05: (0) None; (1) 5 vs 1 & 14; (2) 3 vs. 1, 10 & 14; 5 vs. 10 & 14; (3) 5 vs 1, 7, 10 & 14; (4) 1 vs 3 & 5; (5) 1 vs 5
occlusal surface (pressure side), also manifested by a peak at day 5. Less dramatic trends also occurred on the distal surface (tension side), which followed the same timing as those seen on the pressure side. When comparisons at each location were made between treatment and control groups at each time interval, ANOVA indicated that the values for the treatment group were greater than those for the control group at days 3 and 5 for all resorptive parameters on total sections and on the pressure side. Numerous time intervals showed treatment effects for all resorptive parameters on the mesial (pressure) side; far fewer differences were demonstrable on the distal (tension) side. Formative parameters Data on the bone formation parameters are presented in Table III. For the total alveolar bone section, ANOVA indicated that there were no significant differences in shams over time in either OPm., O.Th., or M.A.R.; but time-specific changes in M.S. and B.F.R. were detectable, characterized by a peak at three days and a depression at seven. In the orthodonticallytreated group, ANOVA indicated that significant differrices existed across time @ < 0.001) in osteoblast perimeter, manifested by values at days 3 and 5 being significantly lower
than those at days 10 and 14 @ < 0.01). Thus a significant peak in osteoblast numbers occurred between days 5 and 10, remaining elevated at the end of the observational period on day 14. These changes were also reflected in osteoid (p < 0.01) perimeter. In the treated group, ANOVA indicated that significant differences in percentage of osteoblasts occurred across time on both the mesial (pressure) an distal (tension) sides. On the mesial side, these differences were manifested by values at day 14 being greater than those at days 3 and 5 0, < 0.05). In the treated group, changes in the osteoid perimeter reflected those seen in osteoblast numbers characterized by a significant peak occurring on the total section by day 10 (p < 0.01). Similar trends are also evident on the mesial and distal. No differences were evident in bone formation rate or osteoid thickness. There was a significant elevation in mineralizing surface on the mesial at day 1, which is consistent with this surface being primarily formative prior to orthodontic force application and with one day being an insufficient period of time for removal of preincorporated tetracycline labels. There were no significant time-specific differences in mineral apposition rate. Changes in osteoblast and osteoclast perimeters are shown graphically in Fig. 3 by subtracting appropriate control values
G. J. King et al.: Bone in orthodontic tooth movement
406
Table III. Mean and standard error of the mean for the formation
Day
0. Pm.
0. Th.
M. S.
M.A.R.
B. F. R.
by group and day
Treatment Group
Control Group
Parameter
Ob. Pm.
parameters
Mesial
Distal
Occl.
Occl.
Total
1 3 5 7 10 14
27.8 43.1 64.3 80.8 80.6 62.3
k 11.9 k 16.3 + 13.6 2 6.4 + 8.3 L 9.1 *, Note 0
19.1 ? 8.2 26.0 ? 12.5 38.9 f 10.2 54.5 * 3.1 61.1 ? 5.3 43.2 c 7.5 **, Note 1
32.2 40.7 49.2 62.9 67.2 48.0
2 7.6 k 15.1 k 10.8 k 3.0 f 4.7 + 5.5 *, Note 0
1 3 5 7 10 14
16.1 30.6 43.3 61.7 59.2 40.3
2 10.3 k 13.7 f 11.0 k 9.9 k 10.4 t 8.9 *, Note 0
10.9 11.7 21.6 26.9 33.1 23.1
22.6 26.1 31.5 35.0 43.6 26.0
k 6.0 k 10.1 t 8.2 k 4.4 2 7.1 2 6.1 NS
1 3 5 7 10 14
0.1 0.4 0.4 0.4 0.4 0.5
1 3 5 7 10 14
42.7 64.5 37.5 3.7 22.2 32.9
2 t + 2 k +
0.1 0.1 0.0 0.0 0.1 0.2 NS
2 14.6 k 11.4 2 13.8 + 1.6 + 11.1 k 6.1 *, Note 6
1 3 5 7 10 14
3.92 4.8 4.1 2.6 2.5 2.5
+ 2 f f +
1 3 5 7 10 14
1.1 2.9 1.6 0.2 1.0 1.3
2 k k + + 2
0.4 0.4 0.3 0.4 0.5 0.4 1.9 37.5 31.6 13.7 9.6 15.5
? 2.0 t 7.0 2 10.0 % 3.1 + 8.6 % 8.1 NS * 2 2 + 2 *
0.1 0.1 0.1 0.0 0.1 0.0 NS
z? 1.9 + 11.3 f 11.8 + 13.7 * 3.4 + 7.8 NS
0.3 0.4 0.4 0.4 0.4 0.4
+ 2 + k ? L
0.0 0.1 0.0 0.0 0.1 0.1 NS
20.8 2 6.3 56.6 + 8.8 36.0 + 10.2 9.2 2 6.1 21.3 + 7.8 22.2 f 2.6 **, Note 6
Mesial
Distal
Occl.
Occl.
Between Group ANOVA’ Mesial Total
Occl.
Distal Occl.
*
29.6 t 9.1 17.7 k 2.2 22.1 2 2.7 45.2 i 11.8 51.6 + 12.8 68.0 k 4.6 ***, Note 2
45.2 k 7.2 42.7 + 9.6 39.5 + 5.5 71.8 f 6.5 73.4 2 2.7 82.1 k 4.6 ***, Note 3
43.7 2 3.9 33.0 2 5.1 35.6 k 4.2 56.4 k 7.3 66.9 k 5.3 70.0 k 4.9 ***, Note 4
NS NS * *
NS NS *
NS NS
NS **
14.6 9.6 15.0 20.9 28.2 36.2
21.7 23.2 21.9 37.0 45.2 48.6
k 6.3 + 6.4 ‘- 4.0 + 5.2 k 3.8 + 15.2 *, Note 0
22.5 + 3.1 16.5 k 3.0 18.1 k 3.2 27.8 + 4.1 39.3 k 5.2 37.0 k 7.6 **, Note 5
NS
NS
NS * **
NS
0.4 0.7 0.4 0.4 0.5 0.6
_f 0.0 5 0.2 5 0.0 k 0.0 + 0.0 ‘- 0.1 NS
0.5 0.5 0.5 0.5 0.6 0.5
t 3.2 i- 2.1 + 3.4 + 6.0 L 8.7 2 6.2 *, Note 0 + * k + + 2
0.1 0.0 0.1 0.0 0.1 0.0 NS
70.9 k 10.0 6.6 + 4.0 24.1 2 7.3 11.8 k 8.1 16.7 f 8.6 28.6 + 8.0 ***, Note 7
21.9 k 11.3 48.6 + 4.9 20.3 k 7.0 64.0 5 8.6 44.0 2 10.4 14.5 k 5.5 **, Note 8
1.0 0.5 1.4 1.2 0.9 0.6 NS
3.0 4.4 4.6 0.5 3.3 1.7
+ + + + + 2
2.3 0.8 0.4 0.5 1.2 0.5 NS
3.4 k 0.6 9.5 + 4.7 3.8 2 0.7 1.3 k 0.5 2.7 2 0.6 2.2 + 0.2 *, Note 0
6.7 2.7 3.1 2.0 2.6 2.2
k 1.1 k 1.2 k 0.8 k 1.3 k 1.1 k 1.0 *, Note 0
3.5 5.5 3.1 9.9 3.3 5.9
0.6 1.0 0.6 0.1 0.5 0.6 NS
0.3 2.4 1.8 0.2 1.7 0.4
f 0.3 ? 1.4 ? 0.9 & 0.2 * 1.1 + 0.3 NS
0.7 f 0.4 3.1 k 1.0 1.7 2 0.5 0.2 f 0.1 1.2 + 0.3 0.9 2 0.2 **, Note 9
4.6 + 1.1 0.4 2 0.2 1.1 k 0.5 1.7 + 1.6 1.1 + 0.9 1.4 f 0.7 *, Note 0
1.1 5.0 0.9 8.2 2.1 1.8
+ 1.0 2 1.2 2 0.7 + 2.2 lr. 1.2 + 1.4 *, Note 0 + 0.8 k 1.5 k 0.2 + 3.0 + 0.9 k 1.1 *, Note 0
0.4 0.5 0.4 0.4 0.6 0.5
k 0.0 2 0.1 f 0.0 k 0.0 2 0.0 2 0.1 *, Note 0
NS NS * NS NS NS NS NS
Total NS NS NS NS NS *
NS NS NS NS
NS NS NS NS NS NS
NS NS NS NS NS NS
NS NS NS NS NS
**
* *
45.4 31.8 33.0 24.7 28.5 24.0
2 + k k + 2
7.5 4.2 4.7 4.9 5.2 6.3 NS
NS ***
4.8 5.0 4.1 5.5 2.6 3.5
5 k k ” 2 k
0.5 0.5 0.9 0.8 0.8 0.7 NS
NS NS NS NS NS NS
NS NS NS **
NS
NS *
NS NS
2.6 3.2 2.5 3.2 1.4 1.4
+ 0.6 ” 0.7 + 0.7 ‘- 0.9 k 0.7 If: 0.5 NS
* *
NS NS NS *
* NS NS *
NS NS
NS NS
NS NS NS NS
NS NS NS NS
NS NS NS * * NS
NS NS NS NS
NS NS **
‘Levels of significance for ANOVA between groups at each day for each location: *p < 0.05; **p < 0.01; ***p < 0.001. Levels of significance for ANOVA among days within group by location: *p < 0.05; **p < 0.01; ***p < 0.001. Notes on results of Scheffe comparisons between days where p < 0.05: (0) None; (1) 1 vs 10; (2) 14 vs 3 & 5; (3) 14 vs 1, 3, & 5; 5 vs. 10; (4) 3 & 5 vs. 10 & 14; (5) 3 vs 10; (6) 3 vs 7; (7) 1 vs 3, 5, 7, 10, & 14; (8) 7 vs 14; (9) 3 vs 1 & 7.
to depict the predominant bone turnover dynamics in the orthodonticahy treated tissues. To facilitate comparisons among locations, the left vertical axes are set to the same scale, as are the right; the zero points of each vertical axis superimpose. In the total tissue (left panel), an early increase in osteoclast perimeter is evident beginning between days 1 and 3 and reversing between days 5 and 7. There is a tendency for this parameter to be inhibited in the later timepoints. Osteoblast surface is already inhibited on day 1. This inhibition is further increased at days 3 and 5, but reversed to stimulation by days 7, 10, and 14.
The middle panel demonstrates changes in the paradental quadrant where the tooth is initially most displaced away from the socket wall (i.e., tension side). Neither osteoblast nor osteoclast perimeter differ appreciably from control until an interval between days 5 and 7, at which time marked stimulation of osteoblast and inhibition of osteoclast perimeter occur. The right panel represents similar data from the pressure quadrant. Osteoclast perimeter shows a marked stimulation beginning between days 1 and 5 and remaining elevated until an interval between days 10 and 14, with a peak period around day 5. Osteoblast perimeter is already significantly inhibited
G. J. King et al.: Bone in orthodontic
tooth movement
on day 1, and remains depressed until late in the tooth movement, when it returns to levels slightly above control.
DlliOIl
The orthodontic mechanics used in this study produce an initial tooth displacement characterized by mesial tipping around a center of rotation located at the root apex. Sites where the tooth root is displaced toward the socket wall have been categorized as being in pressure, and those where the root is moved away from the socket wall as tension (Reitan 1967; Storey 1973). Based on these definitions, the mesial surface would be categorized as a pressure site and the distal as tension in this model. The histomorphometric changes reported here are consistent with numerous reports indicating that sites of orthodontic pressure are primarily bone resorptive, while those in tension are formative (Reitan et al. 1967). In spite of this association, interpretation of tooth displacement data as representing specific strains in the tissues should be approached with caution, as data on alveolar bone strains associated with specific orthodontic tooth displacements are unavailable. For convenience and consistency with the tooth movement literature, we have adopted the ubiquitous pressure/ tension terminology, realizing that it does not adequately describe the biomechanical environment initiated in the tissues by these appliances. The bone turnover dynamics in the total paradental tissues observed in this experiment are inconsistent with the hypothesis that orthodontic tooth movement occurs by two balanced bone modeling processes (i.e., resorption under pressure and formation in tension). Over the short term, bone resorption and formation are not balanced, but instead pass through a period (l-3 days) of little change, a period that is predominantly resorptive (up to days 5-7), and a phase that is primarily formative beginning between days 5-7 and lasting until the end of the experiment. This finding is consistent with the earlier observation of a transient osteopenia in orthodontically-treated paradental alveolar bone (Bridges et al. 1988), and points to a distinct difference from distal molar drift where bone turnover does remain in balance (Vignery et al. 1980). The finding that the activation of bone resorption required two days agrees very well with that for molar migration during tooth egression following removal of an opposing tooth (Tran Van et al. 1982). However, the resorption phase was slightly longer in this study than in other reports using adult rodents (i.e., 1.5-2.8 days; Vignery et al. 1980; Baron et al. 1981; Vignery et al. 1978; Tran Van et al. 1982). The estimate from this study of a bone formation phase longer than 8 days agrees reasonably well with that found in earlier experiments (greater than 5 days; - Tran Van et al. 1982 and 15 days; - Baron et al. 1981). The similarity of activation times between this model and tooth egression suggests that similar mechanisms may be required in both. This is distinct from the significantly shorter activation period (0.5 days) observed in PTH-stimulated rodent alveolar bone (Vignery et al. 1978). It had been previously suggested that this difference may be due to the existence of significant numbers of osteoclast precursors in the remodeling site in the PTH study as compared to the tooth migration model. Sites of future resorption in the orthodontic model are also considered to be poor in osteoclast precursors. Although there is bone resorption occurring around control teeth, this is almost exclusively on the distal alveolar bone surface (Table II). By contrast, the primary site of resorption in the treated animals is on the mesial alveolar bone surface. Another previously cited difference between these animal models is that one
407
utilizes a mechanical signal whereas the other is hormonal. Like tooth egression, orthodontic tooth movement is also initiated by a mechanical signal, and therefore lends further support to the idea that mechanical activation of bone remodeling has a different mechanism than hormonal. The longer bone resorption phase reported in this study may represent an additional difference between the orthodontic and distal dental drift paradigms. In orthodontic tooth movement, a significant amount of tissue damage appears to be unavoidable (Rygh 1972; Rygh 1973; Rygh et ai. 1973). Such changes have not been reported in tooth migration or drift. Microdamage in bone occurring as a result of fatigue has been linked to the initiation of bone remodeling (Burr et al. 1985). This association has also been reported in the osteoporosis seen in conjunction with the stabilization of fractures using implants (Perren et al. 1988) and adjacent to endosseous titanium implants (Roberts et al. 1989). The finding that a greater amount of bone resorption occurs in the occlusal half of the pressure side of the orthodontically-treated teeth further suggests that the resorption response may be related to tissue damage in this model, as areas of localized necrosis of the PDL are seen only in this quadrant during the timeframe of the resorption wave. However, damage and repair does not entirely explain the bone turnover dynamics reported here, because tissue necrosis is not seen in the other quadrants, yet similar bone changes occur. Frost has proposed that skeletal tissue relates to its biomechanical environment through a mechanism he calls the “mechanostat” (Frost 1987). This model proposes that increased mechanical usage causes reduced bone remodeling and increased modeling, with the end result of increased bone density. Conversely, decreased mechanical usage results in bone loss by enhancing remodeling and inhibiting modeling. If the bone turnover dynamics seen in orthodontic treatment are viewed as solely a remodeling type of change, then these findings would seem to contradict the mechanostat principle, because the modeling seen in controls becomes more like remodeling. However, the relationship between mechanical usage and orthodontic tooth movement remains unclear, and the degree to which the bone turnover changes reported here represent modeling, remodeling, damage/repair, or combinations also requires further elucidation. There are several noteworthy issues with regard to the bone turnover at sites of pressure and tension. In control animals, the observations of bone resorption on the distal alveolar surface and the failure to demonstrate any peak in its activity are consistent with an earlier report of socket remodeling at that location facilitating distal drift (Vignery et al. 1980). The application of a mesially-directed orthodontic force rapidly alters this pattern, resulting in the stimulation of a wave of resorption on both mesial and distal surfaces, more pronounced on the mesial . In orthodontically-treated tissues, bone formation is characterized by a stimulation beginning between day 5 and day 7. These changes also occur on both mesial and distal alveolar bone surfaces. Viewed along with the observation that an early resorption peak can be detected on both surfaces, and that these changes are demonstrable in the total tissue as well, the conclusion that both pressure and tension signals stimulate similar events in the alveolar bone seems reasonable. Differences between the two surfaces in the controls occurred as a result of normal distal drift and possibly animal manipulations required for appliance placement (Tran Van et al. 1982). In order to better appreciate the relative amounts of bone resorption and formation occurring at sites of pressure and tension, osteoblast and osteoclast surface perimeters are plotted over time after adjusting for site-specific control values (Fig. 3). Although an early wave of resorption is followed by
G. J. King et al.: Bone in orthodontic
408
a wave of formation, the relative amounts compared to control values at these sites are markedly different. Bone accrues at tension sites due to an inhibition of resorption and a stimulation of formation, with the most marked changes occurring late in the tooth movement. At pressure sites, bone is lost due to the stimulation of resorption and ‘inhibition of formation occurring early in the tooth movement. These data suggest that the normal degree of coupling between bone formation and resorption is disturbed in orthodontic tooth movement, fairly early in pressure sites and later in tension, and that these imbalances facilitate tooth relocation by removing bone ahead of the tooth and depositing it behind. Although we are aware of other instances, such as postmenopausal osteoporosis (for review see Frost 1985), in which quantitative uncoupling in haversian remodeling is thought to occur, it is premature to speculate whether the bone changes reported here have a similar mechanism. Our data further suggest that enhancement of bone formation in orthodontic treatment occurs by increases in the numbers of active bone-forming cells. Whether significant increases in their functioning also occurred is more equivocal. Significant late changes in osteoid perimeter may only reflect increased numbers of osteoblasts, since no changes in osteoid thickness or bone formation rate were detected. Tritiated thymidine studies have shown that orthodontic treatment does enhance cell division in the PDL, and that the primary site toward which these newly divided cells migrate is the alveolar bone (Smith et al. 1980; Roberts & Chase 1981). The mineralization rates and bone formation rates reported in both groups tend to be lower than those previously observed (Vignery et al. 1980). It is likely that this discrepancy is due to age differences between the animals used. The animals in the current study weighed twice as much as those in the former and were considered to be very slowly growing. Age differences in bone turnover parameters have been shown to be of significance (Reid et al. 1987). It is currently unknown whether the bone turnover patterns described here are sensitive to the magnitude of mechanical force or the manner by which it is applied. Studies on the adaptive remodeling of bone to various mechanical stresses do indicate that the process is extremely sensitive to alterations in both the magnitude and distribution of the strain generated within the bone (Rubin & Lanyon 1987). The nature of the feedback mechanism that regulates alterations in bone turnover in response to orthodontic treatment is likely to be complicated. Bone turnover remains in balance as the maxillary molars drift distally. Such a pattern could be genetically programmed (Rubin & Lanyon 1987) or mediated by a low-level net distally-directed strain. When the mesially-directed orthodontic stress is applied, a new strain environment is likely to be established in response to tooth displacement. In addition to the strain itself, which can directly affect bone (Hasagawa et al. 1985), numerous other new conditions with the potential for altering bone turnover also can occur. Matrixbound biologically active molecules like TGF, could be released either directly by tissue strain or by the early resorptive phase of the bone turnover (Seyedin et al. 1985). Other cytokines (Davidovitch et al. 1988) or prostaglandins (Somjen et al. 1980), which could further alter the bone turnover pattern, are also secreted by cells in the area. In addition, fluid flow could be altered in the bone (Kufahl et al. 1990), which could be reflected in fluid sheer stresses stimulating the production of osteoblastic cyclic AMP (Reich et al. 1990) and streaming potentials (Co&ran et al. 1989). These changes could also help to mediate the response to the altered environment caused by orthodontic treatment. Finally, bone proteoglycans have been shown to change orientation in response to stress (Skerry
tooth movement
et al. 1988). These molecular changes also have the potential to mediate alterations of cell behavior (Culp et al. 1986). In conclusion, this histomorphometric study demonstrated that bone turnover in the paradental alveolar bone of orthodontically-treated teeth is not balanced in the short term, but instead is characterized by tandem periods of activation, resorption, reversal, and formation. Furthermore, the study demonstrated that these bone turnover dynamics occur at sites of both pressure and tension, and that bone accretion or removal is controlled by the relative amounts of bone formation and resorption occurring at each site. Acknowledgments:
This
work
was
supported
by National
Institute of
Dental Research Award DE08659 from the National Institutes of Health. The authors wish to thank Ms. Liz McCoy for her technical assistance
and Ms. Kathy Cannon for typing the manuscript.
References Baron, R. A.; Siverglate, R.; ?ross. R.: Vignery, A. Dynamic histomorphometry of trabecular bone remodeling in the rat vertebrae. 1981. Baron, R. A.; Vignery, A.; Neff, L.; Siverglate, A.; Santa Marie, A. Bone histomorphometry: Techniques and interpretation. Reeker, R. R.. ed. CRC Press, Boca Raton 13-35; 1983. Bridges, T.; King, G.; Mohammed, A. The effect of age on tooth movement and mineral density in the alveolar tissues of the rat. Am. .I. Orthod. 93: 245-250; 1988. Burr, D. B.; Martin, R. B.; Schaffer, M. B.; Radin, E. L. Bone remodeling in response to in viva fatigue microdamage. J. Biomech. l&189-200; 1985. Cocbran, G. V. B.; Dell, D. G.; Palmieri. V. R.; Johnson, M. W.: Otter. M. W.; Kadaba, M. P. An improved design of electrodes for measurement of streaming potentials on wet bone in vitro and in viva. J. Biomech. 22:745750; 1989. Gulp, L. A.: Laterra, J.; Lark. M. W.; Beyth. R. J.; Tobey, S. T. Heparan sulphate proteoglycan as mediator of some adhesive responses and cytoskeleta1 reorganization of cells on fibronectin matrices: Independent versus cooperative function. Funcrion of the proteoglycans. Chichester: John Wiley and Sons; 1986. Davidovitch. Z.: Nicolay, 0.: Ngan, P. W.: Shaofeld, J. L. Neurotransmitters. cytokines, and the control of alveolar bone remodeling in orthodontics. Dent.
Clin.
North
Am.
32:411435;
1988.
Frost. H. M. The skeletal intermediary organization: A synthesis. Peck, W. A., ed. Bone and mineral research. Vol. 3. New York: Elsevier Science Publishers; 1985; 49-107. Hasagawa, S.; Sate, S.; Saito, S.: Suzuki, Y.; Brunette. D. M. Mechanical stretching increases the number of cultured bone cells synthesizing DNA and alters their pattern of protein synthesis. C&if. Tin. Inr. 37:431-436; 1985. King, G. J.: Keeling, S. D.: McCoy. E. A.; Ward, T. H. Measuring dental drift and orthodontic tooth movement in response to various initial forces in adult rats. Am. J. Orthod. Dent. Orrhop. 99:45M65; 1991. Kufahl, R. H.; Saha, S. A theoretical model for stress-generated fluid flow in the canaliculi-lacunae network in bone tissue. J. Biomech. 23:171-80: 1990. Lanyon, L. E. Functional strain as a determinant for bone remodeling. C&if. Tiss. Int. 36 (Suppl. 1) S56-661; 1984. Marie, P. J.; Hott, M.; Perheentupa, J. Effects of epidermal growth factor on bone formation and resorption in viva. Physio. Endoc. Metab. 258:E275281:
1990.
Markostamou, M. K.; Baron, R. Etude quantitative de I’ostioclasie su la par01 alv6olaire en orthodontic exptrimentale cher le rat. Orthod. Fr. 44~245-258: 1973.
Parfitt, A. M. The cellular basis of bone remodeling: The quick turn concept reexamined in light of recent advances in the cell biology of bone. C&if. Tin.
Int.
36:537-545;
1984.
Parfitt, M. A.: Drezner, M. K.; Glorieux, F. H.; Kanis, J. A.; Malluche, H.; Meunier, P. J.; Ott, S. M.: Reeker, R. R. Bone histomorphometry: Standardization of nomenclature, symbols, and units. J. Bone Min. Res. 2:595610:
1987.
Perren, S. M.: Cordey, J.; Ralm. B. A.; Gamier. E.: Schneider. E. Early temporary porosis of bone induced by internal fixation implants. A reaction to necrosis, not stress protection? Clin. Orthop. Rel Res. 232:139-51: 1988. Reich. K. M.: Gay, C. V.; Fraozos. J. A. Fluid shear stress as a mediator of
G. J. King et al.: Bone in orthodontic tooth movement osteoblast cyclic adenosine monophosphate production. J. Cell Physiol. 143: 1004: 1990. Reid, S. A.; Boyd, A. Changes in mineral density distribution in human bone with age: Image analysis using backscattered electrons in the SEM. J. Bone Min. Res. 2:13-22; 1987. Reitan, K. Clinical and histological observations on tooth movement during and after orthodontic treatment. Am. J. Orthod. 53:721-745; 1967. Roberts. W. E.; Garetto. L. P.; DeCastro, R. A. Remodeling of devitalized bone threatens periosteal margin integrity of endosseous titanium implants with threaded or smooth surfaces: Indications for provisional loading and axially directed occlusion. J. Ind. DEW. Assoc. 68:19-24; 1989. Roberts, W. E.: Chase, D. C. Kinetics of cell proliferation and migration associated with orthodontically-induced osteogenesis. J. Dent. Res. 60:174-181; 1981. Rubin. C. T.; Lanyon, L. E. Osteoregulatory nature of mechanical stimuli: Function as a determinant for adaptive remodeling in bone. J. Orfhop. Res. 5:300-310; 1987. Rygh. P. Ultrastructural changes in pressure zones of human periodontium incident to orthodontic movement. Acta Odonr. Scond. 31:109-122; 1973. Rygh, P. Ultrastructural changes in pressure zones of rat molar periodontium incident to orthodontic movement. Acm Odont. Stand. 30575-593; 1972. Rygh. P. Ultrastructural changes in tension zones of rat molar periodontium incident to orthodontic movement. Am. J. Orthod. 70:269-281; 1976. Rygh, P.: Selvig, K. Erythrocytic crystallization in rat molar periodontium incident to tooth movement. Stand. J. Dent. Res. 81:62-73; 1973. Sadowsky, C.: BeGole, E. Long-term effects of orthodontic treatment on periodontal health. Am. J. Orthod. 80:156-172; 1981. Saffar, J. L.; Klapisz-Wolikow, Changes in mast cell number during the activation phase of an induced synchronized remodeling sequence in the rat. Bone 11:369-372: 1990.
409 Scheffe, H. A method for judging all contrasts in the analysis of variance. Biometricka 40~87-104: 1953. Seyedin. S. M.; Thomas, T. C.; Thompson, A. Y.; Rosen, D. M.; Pie& K. A. Purification and characterization of two cartilage-inducing factors from bcvine demineralized bone. Proc. N&l. Acad. Sci. USA 82:2267-2271: 1985. Skeny. T. M.; Bitensky. L.; Chayen, J.; Lanyon, L. E. Loading-related reorientation of bone proteoglycan in viva. Strain memory in bone tissue? J. Or‘hop. Res. 6547-51; 1988. Smith, R. K.; Roberts, W. E. Cell kinetics of the initial response to orthodontitally-induced osteogenesis in rat molar periodontal ligament. C&f. En. fnr. 30:51-56: 1980. Somjen, D.: Bindennan, I.; Bergen. E.; Harrell. A. Bone modeling induced by physical stress is prostaglandin mediated. Biochem. Biophys. Acta. 91:627: 1980. Storey. E. The nature of tooth movement. Am. J. Orthod. 63:292-314; 1973. Tran Van, P.; Vignery, A.: Baron, R. Cellular kinetics of the bone remodeling sequence in the rat. Anar. Rec. 202:445-451; 1982. Vignery, A.; Baron, R. Effects of parathyroid hormone on the osteoclastic pool, bone resorption, and formation in rat alveolar bone. C&if. Tim. Res. 26: 23-28; 1978. Vignery, A.; Baron, R. Dynamic histomorphometry of alveolar bone remodeling in the adult rat. Anar. Rec. 196~191-200; 1980.
Date Received: October 25, 1990 Dare Revised: March 22. 1991 Date Accepted: June 6. 1991
Copyright
8756-3282191 $3.00 + .OO 0 1991 Pergamon Press plc
Histomorphometric Study of Alveolar Bone Turnover in Orthodontic Tooth Movement G. J. KING,’
S. D. KEELING’
and T. J. WRONSKI’
‘Department of Orthodontics, University of Florida, College of Dentistry, Gainesville, FL, U.S.A. ‘Department of Physiological Sciences, University of Florida, College of Veterinary Medicine, Gainesville, FL, U.S.A. Address for correspondence and reprints: Gregory J. King, Box J-444, JHMHC,
Abstract Selected histomorphometric parameters were measured in alveolar bone adjacent to rat molars treated with a 40gram tipping force designed to tip the molar mesially and with a sham procedure. Undecalcified parasagittal sections of teeth and surrounding tissues were prepared for static and dynamic hlstomorphometry at 1, 3, 5, 7, 10, and 14 days. Inltlal tooth displacement was assessed by measuring differences between groups in the widths of the day-l perlodontal ligaments (PDL) at various vertical locations and correlating these using linear regression analysis. All parameters were measured in the alveolar bone adjacent to four quadrants around the teeth (mesial-distal; occlusalapical). Means and standard errors for each parameter ln each group were calculated and compared for time- and group-specific differences using ANOVA and pairwlse comparisons with Scheffe’s multiple comparison tests. In shamtreated rats, bone resorption predominated on the distal alveolar surface, but signlficant surface-related differences between mesial and distal surfaces ln bone formation were not demonstrated. Time-specific effects in bone resorption were not evident on either surface in the shams. These findings suggested that molar distal drift in the sham rat is facilitated by resorption in the distal alveolar bone. PDL width changes in orthodontically-treated rats were greatest in the me&l occhtsal half of the root and decayed linearly toward the apex indicating that the greater initial tooth displacement occurred in the occhtsal half of the root surface. Histomorphometric parameters of alveolar bone turnover were also seen to be greater in these locations, suggesting that these processes were sensitive to the increased tooth displacement. In orthodontically-treated rata, the total paradental alveolar bone and both pressure (mesial) and tension (diital) surfaces experienced a wave of resorption that lasted for four days, was preceded by an activation period of two days, and was followed by a wave of formation that persisted until the end of the experiment (14 days). The relative balance between bone formation and resorption waves was location-specific, with formation predominant on the tension side and resorption predomlnant on the pressure. These fhulings are consistent with the conclusion that pressure and tension initiate similar bone turnover events, and that bone deposition or removal is controlled by the degree
Gainesville,
FI 32610, U.S.A.
of balance between the relative amounts of bone formation and resorption that occur at these sites. Key Words: Bone turnover-Tooth Formation -Orthodontic.
movement-Resorption-
Introduction The movement of teeth in the alveolar process requires bone turnover. An extensive literature exists demonstrating that orthodontic treatment has the capacity to alter dramatically a preexisting pattern of alveolar bone turnover, with bone resorption occurring at sites of pressure and formation in areas of tension (Reitan 1967; Storey 1973; Rygh 1972, 1976). However, there are no histomorphometric studies on how these bone turnover dynamics are expressed as functions of time in the total paradental alveolar bone or at locations adjacent to teeth with defined orthodontic displacements. Under routine conditions, orthodontic treatment results in no net long-term changes in the periodontal ligament (PDL), the amount of alveolar bone, or its nature (Sadowsky et al. 1981). This could occur if the amount of bone formation stimulated was simultaneously balanced by an equal amount of resorption. This level of balance has been shown to exist during distal drift of rat molars (Vignery et al. 1980). Alternatively, orthodontic stress may act to stimulate bone turnover events resulting in waves where the resorption and formation occur in tandem. Such a mechanism would predict an early transient bone loss and temporary widening of the PDL, followed by bone formation, which would return the tissue architecture to normal. This view is consistent with the observation that alveolar bone shows a transient osteopenia followed by recovery during orthodontic treatment (Bridges et al. 1988); that a wave of increased osteoclast numbers occurs during the fit few days of orthodontic treatment (Markostamou et al. 1973); that bone cells at sites of formation go through a synchronized remodeling cycle following the mechanical stimulus of removal of an opposing tooth (Tran Van et al. 1982); and that orthodontists report teeth are mobile and display a widened PDL on radiographs during active treatment.
G. J. King
402
et al.: Bone
This study investigated, histomorphometrically, the time course of changes in alveolar bone turnover at locations in the alveolar socket where the tooth displacement geometry was known during a precisely specified orthodontic treatment. Materials and Methods Seventy-two adult male Sprague-Dawley strain rats (89-94 days old) were purchased from Charles River Breeding Laboratories (Wilmington, MA). These were shipped by air freight and acclimatized for at least two weeks under experimental conditions, including housing in plastic cages, a diet of ground laboratory chow and distilled water ad libitum, and a standard 12-hour light/dark cycle. All animal manipulations, including sacrifices, were done at the same time of day. Weights were recorded and anesthesia was obtained using intra-muscular injections of ketamine (87 mg/kg) and xylazine (13 mg/kg). Modified orthodontic cleats were bonded bilaterally to the occlusal surfaces of all maxillary first molars as previously described (King et al. 1991), and the mandibular first and second molars were extracted. The animals were then allowed to recover for one week by monitoring wound healing and weight gain. For appliance activations, the animals were positioned in a head restrainer and orthodontic springs were placed as previously described (King et al. 1991). Briefly, one end of a length of closed coil spring was ligated to the molar cleat while the other was attached to a suspended weight. The anterior end of the coil was then bonded with autocuring methacrylate to the acid-etched lateral surface of the maxillary incisors, followed by removal of the weight and excess coil spring. This method assured a precise initial orthodontic force designed to tip the maxillary first molars to the mesial. The rats were randomly divided into two groups of equal size. One group received the orthodontic force at a level of 40
tooth
movement
Table I. Parameters obtained by histomorphometric
measurements
Parameter
Formula
Abbreviation
Osteoclast number (#/surface lgth.) Osteoclast surface (%) Eroded surface (%) Osteoblast surface (%) Osteoid surface (%) Osteoid thickness (mcm) Mineralizing surface (%)
N.Oc.
N.Oc./B.Pm
Oc.Pm. E.Pm. Ob.Pm. O.Pm. O.Th. M.S.
Mineral apposition rate (mcm/day) Bone formation rate (mcm3/mcm2/day) Reversal surface (%)
M.A.R.
Oc.Pm./B.Pm. E.Pm./B.Pm. Ob.Pm./B.Pm. O.Pm./B.Pm. O.Th./N. [dL Pm. + (sL Pm/2)]/ B.Pm. 1r.L. Th/ 1r.L.t.
B.F.R.
(M.A.R.
Rv.Pm.
E.Pm.
Nomenclature according Additional abbreviations double labeled perimeter, inter-label thickness, 1r.L.t: Fig. 1. Camera lucida drawing of the tissue under investigation. AB - alveolar bone; PDL - periodontal ligament; TR - tooth root; P - pulp chamber; MO - mesio-occlusal quadrant; DO - Disto-occlusal quadrant; MA - mesio-apical quadrant; DA - disto-apical quadrant. Representative non-overlapping fields for histomorphometric analysis at 400 x magnification are displayed as squares in the right DO quadrant.
in orthodontic
X M.S.)/B.Pm.
-
Oc.Pm.
to Parfitt et al. 1987. as follows: B.Pm: bone perimeter, dL Pm: sL Pm: single labeled perimeter, Ir.L.Th: inter-label time.
while the other was sham-treated, receiving all procedures except spring placement. The spring was excluded to allow normal distal drift of the molars. Six animals from each group were sacrificed at each of six timepoints (1, 3, 5, 7, 10, and 14 days). Previous studies have indicated that these orthodontic parameters and time intervals adequately produce and monitor tooth movement kinetics, which exhibit characteristic changes including instantaneous movement, a delay period and late tooth movement (King et al. 1991). All rats were injected intraperitoneally with tetracycline derivatives (15 mg/kg) on two separate occasions. Oxytetracycline hydrochloride (Terramycin; Pfizer, Inc., Brooklyn, New York) was administered five days prior to sacrifice. After a two-day period with no tetracycline administration, all rats were injected again with demeclocycline (Declomycin, Lederle Laboratories, Pearl River, New York) two days before sacrifice. This regimen resulted in the deposition of a double tetracycline label at bone surfaces that were actively mineralizing throughout the injection period (Baron et al. 1983). Animals were killed by decapitation at each of the time intervals. Maxillae were dissected into halves and fixed in 70% ethanol for 24 hours, followed by dehydration in increasing concentrations of ethanol and embedding undecalcified in methyl metacrylate (Baron et al. 1983). The embedded samples were sectioned parallel to the long axis of the first molar teeth at 4 pm thicknesses with an A0 AutocutIJung 1150 Microtome and mounted on 1% gelatinized slides. Alternate sections were stained according to the Von Kossa method with a tetrachrome counterstain. All surface-based parameters were quantified on these sections using a semi-automatic image analyzer (Bioquant, R & M Biometrics, Nashville, TN) coupled to a photomicroscope equipped with epifluorescence (Nikon Labophot). All tetracycline-derived parameters were measured on similarly mounted unstained sections cut at 6 p,m thicknesses. Paradental and inter-radicular alveolar bone immediately adjacent to the periodontal ligament of disto-buccal and middle-buccal maxillary first molar roots was sampled as follows: Parasagittal sections were selected by choosing those demonstrating a radicular pulp to the apical third of the root. The tooth roots, pulp chambers, and alveolar bone surfaces were traced at low magnification using the camera lucida (Fig. 1). A line was drawn on this tracing bisecting the root longitudinally. This line was then divided by one drawn perpendicular to it at the vertical midpoint of the tooth socket. These two lines grams,
G. J. King et al.: Bone in orthodontic
403
tooth movement
Initial root displacement in the sagittal plane
R”2 I 0.898
-20: 0
-
. -6 * . I . 20 40
Difference
.
I . 60
.
I . 80
.
1 100
in mesial PDL width (pm)
Fig. 2. Linear regression analysis between PDL width differences (control minus treated) at 20 equally spaced points vertically along the root.
were used to define equal-sized quadrants around the root (i.e., mesio-occlusal, mesio-apical, disto-occlusal, and disto-apical). Multiple non-overlapping fields at 400 X were then sampled from each quadrant. The bone parameters that were measured are shown in Table I and are defined according to Pa&t et al. (1987). One section was measured for each hemimaxilla, thereby preventing the sampling of the same area twice. The means and standard errors of each parameter were calculated for each group, by location (i.e., total section, side, quadrant) and timepoint. An analysis of variance (ANOVA) was performed to examine differences among force groups at each timepoint and across times within each group. Pairwise (Scheffe, 1953) comparisons were performed between groups, locations, and days when ANOVA indicated that significant differences existed (p < 0.05). Since bone remodeling does not commence until after day 1, alterations in PDL width were considered to adequately represent the initial tooth displacement. This was assessed on the mesial PDL space by measuring the difference in width between control and treated sections from day 1 at each of 20 equally spaced locations vertically along the root from the crestal bone to the apex. A linear regression line was then used to depict the initial tooth displacement and its center of rotation.
RMlftS Initial tooth displacement There was a highly significant correlation (R2 = 0.90, p = 0.0001) between vertical location on the root and mean difference in PDL width between control and treatment groups at day 1 (Fig. 2). The greatest reduction occurred at the alveolar
crest (i.e., occlusal) and the least at the apex. This root displacement is consistent with the teeth being tipped toward the mesial with a center of rotation at or near the root apex. Using traditional orthodontic terminology, the mesial-occlusal quadrant is in “pressure” and the disto-occlusal is in “tension.” The observations obtained from total alveolar sections, the mesial occlusal quadrant, and the distal occlusal quadrant are reported below. The occlusal quadrants are displayed because they represented the sites of greater tooth displacement.
Resorptive parameters Data on the resorptive parameters are presented in Table II. For the total alveolar bone section, ANOVA indicated no differences across time in the control group. However, significant differences across time existed in the treatment group for osteoclast perimeter, osteoclast numbers, and erosion perimeter, where Scheffe comparisons indicated that day 5 represented a consistent peak. Thus, the alveolar bone around the orthodontically-treated teeth experienced a period, lasting from l-3 days, when there were no increases in these parameters. This was then followed by a wave of increase that reversed between 5 and 7 days. A similar trend also exists in reversal perimeter but this did not achieve statistical significance. When comparisons between quadrants (i.e., mesial-occlusal, distal-occlusal) are considered, the control group had consistently higher resorptive parameters on the distal alveolar bone surface than the mesial at each timepoint, but there were no significant differences across time at either location. In the orthodontically-treated group, ANOVA indicated that significant differences in osteoclast perimeter, osteoclast numbers, and erosion perimeter across time existed on the mesial-
0
2
4
6
8
10 12 14
-50
-40
-30
30
40
50
0
E
8
S
t
m
3
1
0
2
-30,1-50
-25 -1
-20
-15
-10
-5
0
5
10
3 g r
15
-a-
20
25
4
6
:
: :
8
, _--
-40
t
t 40
-30
-20
-30
1:
10 12 14
4’ 4’
13 ,.
Tension side: distal occlusal quadrant
1
-25-
0
2
-30 I-50
\ \, .
4
/' f
6
8
10 12 14
- -40
Pressure side: mesial occlusal quadrant
Fig.3. Bone turnover as result of orthodontic force, depicted by changes in osteoclast and osteoblast perimeter minus control values over time. Left panel - Total pamdental tissue. Differences across time existed in % osteoclast surface (p < .OOOl), manifested by a peak at days 3 and 5 (versus days I, 10, and 14; p < .05). Differences across time existed in % osteoblast surface @ < .OOOl), characterized by a peak at days 10 and 14 (versus days 3 and 5, p < .05). Middle panel - Tension side: No differences were detected in % osteoclast surface over time @ > .05). Differences across time existed in % osteoblast surface @ < .OOl), manifested by a peak at day 14 (versus days 1, 3, and 5: p < .05). Right panel - Pressure side: Differences across time were detected in % osteoclast surface @ < .OOl), manifested by a peak at day 5 (versus days I and 14; p < .05). Differences across time were also detected in % osteoblast surface @ < ,001). where a peak occurred at day 14 (versus days 3 and 5; p < .05).
-30
-25
-20
-15
20
25
30
(40 gram dafa adjusted tar moan conlrol valuoa)
Total paradental tissue
G. J. King et al.: Bone in orthodontic
tooth movement
405
Table II. Mean and standard error of the mean for resorptive parameters by group and day Control group
Day Oc.Pm 3 5 10 14
Mesial Occl. 0.0 4.7 3.6 0.0 1.8 1.0
2
* ‘5 i k NS
Distal Occl.
0.0 2.8 2.3 0.0 1.8 0.5
14.8 20.9 9.0 11.1 15.2 17.8
k 2.9 f 4.8 f 1.1 k 1.3 k 4.9 ~fr 3.4 NS
Treatment group Total
Mesial Occl.
Distal Occl. 11.2 11.3 12.5 3.8 3.9 1.6
8.1 7.5 6.5 6.8 1.3 10.7
k 2 k 2 k ‘NS
1.3 1.1 0.7 0.9 1.7 2.2
3.8 18.4 25.2 11.7 9.1 4.8 ***,
k 0.9 2 4.2 k 2.3 2 5.1 2 3.3 k 2.0 Note 1
2 2 2 2 k k NS
Between Group ANOVA’ Total
3.1 5.7 1.6 1.8 3.4 1.6
6.6 16.5 16.2 6.9 5.2 3.7 ***,
k 1.1 k 2.4 + 1.7 L 2.0 2 2.7 k 1.7 Note 2
Mesial
Distal
Occl
Occl.
Total
* *** *
NS NS NS *
NS *
NS NS
NS **
** NS NS *
**
0.3 1.5 1.9 0.2 0.5 0.9
2 2 t t ? k NS
0.3 1.0 0.8 0.2 0.5 0.4
5.4 7.3 4.4 5.3 5.3 6.4
f k t + + + NS
1.3 1.0 1.1 1.0 1.9 1.0
3.3 2.8 2.7 3.0 2.5 4.6
” 2 2 2 k k NS
0.8 0.5 0.4 0.6 0.6 0.9
1.7 6.6 12.4 3.5 3.1 1.9 ***,
2 0.4 2 1.6 2 1.4 k 1.3 2 1.1 k 1.1 Note 3
3.8 k 1.1 2.3 k 1.2 5.4 2 1.1 1.3 k 0.6 1.1 k 0.7 0.8 2 0.6 *, Note 0
2.1 4.9 6.7 2.1 1.9 1.4 ***,
k 0.3 + 0.6 2 0.6 + 0.6 k 0.8 ” 0.7 Note 3
* * *** *
NS *
NS *
NS **
NS NS
NS **
** NS NS *
+ k + 2 2 i NS
10.1 7.7 6.6 3.0 7.3 4.1
50.0 48.7 41.9 56.1 49.6 59.9
k -c + 2 k ?I NS
5.3 2.2 7.5 4.1 8.1 6.8
27.6 25.1 26.5 30.6 26.1 38.7
2 k 2 k k 2 NS
2.3 4.8 2.1 3.5 2.8 6.8
20.4 56.9 65.0 41.6 35.7 40.4 **,
+ 1.3 + 5.2 + 12.2 + 11.2 k 7.2 k 5.7 Note 4
54.9 k 7.8 44.2 i- 8.6 57.3 k 3.3 37.7 k 5.4 24.9 2 11.0 28.3 + 10.3 *, Note 0
31.8 51.2 60.5 43.0 34.8 35.0 **,
k 5.4 L 4.0 k 3.6 k 5.1 k 8.6 + 6.3 Note 5
NS *** *** * * *
NS NS NS *
14
10.1 12.1 16.2 4.9 8.7 19.0
NS *
NS ** *** NS NS NS
5 I 10 14
10.1 7.4 12.6 4.9 6.9 18.0
+ k + k k k NS
10.1 6.3 7.2 3.0 5.5 3.9
35.2 27.8 32.9 45.1 34.4 42.1
k 2 2 2 -t2 NS
4.5 6.8 6.6 4.3 6.6 4.2
19.6 17.5 20.0 23.9 18.8 27.9
2 * 2 2 2 2 NS
2.6 5.0 2.1 3.2 2.1 5.0
16.6 38.5 39.9 30.0 26.6 35.6
k k k f 2 k NS
43.1 32.9 44.8 34.0 21.1 26.7
25.2 34.6 44.3 36.1 29.6 31.3
+ k 5 k k t NS
NS * * * * *
NS NS NS NS NS NS
NS * ** NS NS NS
N.Oc. 5 7 10
14 E.Pm. 5 10
Rv.Pm
6.9 7.1 4.9 9.7 6.0 5.1
+ + k * 2 k NS
9.0 4.8 3.7 5.1 9.6 9.4
4.8 3.0 4.7 5.3 7.5 5.4
‘Levels of significance for ANOVA between groups at each day for each location: *p < 0.05; **p < 0.01; ***p < 0.001. Levels of significance for ANOVA among days within each group by location: *p < 0.05; **p < 0.01; ***p < 0.001. Notes on results of Scheffe comparisons between days where p < 0.05: (0) None; (1) 5 vs 1 & 14; (2) 3 vs. 1, 10 & 14; 5 vs. 10 & 14; (3) 5 vs 1, 7, 10 & 14; (4) 1 vs 3 & 5; (5) 1 vs 5
occlusal surface (pressure side), also manifested by a peak at day 5. Less dramatic trends also occurred on the distal surface (tension side), which followed the same timing as those seen on the pressure side. When comparisons at each location were made between treatment and control groups at each time interval, ANOVA indicated that the values for the treatment group were greater than those for the control group at days 3 and 5 for all resorptive parameters on total sections and on the pressure side. Numerous time intervals showed treatment effects for all resorptive parameters on the mesial (pressure) side; far fewer differences were demonstrable on the distal (tension) side. Formative parameters Data on the bone formation parameters are presented in Table III. For the total alveolar bone section, ANOVA indicated that there were no significant differences in shams over time in either OPm., O.Th., or M.A.R.; but time-specific changes in M.S. and B.F.R. were detectable, characterized by a peak at three days and a depression at seven. In the orthodonticallytreated group, ANOVA indicated that significant differrices existed across time @ < 0.001) in osteoblast perimeter, manifested by values at days 3 and 5 being significantly lower
than those at days 10 and 14 @ < 0.01). Thus a significant peak in osteoblast numbers occurred between days 5 and 10, remaining elevated at the end of the observational period on day 14. These changes were also reflected in osteoid (p < 0.01) perimeter. In the treated group, ANOVA indicated that significant differences in percentage of osteoblasts occurred across time on both the mesial (pressure) an distal (tension) sides. On the mesial side, these differences were manifested by values at day 14 being greater than those at days 3 and 5 0, < 0.05). In the treated group, changes in the osteoid perimeter reflected those seen in osteoblast numbers characterized by a significant peak occurring on the total section by day 10 (p < 0.01). Similar trends are also evident on the mesial and distal. No differences were evident in bone formation rate or osteoid thickness. There was a significant elevation in mineralizing surface on the mesial at day 1, which is consistent with this surface being primarily formative prior to orthodontic force application and with one day being an insufficient period of time for removal of preincorporated tetracycline labels. There were no significant time-specific differences in mineral apposition rate. Changes in osteoblast and osteoclast perimeters are shown graphically in Fig. 3 by subtracting appropriate control values
G. J. King et al.: Bone in orthodontic tooth movement
406
Table III. Mean and standard error of the mean for the formation
Day
0. Pm.
0. Th.
M. S.
M.A.R.
B. F. R.
by group and day
Treatment Group
Control Group
Parameter
Ob. Pm.
parameters
Mesial
Distal
Occl.
Occl.
Total
1 3 5 7 10 14
27.8 43.1 64.3 80.8 80.6 62.3
k 11.9 k 16.3 + 13.6 2 6.4 + 8.3 L 9.1 *, Note 0
19.1 ? 8.2 26.0 ? 12.5 38.9 f 10.2 54.5 * 3.1 61.1 ? 5.3 43.2 c 7.5 **, Note 1
32.2 40.7 49.2 62.9 67.2 48.0
2 7.6 k 15.1 k 10.8 k 3.0 f 4.7 + 5.5 *, Note 0
1 3 5 7 10 14
16.1 30.6 43.3 61.7 59.2 40.3
2 10.3 k 13.7 f 11.0 k 9.9 k 10.4 t 8.9 *, Note 0
10.9 11.7 21.6 26.9 33.1 23.1
22.6 26.1 31.5 35.0 43.6 26.0
k 6.0 k 10.1 t 8.2 k 4.4 2 7.1 2 6.1 NS
1 3 5 7 10 14
0.1 0.4 0.4 0.4 0.4 0.5
1 3 5 7 10 14
42.7 64.5 37.5 3.7 22.2 32.9
2 t + 2 k +
0.1 0.1 0.0 0.0 0.1 0.2 NS
2 14.6 k 11.4 2 13.8 + 1.6 + 11.1 k 6.1 *, Note 6
1 3 5 7 10 14
3.92 4.8 4.1 2.6 2.5 2.5
+ 2 f f +
1 3 5 7 10 14
1.1 2.9 1.6 0.2 1.0 1.3
2 k k + + 2
0.4 0.4 0.3 0.4 0.5 0.4 1.9 37.5 31.6 13.7 9.6 15.5
? 2.0 t 7.0 2 10.0 % 3.1 + 8.6 % 8.1 NS * 2 2 + 2 *
0.1 0.1 0.1 0.0 0.1 0.0 NS
z? 1.9 + 11.3 f 11.8 + 13.7 * 3.4 + 7.8 NS
0.3 0.4 0.4 0.4 0.4 0.4
+ 2 + k ? L
0.0 0.1 0.0 0.0 0.1 0.1 NS
20.8 2 6.3 56.6 + 8.8 36.0 + 10.2 9.2 2 6.1 21.3 + 7.8 22.2 f 2.6 **, Note 6
Mesial
Distal
Occl.
Occl.
Between Group ANOVA’ Mesial Total
Occl.
Distal Occl.
*
29.6 t 9.1 17.7 k 2.2 22.1 2 2.7 45.2 i 11.8 51.6 + 12.8 68.0 k 4.6 ***, Note 2
45.2 k 7.2 42.7 + 9.6 39.5 + 5.5 71.8 f 6.5 73.4 2 2.7 82.1 k 4.6 ***, Note 3
43.7 2 3.9 33.0 2 5.1 35.6 k 4.2 56.4 k 7.3 66.9 k 5.3 70.0 k 4.9 ***, Note 4
NS NS * *
NS NS *
NS NS
NS **
14.6 9.6 15.0 20.9 28.2 36.2
21.7 23.2 21.9 37.0 45.2 48.6
k 6.3 + 6.4 ‘- 4.0 + 5.2 k 3.8 + 15.2 *, Note 0
22.5 + 3.1 16.5 k 3.0 18.1 k 3.2 27.8 + 4.1 39.3 k 5.2 37.0 k 7.6 **, Note 5
NS
NS
NS * **
NS
0.4 0.7 0.4 0.4 0.5 0.6
_f 0.0 5 0.2 5 0.0 k 0.0 + 0.0 ‘- 0.1 NS
0.5 0.5 0.5 0.5 0.6 0.5
t 3.2 i- 2.1 + 3.4 + 6.0 L 8.7 2 6.2 *, Note 0 + * k + + 2
0.1 0.0 0.1 0.0 0.1 0.0 NS
70.9 k 10.0 6.6 + 4.0 24.1 2 7.3 11.8 k 8.1 16.7 f 8.6 28.6 + 8.0 ***, Note 7
21.9 k 11.3 48.6 + 4.9 20.3 k 7.0 64.0 5 8.6 44.0 2 10.4 14.5 k 5.5 **, Note 8
1.0 0.5 1.4 1.2 0.9 0.6 NS
3.0 4.4 4.6 0.5 3.3 1.7
+ + + + + 2
2.3 0.8 0.4 0.5 1.2 0.5 NS
3.4 k 0.6 9.5 + 4.7 3.8 2 0.7 1.3 k 0.5 2.7 2 0.6 2.2 + 0.2 *, Note 0
6.7 2.7 3.1 2.0 2.6 2.2
k 1.1 k 1.2 k 0.8 k 1.3 k 1.1 k 1.0 *, Note 0
3.5 5.5 3.1 9.9 3.3 5.9
0.6 1.0 0.6 0.1 0.5 0.6 NS
0.3 2.4 1.8 0.2 1.7 0.4
f 0.3 ? 1.4 ? 0.9 & 0.2 * 1.1 + 0.3 NS
0.7 f 0.4 3.1 k 1.0 1.7 2 0.5 0.2 f 0.1 1.2 + 0.3 0.9 2 0.2 **, Note 9
4.6 + 1.1 0.4 2 0.2 1.1 k 0.5 1.7 + 1.6 1.1 + 0.9 1.4 f 0.7 *, Note 0
1.1 5.0 0.9 8.2 2.1 1.8
+ 1.0 2 1.2 2 0.7 + 2.2 lr. 1.2 + 1.4 *, Note 0 + 0.8 k 1.5 k 0.2 + 3.0 + 0.9 k 1.1 *, Note 0
0.4 0.5 0.4 0.4 0.6 0.5
k 0.0 2 0.1 f 0.0 k 0.0 2 0.0 2 0.1 *, Note 0
NS NS * NS NS NS NS NS
Total NS NS NS NS NS *
NS NS NS NS
NS NS NS NS NS NS
NS NS NS NS NS NS
NS NS NS NS NS
**
* *
45.4 31.8 33.0 24.7 28.5 24.0
2 + k k + 2
7.5 4.2 4.7 4.9 5.2 6.3 NS
NS ***
4.8 5.0 4.1 5.5 2.6 3.5
5 k k ” 2 k
0.5 0.5 0.9 0.8 0.8 0.7 NS
NS NS NS NS NS NS
NS NS NS **
NS
NS *
NS NS
2.6 3.2 2.5 3.2 1.4 1.4
+ 0.6 ” 0.7 + 0.7 ‘- 0.9 k 0.7 If: 0.5 NS
* *
NS NS NS *
* NS NS *
NS NS
NS NS
NS NS NS NS
NS NS NS NS
NS NS NS * * NS
NS NS NS NS
NS NS **
‘Levels of significance for ANOVA between groups at each day for each location: *p < 0.05; **p < 0.01; ***p < 0.001. Levels of significance for ANOVA among days within group by location: *p < 0.05; **p < 0.01; ***p < 0.001. Notes on results of Scheffe comparisons between days where p < 0.05: (0) None; (1) 1 vs 10; (2) 14 vs 3 & 5; (3) 14 vs 1, 3, & 5; 5 vs. 10; (4) 3 & 5 vs. 10 & 14; (5) 3 vs 10; (6) 3 vs 7; (7) 1 vs 3, 5, 7, 10, & 14; (8) 7 vs 14; (9) 3 vs 1 & 7.
to depict the predominant bone turnover dynamics in the orthodonticahy treated tissues. To facilitate comparisons among locations, the left vertical axes are set to the same scale, as are the right; the zero points of each vertical axis superimpose. In the total tissue (left panel), an early increase in osteoclast perimeter is evident beginning between days 1 and 3 and reversing between days 5 and 7. There is a tendency for this parameter to be inhibited in the later timepoints. Osteoblast surface is already inhibited on day 1. This inhibition is further increased at days 3 and 5, but reversed to stimulation by days 7, 10, and 14.
The middle panel demonstrates changes in the paradental quadrant where the tooth is initially most displaced away from the socket wall (i.e., tension side). Neither osteoblast nor osteoclast perimeter differ appreciably from control until an interval between days 5 and 7, at which time marked stimulation of osteoblast and inhibition of osteoclast perimeter occur. The right panel represents similar data from the pressure quadrant. Osteoclast perimeter shows a marked stimulation beginning between days 1 and 5 and remaining elevated until an interval between days 10 and 14, with a peak period around day 5. Osteoblast perimeter is already significantly inhibited
G. J. King et al.: Bone in orthodontic
tooth movement
on day 1, and remains depressed until late in the tooth movement, when it returns to levels slightly above control.
DlliOIl
The orthodontic mechanics used in this study produce an initial tooth displacement characterized by mesial tipping around a center of rotation located at the root apex. Sites where the tooth root is displaced toward the socket wall have been categorized as being in pressure, and those where the root is moved away from the socket wall as tension (Reitan 1967; Storey 1973). Based on these definitions, the mesial surface would be categorized as a pressure site and the distal as tension in this model. The histomorphometric changes reported here are consistent with numerous reports indicating that sites of orthodontic pressure are primarily bone resorptive, while those in tension are formative (Reitan et al. 1967). In spite of this association, interpretation of tooth displacement data as representing specific strains in the tissues should be approached with caution, as data on alveolar bone strains associated with specific orthodontic tooth displacements are unavailable. For convenience and consistency with the tooth movement literature, we have adopted the ubiquitous pressure/ tension terminology, realizing that it does not adequately describe the biomechanical environment initiated in the tissues by these appliances. The bone turnover dynamics in the total paradental tissues observed in this experiment are inconsistent with the hypothesis that orthodontic tooth movement occurs by two balanced bone modeling processes (i.e., resorption under pressure and formation in tension). Over the short term, bone resorption and formation are not balanced, but instead pass through a period (l-3 days) of little change, a period that is predominantly resorptive (up to days 5-7), and a phase that is primarily formative beginning between days 5-7 and lasting until the end of the experiment. This finding is consistent with the earlier observation of a transient osteopenia in orthodontically-treated paradental alveolar bone (Bridges et al. 1988), and points to a distinct difference from distal molar drift where bone turnover does remain in balance (Vignery et al. 1980). The finding that the activation of bone resorption required two days agrees very well with that for molar migration during tooth egression following removal of an opposing tooth (Tran Van et al. 1982). However, the resorption phase was slightly longer in this study than in other reports using adult rodents (i.e., 1.5-2.8 days; Vignery et al. 1980; Baron et al. 1981; Vignery et al. 1978; Tran Van et al. 1982). The estimate from this study of a bone formation phase longer than 8 days agrees reasonably well with that found in earlier experiments (greater than 5 days; - Tran Van et al. 1982 and 15 days; - Baron et al. 1981). The similarity of activation times between this model and tooth egression suggests that similar mechanisms may be required in both. This is distinct from the significantly shorter activation period (0.5 days) observed in PTH-stimulated rodent alveolar bone (Vignery et al. 1978). It had been previously suggested that this difference may be due to the existence of significant numbers of osteoclast precursors in the remodeling site in the PTH study as compared to the tooth migration model. Sites of future resorption in the orthodontic model are also considered to be poor in osteoclast precursors. Although there is bone resorption occurring around control teeth, this is almost exclusively on the distal alveolar bone surface (Table II). By contrast, the primary site of resorption in the treated animals is on the mesial alveolar bone surface. Another previously cited difference between these animal models is that one
407
utilizes a mechanical signal whereas the other is hormonal. Like tooth egression, orthodontic tooth movement is also initiated by a mechanical signal, and therefore lends further support to the idea that mechanical activation of bone remodeling has a different mechanism than hormonal. The longer bone resorption phase reported in this study may represent an additional difference between the orthodontic and distal dental drift paradigms. In orthodontic tooth movement, a significant amount of tissue damage appears to be unavoidable (Rygh 1972; Rygh 1973; Rygh et ai. 1973). Such changes have not been reported in tooth migration or drift. Microdamage in bone occurring as a result of fatigue has been linked to the initiation of bone remodeling (Burr et al. 1985). This association has also been reported in the osteoporosis seen in conjunction with the stabilization of fractures using implants (Perren et al. 1988) and adjacent to endosseous titanium implants (Roberts et al. 1989). The finding that a greater amount of bone resorption occurs in the occlusal half of the pressure side of the orthodontically-treated teeth further suggests that the resorption response may be related to tissue damage in this model, as areas of localized necrosis of the PDL are seen only in this quadrant during the timeframe of the resorption wave. However, damage and repair does not entirely explain the bone turnover dynamics reported here, because tissue necrosis is not seen in the other quadrants, yet similar bone changes occur. Frost has proposed that skeletal tissue relates to its biomechanical environment through a mechanism he calls the “mechanostat” (Frost 1987). This model proposes that increased mechanical usage causes reduced bone remodeling and increased modeling, with the end result of increased bone density. Conversely, decreased mechanical usage results in bone loss by enhancing remodeling and inhibiting modeling. If the bone turnover dynamics seen in orthodontic treatment are viewed as solely a remodeling type of change, then these findings would seem to contradict the mechanostat principle, because the modeling seen in controls becomes more like remodeling. However, the relationship between mechanical usage and orthodontic tooth movement remains unclear, and the degree to which the bone turnover changes reported here represent modeling, remodeling, damage/repair, or combinations also requires further elucidation. There are several noteworthy issues with regard to the bone turnover at sites of pressure and tension. In control animals, the observations of bone resorption on the distal alveolar surface and the failure to demonstrate any peak in its activity are consistent with an earlier report of socket remodeling at that location facilitating distal drift (Vignery et al. 1980). The application of a mesially-directed orthodontic force rapidly alters this pattern, resulting in the stimulation of a wave of resorption on both mesial and distal surfaces, more pronounced on the mesial . In orthodontically-treated tissues, bone formation is characterized by a stimulation beginning between day 5 and day 7. These changes also occur on both mesial and distal alveolar bone surfaces. Viewed along with the observation that an early resorption peak can be detected on both surfaces, and that these changes are demonstrable in the total tissue as well, the conclusion that both pressure and tension signals stimulate similar events in the alveolar bone seems reasonable. Differences between the two surfaces in the controls occurred as a result of normal distal drift and possibly animal manipulations required for appliance placement (Tran Van et al. 1982). In order to better appreciate the relative amounts of bone resorption and formation occurring at sites of pressure and tension, osteoblast and osteoclast surface perimeters are plotted over time after adjusting for site-specific control values (Fig. 3). Although an early wave of resorption is followed by
G. J. King et al.: Bone in orthodontic
408
a wave of formation, the relative amounts compared to control values at these sites are markedly different. Bone accrues at tension sites due to an inhibition of resorption and a stimulation of formation, with the most marked changes occurring late in the tooth movement. At pressure sites, bone is lost due to the stimulation of resorption and ‘inhibition of formation occurring early in the tooth movement. These data suggest that the normal degree of coupling between bone formation and resorption is disturbed in orthodontic tooth movement, fairly early in pressure sites and later in tension, and that these imbalances facilitate tooth relocation by removing bone ahead of the tooth and depositing it behind. Although we are aware of other instances, such as postmenopausal osteoporosis (for review see Frost 1985), in which quantitative uncoupling in haversian remodeling is thought to occur, it is premature to speculate whether the bone changes reported here have a similar mechanism. Our data further suggest that enhancement of bone formation in orthodontic treatment occurs by increases in the numbers of active bone-forming cells. Whether significant increases in their functioning also occurred is more equivocal. Significant late changes in osteoid perimeter may only reflect increased numbers of osteoblasts, since no changes in osteoid thickness or bone formation rate were detected. Tritiated thymidine studies have shown that orthodontic treatment does enhance cell division in the PDL, and that the primary site toward which these newly divided cells migrate is the alveolar bone (Smith et al. 1980; Roberts & Chase 1981). The mineralization rates and bone formation rates reported in both groups tend to be lower than those previously observed (Vignery et al. 1980). It is likely that this discrepancy is due to age differences between the animals used. The animals in the current study weighed twice as much as those in the former and were considered to be very slowly growing. Age differences in bone turnover parameters have been shown to be of significance (Reid et al. 1987). It is currently unknown whether the bone turnover patterns described here are sensitive to the magnitude of mechanical force or the manner by which it is applied. Studies on the adaptive remodeling of bone to various mechanical stresses do indicate that the process is extremely sensitive to alterations in both the magnitude and distribution of the strain generated within the bone (Rubin & Lanyon 1987). The nature of the feedback mechanism that regulates alterations in bone turnover in response to orthodontic treatment is likely to be complicated. Bone turnover remains in balance as the maxillary molars drift distally. Such a pattern could be genetically programmed (Rubin & Lanyon 1987) or mediated by a low-level net distally-directed strain. When the mesially-directed orthodontic stress is applied, a new strain environment is likely to be established in response to tooth displacement. In addition to the strain itself, which can directly affect bone (Hasagawa et al. 1985), numerous other new conditions with the potential for altering bone turnover also can occur. Matrixbound biologically active molecules like TGF, could be released either directly by tissue strain or by the early resorptive phase of the bone turnover (Seyedin et al. 1985). Other cytokines (Davidovitch et al. 1988) or prostaglandins (Somjen et al. 1980), which could further alter the bone turnover pattern, are also secreted by cells in the area. In addition, fluid flow could be altered in the bone (Kufahl et al. 1990), which could be reflected in fluid sheer stresses stimulating the production of osteoblastic cyclic AMP (Reich et al. 1990) and streaming potentials (Co&ran et al. 1989). These changes could also help to mediate the response to the altered environment caused by orthodontic treatment. Finally, bone proteoglycans have been shown to change orientation in response to stress (Skerry
tooth movement
et al. 1988). These molecular changes also have the potential to mediate alterations of cell behavior (Culp et al. 1986). In conclusion, this histomorphometric study demonstrated that bone turnover in the paradental alveolar bone of orthodontically-treated teeth is not balanced in the short term, but instead is characterized by tandem periods of activation, resorption, reversal, and formation. Furthermore, the study demonstrated that these bone turnover dynamics occur at sites of both pressure and tension, and that bone accretion or removal is controlled by the relative amounts of bone formation and resorption occurring at each site. Acknowledgments:
This
work
was
supported
by National
Institute of
Dental Research Award DE08659 from the National Institutes of Health. The authors wish to thank Ms. Liz McCoy for her technical assistance
and Ms. Kathy Cannon for typing the manuscript.
References Baron, R. A.; Siverglate, R.; ?ross. R.: Vignery, A. Dynamic histomorphometry of trabecular bone remodeling in the rat vertebrae. 1981. Baron, R. A.; Vignery, A.; Neff, L.; Siverglate, A.; Santa Marie, A. Bone histomorphometry: Techniques and interpretation. Reeker, R. R.. ed. CRC Press, Boca Raton 13-35; 1983. Bridges, T.; King, G.; Mohammed, A. The effect of age on tooth movement and mineral density in the alveolar tissues of the rat. Am. .I. Orthod. 93: 245-250; 1988. Burr, D. B.; Martin, R. B.; Schaffer, M. B.; Radin, E. L. Bone remodeling in response to in viva fatigue microdamage. J. Biomech. l&189-200; 1985. Cocbran, G. V. B.; Dell, D. G.; Palmieri. V. R.; Johnson, M. W.: Otter. M. W.; Kadaba, M. P. An improved design of electrodes for measurement of streaming potentials on wet bone in vitro and in viva. J. Biomech. 22:745750; 1989. Gulp, L. A.: Laterra, J.; Lark. M. W.; Beyth. R. J.; Tobey, S. T. Heparan sulphate proteoglycan as mediator of some adhesive responses and cytoskeleta1 reorganization of cells on fibronectin matrices: Independent versus cooperative function. Funcrion of the proteoglycans. Chichester: John Wiley and Sons; 1986. Davidovitch. Z.: Nicolay, 0.: Ngan, P. W.: Shaofeld, J. L. Neurotransmitters. cytokines, and the control of alveolar bone remodeling in orthodontics. Dent.
Clin.
North
Am.
32:411435;
1988.
Frost. H. M. The skeletal intermediary organization: A synthesis. Peck, W. A., ed. Bone and mineral research. Vol. 3. New York: Elsevier Science Publishers; 1985; 49-107. Hasagawa, S.; Sate, S.; Saito, S.: Suzuki, Y.; Brunette. D. M. Mechanical stretching increases the number of cultured bone cells synthesizing DNA and alters their pattern of protein synthesis. C&if. Tin. Inr. 37:431-436; 1985. King, G. J.: Keeling, S. D.: McCoy. E. A.; Ward, T. H. Measuring dental drift and orthodontic tooth movement in response to various initial forces in adult rats. Am. J. Orthod. Dent. Orrhop. 99:45M65; 1991. Kufahl, R. H.; Saha, S. A theoretical model for stress-generated fluid flow in the canaliculi-lacunae network in bone tissue. J. Biomech. 23:171-80: 1990. Lanyon, L. E. Functional strain as a determinant for bone remodeling. C&if. Tiss. Int. 36 (Suppl. 1) S56-661; 1984. Marie, P. J.; Hott, M.; Perheentupa, J. Effects of epidermal growth factor on bone formation and resorption in viva. Physio. Endoc. Metab. 258:E275281:
1990.
Markostamou, M. K.; Baron, R. Etude quantitative de I’ostioclasie su la par01 alv6olaire en orthodontic exptrimentale cher le rat. Orthod. Fr. 44~245-258: 1973.
Parfitt, A. M. The cellular basis of bone remodeling: The quick turn concept reexamined in light of recent advances in the cell biology of bone. C&if. Tin.
Int.
36:537-545;
1984.
Parfitt, M. A.: Drezner, M. K.; Glorieux, F. H.; Kanis, J. A.; Malluche, H.; Meunier, P. J.; Ott, S. M.: Reeker, R. R. Bone histomorphometry: Standardization of nomenclature, symbols, and units. J. Bone Min. Res. 2:595610:
1987.
Perren, S. M.: Cordey, J.; Ralm. B. A.; Gamier. E.: Schneider. E. Early temporary porosis of bone induced by internal fixation implants. A reaction to necrosis, not stress protection? Clin. Orthop. Rel Res. 232:139-51: 1988. Reich. K. M.: Gay, C. V.; Fraozos. J. A. Fluid shear stress as a mediator of
G. J. King et al.: Bone in orthodontic tooth movement osteoblast cyclic adenosine monophosphate production. J. Cell Physiol. 143: 1004: 1990. Reid, S. A.; Boyd, A. Changes in mineral density distribution in human bone with age: Image analysis using backscattered electrons in the SEM. J. Bone Min. Res. 2:13-22; 1987. Reitan, K. Clinical and histological observations on tooth movement during and after orthodontic treatment. Am. J. Orthod. 53:721-745; 1967. Roberts. W. E.; Garetto. L. P.; DeCastro, R. A. Remodeling of devitalized bone threatens periosteal margin integrity of endosseous titanium implants with threaded or smooth surfaces: Indications for provisional loading and axially directed occlusion. J. Ind. DEW. Assoc. 68:19-24; 1989. Roberts, W. E.: Chase, D. C. Kinetics of cell proliferation and migration associated with orthodontically-induced osteogenesis. J. Dent. Res. 60:174-181; 1981. Rubin. C. T.; Lanyon, L. E. Osteoregulatory nature of mechanical stimuli: Function as a determinant for adaptive remodeling in bone. J. Orfhop. Res. 5:300-310; 1987. Rygh. P. Ultrastructural changes in pressure zones of human periodontium incident to orthodontic movement. Acta Odonr. Scond. 31:109-122; 1973. Rygh, P. Ultrastructural changes in pressure zones of rat molar periodontium incident to orthodontic movement. Acm Odont. Stand. 30575-593; 1972. Rygh. P. Ultrastructural changes in tension zones of rat molar periodontium incident to orthodontic movement. Am. J. Orthod. 70:269-281; 1976. Rygh, P.: Selvig, K. Erythrocytic crystallization in rat molar periodontium incident to tooth movement. Stand. J. Dent. Res. 81:62-73; 1973. Sadowsky, C.: BeGole, E. Long-term effects of orthodontic treatment on periodontal health. Am. J. Orthod. 80:156-172; 1981. Saffar, J. L.; Klapisz-Wolikow, Changes in mast cell number during the activation phase of an induced synchronized remodeling sequence in the rat. Bone 11:369-372: 1990.
409 Scheffe, H. A method for judging all contrasts in the analysis of variance. Biometricka 40~87-104: 1953. Seyedin. S. M.; Thomas, T. C.; Thompson, A. Y.; Rosen, D. M.; Pie& K. A. Purification and characterization of two cartilage-inducing factors from bcvine demineralized bone. Proc. N&l. Acad. Sci. USA 82:2267-2271: 1985. Skeny. T. M.; Bitensky. L.; Chayen, J.; Lanyon, L. E. Loading-related reorientation of bone proteoglycan in viva. Strain memory in bone tissue? J. Or‘hop. Res. 6547-51; 1988. Smith, R. K.; Roberts, W. E. Cell kinetics of the initial response to orthodontitally-induced osteogenesis in rat molar periodontal ligament. C&f. En. fnr. 30:51-56: 1980. Somjen, D.: Bindennan, I.; Bergen. E.; Harrell. A. Bone modeling induced by physical stress is prostaglandin mediated. Biochem. Biophys. Acta. 91:627: 1980. Storey. E. The nature of tooth movement. Am. J. Orthod. 63:292-314; 1973. Tran Van, P.; Vignery, A.: Baron, R. Cellular kinetics of the bone remodeling sequence in the rat. Anar. Rec. 202:445-451; 1982. Vignery, A.; Baron, R. Effects of parathyroid hormone on the osteoclastic pool, bone resorption, and formation in rat alveolar bone. C&if. Tim. Res. 26: 23-28; 1978. Vignery, A.; Baron, R. Dynamic histomorphometry of alveolar bone remodeling in the adult rat. Anar. Rec. 196~191-200; 1980.
Date Received: October 25, 1990 Dare Revised: March 22. 1991 Date Accepted: June 6. 1991
