Aim was to evaluate effect of unilateral distraction osteogenesis (DO) on mandibular morphology in rabbits with antigen-induced arthritis in the temporomandibular joint (TMJ). Forty 8-week-old rabbits were divided into four groups. In groups A,C, arthritis was induced in the right TMJ. Groups A,B underwent DO. Group D served as control group. Cephalometric analysis of mandibular angle, mandibular ramus height, mandibular collum height, and total posterior mandibular height was done on CT-scans preoperatively (T0), after distraction (T1), and at euthanasia (T2). Two-factor ANOVA evaluated the effect of DO and antigen-induced arthritis. No effect of DO or arthritis was observed on mandibular angle or mandibular collum height. For T0–T1, DO increased mandibular ramus height 12.3% (95% CI 5.2–19.4%) in group B ( P = 0.001) and total posterior mandibular height 6.2% (95% CI 0.3–12.1%) in group A ( P = 0.04) and 10.0% (95% CI 4.3–15.7%) in group B ( P = 0.001). For T1–T2, no significant changes occurred in arthritic rabbits (group A). In conclusion, DO increased total posterior mandibular height in rabbits with arthritis. Postoperatively, no significant effect of DO was observed in rabbits with arthritis. Mandibular DO could be a viable treatment modality in patients with TMJ-arthritis.
Juvenile idiopathic arthritis (JIA) is a chronic condition characterized by inflammation of the synovial tissue, with onset before the age of 16 years and a persistence of more than 6 weeks. The incidence of JIA varies between 7 and 23/100,000 children per year in the Scandinavian countries. Arthritis of the temporomandibular joint (TMJ) in JIA affects the mandibular intracapsular growth zones, thereby inducing unilateral or bilateral degenerative changes of the mandibular morphology. Diagnostic imaging has revealed morphological changes of the mandibular condyle in 62–87% of patients with JIA. TMJ arthritis causes growth disturbances of the entire craniomandibular complex with high variation in severity depending on the time of TMJ arthritis onset and overall craniofacial growth pattern. Cephalometric studies have shown a high prevalence of mandibular retrognathism, asymmetry, a clockwise rotational skeletal growth pattern with increased anterior lower face height, reduced posterior lower face height, and open bite.
Growth disturbances may be reduced by early interceptive treatment, which promotes normal growth, reduces orofacial symptoms, and normalizes facial proportions. However, severe growth disturbances may necessitate surgical procedures, e.g. conventional orthognathic surgery, alloplastic joint prostheses, and costochondral grafts to normalize occlusion, facial proportions, and jaw relations. Distraction osteogenesis (DO) is a gradual elongation of segmented bone by the use of an external force and the creation of new bone in the distraction gap. Studies have shown consistent results following mandibular DO (MDO), which has been advocated as a predictable treatment modality with few complications.
Several authors have argued that MDO is a viable treatment modality in patients with JIA. A low rate of degenerative changes in the TMJ of JIA patients following MDO has been reported in two studies, and only minor subjective complaints and limited objective changes in various functional parameters following MDO in patients with JIA have been reported. In finite element analyses of MDO, it has been reported that DO changes the stress distribution in the TMJ, but that the induced stress is low and probably results in only insignificant changes in the morphology of the condyle and TMJ. In experimental studies involving healthy animals, it has been indicated that the direction of the vector may influence the effect of DO on the TMJ. Furthermore, adaptive changes in the TMJ following MDO at physiological levels have been documented in healthy animals. Studies assessing the effect of MDO in animals with pathological conditions in the TMJ are currently not available; therefore the purpose of the present study was to evaluate the effect of unilateral MDO on mandibular morphology in rabbits with antigen-induced TMJ arthritis.
Materials and methods
The study was conducted in accordance with national guidelines for the care and use of animals in research. The research protocol was approved by the Animal Experiments Inspectorate in Denmark. Forty 8-week-old New Zealand white rabbits were used ( Oryctolagus cuniculus ; KB Lidköpings Kaninfarm, Sweden). Before commencing the study, the rabbits were allowed to acclimatize for 14 days. All rabbits were housed in pairs in approved caging with slotted floors in a temperature and humidity-controlled room on a standardized light–dark routine. A laboratory rabbit diet supplemented with fresh hay and tap water was provided ad libitum throughout the study. Environmental enrichment consisted of a variety of daily fresh vegetables.
The rabbits were randomized into four groups by block randomization (groups A, B, C, and D) ( Fig. 1 ). Arthritis was induced in the right TMJ of the animals in groups A and C by the method introduced by Kapila et al. DO of the right ramus was performed in groups A and B. Group D served as a healthy, non-operated TMJ control group.
At the age of 10 weeks, the rabbits in groups A and C were pre-sensitized to ovalbumin using 1 ml of ovalbumin (1 mg/ml; Sigma Chemicals, Germany) dissolved in physiological saline and Freund’s complete adjuvant (Sigma Chemicals). Two weeks later, the procedure was repeated using 1 ml of ovalbumin dissolved in physiological saline and Freund’s incomplete adjuvant. After a further 1 week, the rabbits in groups A and C were tested for sensitivity to ovalbumin by subcutaneous injection of 1 ml (1 mg/ml) ovalbumin on a shaved spot (3 cm × 3 cm) on the back. Twelve hours later, the shaved spot was inspected to detect clinical signs of inflammation. Redness (more than 1 cm × 1 cm) or swelling (more than 0.5 cm × 0.5 cm) was an indication of successful sensitization. One week after sensitization and every 3 weeks during the remaining study period, arthritis was induced in the right TMJ by injection of sterile, filtered 0.1 ml ovalbumin dissolved in physiological saline (10 mg/ml). The rabbits were sedated (Rompun Vet; Bayer Animal Health GmbH, Germany) to allow manipulation of the mandible and palpation of the condylar head before any intra-articular injection was performed. At the times corresponding to intra-articular injections of ovalbumin in groups A and C, the rabbits in groups B and D were sham-treated with intra-articular injections of 0.1 ml sterile and physiological saline in the right TMJ.
At the age of 15 weeks, a surgical procedure was performed in groups A and B to allow MDO of the right mandibular ramus. After general anaesthesia (Rompun Vet), the skin lateral to the right mandibular ramus was shaved, and through a submandibular incision followed by sub-periosteal dissection, the lateral part of the mandible was exposed. The vector of the distraction device (Alveolar Ridge Distractor; Synthes, USA) was determined and placed at a 90° angle to the occlusal plane. After adjusting and determining the final position of the distraction device relative to the lateral surface of the ramus, the central part of the horizontal ramus osteotomy corresponding to the location of the distraction device was performed using piezoelectric surgery (Piezosurgery; Mectron s.p.a., Carasco, Italy). Next, the distraction device was fixed with six 4-mm-long 1.3-mm-diameter self-tapping screws (Synthes), and the osteotomy was completed to the posterior and anterior borders of the ramus ( Fig. 2 ). The functioning of the distraction device and the absence of bony resistance were tested prior to layered suturing of the skin (Vicryl 4–0, Ethicon, USA and Monosof 3–0, Covidien Medical Supplies, USA). The activating arm (10 mm) of the distraction device penetrated the skin near the mandibular angle allowing good access for activation. After a latency period of 5 days, the distraction phase was initiated with activation of the distraction device by 1.3 mm every second day until a total of approximately 7 mm of distraction was achieved. During activation, the rabbits were sedated (Rompun Vet) to allow safe activation of the device and reduce the stress levels of the animals.
At the age of 25 weeks and after 8 weeks of consolidation, the rabbits were sedated (Rompun Vet) and euthanized by the administration of an overdose of intravenous pentobarbital (200 mg/ml) through an ear vein.
Computed tomography (CT) scans of the cranium were obtained with 1-mm slice thickness and a distance of 1 mm between slices (Sensation 10; Siemens Medical Solutions, Forchheim, Germany). Three CT scans were obtained: the first at T0 (week 14, before surgery), the second at T1 (week 17, after DO), and the third at T3 (week 25, at euthanasia). The scanned images were digitized by one of the authors (KA) using Mimics 15 software (Materialise, Belgium). The following anatomical landmarks and lines were defined ( Fig. 2 ): the condylar midpoint (Cm) defined as the midpoint of the condylar head; the articulare (Ar) defined as the most inferior point on the convex margin located dorsal to the mandibular collum; the gonion (Go) defined as the most inferior and anterior point on the ramus; the basion (Ba) defined as the most inferior point located anteriorly on the lower border of the mandibular body; and the mandibular line (ML) defined as the line between Go and Ba.
The digitized CT scans were traced on the right side of the mandible using a method modified from Stoustrup et al., to include the measurement of (1) total posterior mandibular height (TPMH), Cm–Go; (2) ramus mandibular height, Ar–Go; (3) collum mandibular height, Cm–Ar; and (4) mandibular angle, ∠Cm–Go–Ba. Fig. 3 shows the lines and angle constructed and measured to assess the mandibular morphology.
Percentage changes (Δ%) in the relationship between T0 and T1 for the mandibular morphology variables were estimated using the formula (VT1 − VT0) × 100/VT0, while percentage changes (Δ%) between T1 and T2 in the mandibular morphology variables were estimated using the formula (VT2 − VT1) × 100/VT1, where VT0 is the length or angle of a given variable at T0, VT1 is the length or angle of the same variable at T1, and VT2 is the length or angle of the same variable at T2. A positive value for Δ% denotes an increase in the variable (%), and a negative value in Δ% denotes a reduction in the variable (%).
Data were processed using the software program SPSS version 18 (SPSS Inc., Chicago, IL, USA). Descriptive statistics were used to describe the mean of changes between the outcome variables. For the periods T0–T1 and T1–T2, a two-factor analysis of variance (ANOVA) was conducted to examine the separate effects of DO and antigen-induced arthritis on each of the four outcome variables (mandibular angle, ramus mandibular height, collum mandibular height, and total posterior mandibular height). Furthermore, the interaction between DO and antigen-induced arthritis was estimated. A P -value of <0.05 was considered statistically significant. Intra-observer variation was evaluated by repeated measurement of all outcome variables on 20 randomly chosen CT scans within an interval of 1 month. Statistical analyses were undertaken using a paired t -test to determine the intra-observer difference.
Rabbits in groups A and C were successfully sensitized to ovalbumin. Eight animals died during the administration of general anaesthesia (one in group A, three in group B, none in group C, and four in group D). During the study, the well-being of the remaining rabbits in all groups was preserved. No differences in daily food intake were noticed.
No technical complications in relation to the distraction device were noted during the surgical procedure or postoperatively. An ANOVA test of the post-mortem weight was conducted, and no significant inter-group weight differences were found. CT data from the scanning of one rabbit was irretrievable and, consequently, the rabbit was excluded (group C). No significant intra-observer variation was found.
Linear and angular measurements are presented in Table 1 . Between T0 and T1 ( Table 2 ) and between T1 and T2 ( Table 3 ), no significant effect of DO or arthritis on the parameters mandibular angle and collum mandibular height was observed in any of the groups.
( n = 8)
( n = 8)
( n = 7)
( n = 8)
|Mandibular angle (°)||106.8 (4.6)||105.5 (8.3)||109.0 (2.6)||108.8 (2.1)|
|Ramus mandibular height (mm)||29.7 (1.5)||29.7 (5.2)||30.4 (1.8)||30.0 (1.4)|
|Collum mandibular height (mm)||23.6 (0.9)||23.0 (1.3)||24.4 (1.0)||23.8 (0.8)|
|Total posterior mandibular height (mm)||44.9 (2.0)||44.9 (3.4)||45.7 (1.7)||45.4 (1.6)|
|Mandibular angle (°)||106.5 (3.7)||106.3 (3.7)||109.6 (2.5)||107.6 (3.1)|
|Ramus mandibular height (mm)||32.5 (2.8)||33.8 (2.1)||31.1 (1.0)||29.7 (1.3)|
|Collum mandibular height (mm)||24.5 (1.6)||23.2 (1.2)||24.8 (0.8)||23.9 (0.7)|
|Total posterior mandibular height (mm)||49.2 (3.8)||50.1 (2.2)||47.0 (1.0)||45.1 (1.3)|
|Mandibular angle (°)||106.9 (2.9)||106.7 (2.6)||110.6 (2.9)||108.8 (2.1)|
|Ramus mandibular height (mm)||33.0 (2.5)||33.7 (2.4)||31.3 (1.5)||31.0 (1.1)|
|Collum mandibular height (mm)||24.8 (1.4)||23.4 (0.8)||24.4 (1.4)||24.3 (0.8)|
|Total posterior mandibular height (mm)||50.1 (3.2)||50.4 (2.5)||46.6 (1.4)||47.0 (1.4)|
a Mandibular angle: ∠Cm–Go–Ba. Ramus mandibular height: Ar–Go. Collum mandibular height: Cm–Ar. Total posterior mandibular height: Cm–Go. Group A were treated with distraction osteogenesis and antigen-induced arthritis. Group B were treated with distraction osteogenesis. Group C were treated with antigen-induced arthritis. Group D was the control group.
|DO treatment effect, healthy animals
(Group B) ( n = 8)
|DO treatment effect, animals with arthritis
(Group A) ( n = 8)
|Arthritis effect, untreated animals
(Group C) ( n = 7)
|Arthritis effect, DO treated animals
(Group A) ( n = 8)
|Interaction between DO and arthritis
(Group A) ( n = 8)
(95% CI −2.6 to 2.9%)
P = 0.91
(95% CI −3.7 to 1.9%)
P = 0.52
(95% CI −1.1 to 4.5%)
P = 0.23
(95% CI −2.0 to 3.4%)
P = 0.63
(95% CI −5.0 to 2.9%)
P = 0.59
|Ramus mandibular height||12.3%
(95% CI 5.2 to 19.4%)
P = 0.001 *
(95% CI −0.9 to 13.9%)
P = 0.08
(95% CI −3.9 to 10.9%)
P = 0.34
(95% CI −9.4 to 4.8%)
P = 0.51
(95% CI −16.1 to 4.3%)
P = 0.04 *
|Collum mandibular height||0.6%
(95% CI −3.5 to 4.6%)
P = 0.78
(95% CI −2.3 to 6.1%)
P = 0.36
(95% CI −2.9 to 5.4%)
P = 0.54
(95% CI −1.4 to 6.7%)
P = 0.19
(95% CI −4.5 to 7.2%)
P = 0.64
|Total posterior mandibular height||10.0%
(95% CI 4.3 to 15.7%)
P = 0.001 *
(95% CI 0.3 to 12.1%)
P = 0.04 *
(95% CI −2.5 to 9.3%)
P = 0.25
(95% CI −6.1 to 5.3%)
P = 0.89
(95% CI −11.9 to 4.4%)
P = 0.35
a Data are presented as the mean percentage change and the 95% confidence interval (mean Δ% (95% CI)) with P -values. Mandibular angle: ∠Cm–Go–Ba. Ramus mandibular height: Ar–Go. Collum mandibular height: Cm–Ar. Total posterior mandibular height: Cm–Go. Group A were treated with distraction osteogenesis and antigen-induced arthritis. Group B were treated with distraction osteogenesis. Group C were treated with antigen-induced arthritis. Group D was the control group.