Effect of low-intensity pulsed ultrasound on injured temporomandibular joints with or without articular disc removal in a rabbit model.

BACKGROUND/
PURPOSE: Dynamic stimulation can induce bone and cartilage growth. The purpose of this study was to examine the effect of low-intensity pulsed ultrasound (LIPUS) on injured temporomandibular joints (TMJs) in a rabbit model.
MATERIALS AND METHODS: Twenty-four female Japanese white rabbits (age: 12-16 weeks, weight: 2.0-2.5 kg) were equally divided into 4 groups. In two groups, discectomy was performed with (the LD group) and without (the D group) subsequent LIPUS treatment. In the other groups, a sham operation was performed with (the LC group) and without (the C group) subsequent LIPUS treatment. Two animals in each group were sacrificed at each time point (2, 4, and 8 weeks postoperatively). Mandibular measurements were made using three-dimensional computed tomography. We performed histological and immunohistochemical examination of the articular disc, and the cartilage layer and bone at the 30- and 60-degree sites in each condyle.
RESULTS: There were no statistically significant differences among the groups in terms of thickness of the disc or the fibrous articular zone, or the number of BMP-2 positive cells. In terms of mandibular length, there were differences among the groups after 4 (P = 0.0498) and 8 weeks (P = 0.0260). Specifically, there was a difference between the LC group and the C group after 4 weeks (P = 0.014) and 8 weeks (P = 0.029).
CONCLUSIONS: This study suggests that LIPUS has little effect on cartilage after TMJ injury. It may promote bone growth in a normal TMJ, although discectomy seems to reduce this effect.

Introduction

Many researchers have investigated maxillofacial trauma, due to its high incidence.1, 2, 3 In particular, the temporomandibular joint (TMJ) is reported to be the most likely site of fracture., Fractures of the TMJ can be caused by both direct and indirect trauma. Acute or chronic inflammation of the TMJ following maxillofacial trauma results in trismus and/or pain. For TMJ fractures, conservative rather than surgical treatment is often selected, depending on the fracture site or condition. It has been reported that postoperative complications, such as temporomandibular disorders, may occur after condylar fracture. These may be due to damage to the surface of the mandibular condyle and articular disc. The promotion of fracture and cartilage healing after maxillofacial trauma accelerates a patient's return to normal social activity and reduces complications.

It has been reported that dynamic stimulation, such as by low-intensity pulsed ultrasound (LIPUS), can induce bone and cartilage growth. In an in vivo pilot study, El-Bialy et al. reported the use of LIPUS to enhance the performance of tissue-engineered mandibular condyles in baboons. Oyonarte et al. concluded that the application of LIPUS to the TMJ region of growing rats promoted sagittal and transverse condylar growth. However, whether the with or without of the articular disk affects LIPUS treatment of the TMJ was not investigated.

The purpose of the present study was to examine, histologically and macroscopically, the effects of LIPUS on an injured TMJ, using a rabbit model.

Materials and methods

Surgical procedure

The subjects were 24 female Japanese white rabbits (weight: 2.0–2.5 kg, age: 12–16 weeks). They were divided into four equal groups: the LIPUS after discectomy (LD) group, the LIPUS only (LC) group, the discectomy only (D) group, and the control (C) group (neither LIPUS nor discectomy).

All animal experiments were conducted according to the European Commission Directive 86/609/EEC and with the approval of the Ethical Committee xxx (approval number: xxx). In the LD and D groups, experimental surgery was performed under sedation obtained with sodium pentobarbital (25 mg/kg) injection into the ear vein. The hair in the TMJ region was shaved, and approximately 1.8 ml of 2% lidocaine, containing 1/80,000 epinephrine, was infiltrated. A 1.5-cm horizontal skin incision was made over the pre-auricular region. The zygomatic arch was partially resected, at the superior border, and the TMJ capsule was opened by horizontal incision to expose the condyle. The condylar cartilage was exposed and the articular disc was removed (Fig. 1). The wound was closed with a 4-0 absorbable suture. The same operation was performed on the other temporomandibular joint. In the LC and C groups, a sham procedure was conducted; after exposure of the condyle and disc, the wound was closed in the same way without disc removal.

Figure 1

Intra-operative photograph. D: disc, C: condyle.

Ultrasound treatment

A LIPUS device (BR-sonic, ITO Co., Tokyo, Japan) was used in the experiment, with the following settings (Fig. 2): frequency, 3.0 MHz; duty, 20%; and intensity, 240 mW/cm2. LIPUS application was performed after the head was fixed to restrict excessive movement. One 20-min treatment was performed on the first day after surgery and repeated daily for 2 weeks, as in previous studies.,

Figure 2

A) The low-intensity pulsed ultrasound (LIPUS) device (BR-sonic, ITO Co., Tokyo, Japan). B) LIPUS application was performed to restrict rabbit excessive movement.

Postoperative morphological and histological evaluation

Macroscopic and microscopic changes were assessed in all experimental groups. Two rabbits in each group (eight in total, per time point) were sacrificed via sodium pentobarbital injection into the lateral ear vein at 2, 4, and 8 weeks postoperatively.

LD group: 2 weeks (2 rabbits/4 TMJs), 4 weeks (2 rabbits/4 TMJs), 8 weeks (2 rabbits/4 TMJs)

D group: 2 weeks (2 rabbits/4 TMJs), 4 weeks (2 rabbits/4 TMJs), 8 weeks (2 rabbits/4 TMJs)

LC group: 2 weeks (2 rabbits/4 TMJs), 4 weeks (2 rabbits/4 TMJs), 8 weeks (2 rabbits/4 TMJs)

C group: 2 weeks (2 rabbits/4 TMJs), 4 weeks (2 rabbits/4 TMJs), 8 weeks (2 rabbits/4 TMJs)

Computed tomography measurement

Immediately after sacrifice, computed tomography (CT) scanning of the head was performed in the radiology department using a high-speed CT scanner (Aquilion One; Toshiba Medical Systems Corp., Tochigi, Japan), for 0.5-mm sections (120 kV; 130 mA; 0.75 s/rotation; 0.641 pitch factor). The CT image was three-dimensionally reconstructed, and the items listed below were measured with the Materialise ProPlan CMF imaging software (Materialise Dental NV, Leuven, Belgium). The mean of five measurements was used in subsequent analyses, to increase reproducibility.

Mandibular ramus height: distance between the most superior point of the condyle and the mandibular inferior border (Fig. 3A).

Figure 3

Three-dimensional computed tomography measurement. A) Lateral view: a) mandibular ramus height and b) mandibular length. B) Superior view: c) condylar length and d) condylar width.

Mandibular length: distance between the most anterior point of the condyle and the top edge of the incisor (Fig. 3A).

Condylar length: distance between the anterior and posterior points of the condyle (Fig. 3B).

Condylar width: distance between the most lateral point and the medial point of the condyle (Fig. 3B).

Microscopic condylar cartilage changes

After CT scanning, tissues were immersed in saline and perfused with phosphate-buffered 10% formalin through a catheter placed in the left ventricle of the heart, for fixation. The heads were fixed in 10% phosphate-buffered formalin for 2 h, demineralized in ethylenediaminetetraacetic acid for 4 weeks, embedded in paraffin, serially sectioned in the sagittal plane at a thickness of 5 μm, and stained with hematoxylin and eosin. The histological features of the condyle were assessed by light microscopy. Thickness of the disc, the fibrous articular zone, and the cell proliferation-rich hypertrophic chondrocyte zone were measured at the 30- and 60-degree sites (Fig. 4). The mean of five measurements was used in subsequent analyses.

Figure 4

Photomicrographs of the control group after 2 weeks. A) Hematoxylin and eosin (H&E) staining, original 10 × magnification; the thickness was measured at the 30- and 60-degree sites of the condyle. B) H&E staining, original 100 × magnification; C: condyle, D: disc, E: articular eminence, F: fibrous articular zone, H: cell proliferation-rich hypertrophic chondrocyte zone.

Bone morphogenetic protein (BMP)-2 immunohistochemical staining was performed according to the method described in a previous paper., Positively stained cells per 1000 cells were counted in the region between the 30- and 60-degree sites using 100×-magnification photomicrography (Fig. 5). The mean of five measurements was used in subsequent analyses.

Figure 5

Photomicrographs after 2 weeks. A) The red arrows indicate bone morphogenetic protein (BMP)-2-stained cells in the cartilage layer of the condyle. B) The red arrows indicate BMP-2-stained cells in the bone areas of the condyle.

Statistical analyses

Data were analyzed using IBM SPSS Statistics for Windows, Version 25 (IBM Corp., Armonk, NY, USA). The Kruskal–Wallis test and the Mann-Whiney U test were used with Bonferroni correction to compare the groups at each time point. P-values < 0.05 were considered statistically significant.

Results

Histological analysis of the D and LD groups after discectomy revealed partial degenerative changes in the condyle and articular eminence after 2 weeks (Fig. 6A). After 4 weeks, the cartilage layer on the articular surface exhibited partial resorption and proliferation, and the chondrocyte zone was partially exposed (Fig. 6B). We observed inflammatory cell infiltration into the exposed marrow and resorption of subchondral bone. The condylar surface, including the osseous components, was flattened. After 8 weeks, the articular surfaces exhibited an almost normal amount of cartilage, although clustered chondrocytes were scattered throughout the extracellular matrix (Fig. 6C). In the C and LC groups, normal cartilage, bone, and disc were observed at 2, 4, and 8 weeks post-operatively.

Figure 6

Photomicrographs at 2, 4, and 8 weeks after discectomy (discectomy-only group). A) The condyle and eminence exhibited partial degenerative changes after 2 weeks, in the area bordered by the yellow line. B) The cartilage layer on the articular surface exhibited partial resorption and proliferation, and the chondrocyte zone was partially exposed, after 4 weeks. C) The articular surfaces exhibited almost normal amounts of cartilage after 8 weeks. C: condyle, E: articular eminence.

Comparisons among the groups in each period

Thickness of the disc

There were no statistically significant differences in the thickness of the articular disc between the LC and C groups at any time point (Table 1).

Table 1

Thickness of the articular disc after 2, 4, and 8 weeks, denoted as the mean ± standard deviation. C: the control group (neither LIPUS nor discectomy), LC: the LIPUS only group, LIPUS: low-intensity pulsed ultrasound.

2w	0°	30°	60°	90°
LC (μm)	179.7 ± 47.4	290.9 ± 66.2	246.6 ± 51.6	156.7 ± 45.9
C (μm)	133.5 ± 13.8	257.3 ± 48.6	255.5 ± 74.8	109.9 ± 21.7
4w	0°	30°	60°	90°
LC (μm)	135.5 ± 20.8	295.3 ± 54.6	342.7 ± 44.6	179.3 ± 59.0
C (μm)	126.6 ± 13.4	291.8 ± 46.0	271.2 ± 20.8	143.1 ± 23.7
8w	0°	30°	60°	90°
LC (μm)	110.8 ± 6.9	300.2 ± 12.4	332.2 ± 74.3	126.0 ± 9.8
C (μm)	126.4 ± 11.8	298.1 ± 36.0	251.4 ± 23.5	120.3 ± 11.0

Thickness of the fibrous articular zone

There were no statistically significant differences among the groups at any time point in the thickness at either the 30- or 60-degree sites (Fig. 7).

Figure 7

Thickness of the fibrous articular zone. A) 2 weeks postoperatively. B) 4 weeks postoperatively. C) 8 weeks postoperatively. The column height represents the median value, and the error bars represent the range. ∗ indicates a statistically significant difference among the groups (P < 0.05). C: the control group (neither LIPUS nor discectomy), D: the discectomy only group, LC: the LIPUS only group, LD: the LIPUS after discectomy group, LIPUS: low-intensity pulsed ultrasound.

Thickness of the cell proliferation-rich hypertrophic zone

There were no statistically significant differences among the groups after 2 and 4 weeks (Fig. 8A and B). A statistically significant difference among the groups was observed at the 30-degree site (P = 0.0487) after 8 weeks. Specifically, the difference between the LD group and the C group was statistically significant (P = 0.036) (Fig. 8C).

Figure 8

Thickness of the cell proliferation-rich hypertrophic chondrocyte zone. A) 2 weeks postoperatively. B) 4 weeks postoperatively. C) 8 weeks postoperatively. The column height represents the median value, and the error bars represent the range. ∗ indicates a statistically significant difference among the groups (P < 0.05). C: the control group (neither LIPUS nor discectomy), D: the discectomy only group, LC: the LIPUS only group, LD: the LIPUS after discectomy group, LIPUS: low-intensity pulsed ultrasound.

Number of BMP-2 positive cells

There were no statistically significant differences among the groups in the number of BMP-2 positive cells in the cartilage or bone areas (Fig. 9).

Figure 9

Number of bone morphogenetic protein (BMP)-2-positive cells. A) The cartilage area. B) The bone area. The column height represents the median value, and the error bars represent the range. C: the control group (neither LIPUS nor discectomy), D: the discectomy only group, LC: the LIPUS only group, LD: the LIPUS after discectomy group, LIPUS: low-intensity pulsed ultrasound.

Mandibular morphology by CT imaging

There were no statistically significant differences among the groups in terms of mandibular ramus height, condylar length, or condylar width (Fig. 10A, C, D). Regarding mandibular length, there were statistically significant differences among the groups after 4 weeks (P = 0.0498) and 8 weeks (P = 0.0260). Specifically, there were statistically significant differences between the LC group and the C group after 4 weeks (P = 0.014) and 8 weeks (P = 0.029) (Fig. 10B).

Figure 10

Comparison of mandibular morphology by computed tomography imaging in the experimental groups. A) Mandibular ramus height. B) Mandibular length. C) Condylar length. D) Condylar width. The column height represents the median value, and the error bars represent the range. ∗ indicates a statistically significant difference among the groups (P < 0.05). C: the control group (neither LIPUS nor discectomy), D: the discectomy only group, LC: the LIPUS only group, LD: the LIPUS after discectomy group, LIPUS: low-intensity pulsed ultrasound.

Discussion

Internal derangement of the TMJ can be caused directly or indirectly by maxillofacial trauma. Maxillofacial trauma may cause damage to the cartilage or bone tissue of the condyle, even in the absence of prominent damage such as a fracture. Similarly, the surface of the mandibular condyle may be exposed due to the trauma-induced derangement of the articular disc, which manifests as fragility during mandibular movement. Clinical and experimental studies have revealed that discectomy may cause degenerative changes of bone and cartilage of the condyle and the articular eminence., Takatsuka et al. reported resorption of the cartilage layer on the articular surface at 4 weeks after discectomy, in a rabbit model. However, 6 weeks after discectomy, there was regeneration of the fibrous cartilage layer and an almost normal amount of cartilage at the articular surface. In the study by Sato et al., also based on a rabbit model, osteoarthritic changes and a significant increase in the number of elastic fibers were observed in the condyle and articular eminence, 1 month after discectomy; these changes tended to revert to the same condition as that of controls, 3 months after surgery. The present study was the first in which the effect of LIPUS on trauma-induced displacement or injury of the joint disk was examined.

The inner quality of the maxillofacial bone, including the TMJ, may be strongly influenced by the forces exerted by the masticatory muscles. The area between the 30- and 60-degree sites was selected for measurement in the present study, as the largest load is generated in this area. We measured the thickness of the fibrous articular zone and the cell-proliferation rich hypertrophic chondrocyte zone separately for these two sites.

There have been few reports regarding the optimization of LIPUS intensity. Pilla et al. compared different LIPUS intensities (1.5, 7.5, 15, 30, and 45 mW/cm2) in a rabbit fracture model. They showed a consistent, optimal LIPUS effect at an intensity of 30 mW/cm2. The penetration depth of ultrasound waves depends on their frequency: the half-value depths at 1 MHz and 3 MHz are 50 mm and 16.5 mm, respectively, in the fatty tissue layer, and 9 mm and 3 mm, respectively, in the muscular layer. In physical therapy, 1 MHz is applied to deep sites and 3 MHz to surface sites. A previous study demonstrated that 3 MHz promoted bone regeneration around dental implants more rapidly than 1 MHz did. The LIPUS conditions in the present study (frequency: 3 MHz, and intensity: 240 mW/cm2) were similar to that of a previous study.

We observed no statistically significant differences between the LC and C groups in terms of the thickness of the articular disc, which suggests that LIPUS does not effectively stimulate the disc in its normal position. Additionally, there were no statistically significant differences in the thickness of the fibrous articular zone at either site, among any of the groups, which suggests that LIPUS does not effectively stimulate the fibrous cartilage layer. However, after 4 and 8 weeks, the LD group exhibited a tendency towards a thicker fibrous cartilage layer than the other groups, at the 60-degree site. In the cell proliferation-rich hypertrophic chondrocyte zone, no effect of LIPUS could be detected, although the LD group exhibited greater thickness than the C group at the 30-degree site after 8 weeks. It is thus possible that LIPUS may promote the healing of cartilage injury after discectomy, restoring mandibular condylar gliding.

The LC group exhibited a tendency toward larger mandibular length compared with the C group, although there were no differences between the LD and D groups after 4 or 8 weeks, suggesting that LIPUS may prompt mandibular growth in normal TMJs. There were no other differences between any of the groups in terms of CT measurements. On the other hand, the absence of the articulating disc increases the pressure during mandibular condylar movement, which may have reduced the effect of LIPUS on mandible length in the LC group.

BMPs are important soluble mediators triggering a variety of cellular processes, including cell proliferation and differentiation. In particular, BMP-2 and BMP-7 are clinically approved and have been widely evaluated for cartilage and bone tissue regenerative purposes.18, 19, 20 BMPs are involved in all phases of chondrogenic differentiation and are able to directly regulate the expression of several chondrocyte-specific genes. BMP-2 is known to promote cell proliferation and matrix synthesis in human articular chondrocytes, growth plate chondrocytes, and developing mouse limbs.21, 22, 23 Si et al. discovered that the BMP-2 signal was the greatest in undifferentiated mesenchymal cells, differentiating osteoblasts, and chondroblasts at the stage of intramembranous bone formation and early chondrogenesis, suggesting that BMP-2 mediates the differentiation of mesenchymal cells into osteoblasts and chondroblasts. Suzuki et al. demonstrated that daily LIPUS treatment significantly increased the expression of BMP-2, -4, and -7, and of their receptors, in an in vitro study. LIPUS was also reported to stimulate proteoglycan synthesis in rat chondrocytes by increasing the expression of the aggrecan gene. In the present study, the maximum number of BMP-2 positive cells was observed after 2 weeks in both cartilage and bone tissue in all groups. However, we could not detect a statistically significant effect of LIPUS on BMP-2 expression in condylar cartilage or bone tissue. Previous studies have demonstrated that LIPUS and discectomy lead to condylar tissue changes.,,, We discovered that, at 2 weeks, BMP-2 levels tended to be higher than at the other time points, indicating that surgical invasion may trigger BMP-2 activation. Our results also suggested that LIPUS has only a minor effect on BMP-2 activation, compared to that of surgical stimulation.

The main limitation of this study was the surgical treatment we performed to reproduce trauma-induced injury of the TMJ; there are inevitable differences between the procedure and actual trauma-induced injury of the TMJ. The nature of such injuries makes it difficult to create a uniform rabbit model. Therefore, our study does not provide conclusive evidence that LIPUS as a post-traumatic treatment would provide the same effect in human trauma-induced TMJ injury. In addition, in this experiment, the articular disc was completely removed; we did not include a model of articular derangement or other damage to the articular disc without removal. Before LIPUS can be recommended for the treatment of TMJ trauma, it is necessary to examine whether it is effective in such cases.

In conclusion, the results of this study suggest that LIPUS has little effect on TMJ cartilage after trauma, with or without disc removal. Conversely, LIPUS had a statistically significant effect in promoting bone growth in a normal TMJ in the present study, although this effect was reduced after disc removal.

Declaration of Competing Interest

None.