- Email: [email protected]
The effects of therapeutic vs. high-intensity ultrasound on the rabbit growth plate
ELSEVIER
Journal of Orthopaedic Research
Journal of Orthopaedic Research 21 (2003) 865-871
www.elsevier.com/locate/ort hres
The effects of therapeutic vs. high-intensity ultrasound on the rabbit growth plate Roger Lyon *, Xue C. Liu, Joshua Meier Dqiurtnwnt of Orthopedic Surgery, Children’s Hospital of Wisconsin, Medical College of Wisconsin, 8701 Wutertowm Plunk R o d , Suite 3018. Milwuukee, WI 53226, U S A Received 14 June 2002; accepted 6 February 2003
Abstract
Six of 6-week-old NZ rabbits underwent ultrasound treatment using a therapeutic dose (0.5 W/cmZ)and other six were treated with a higher dose (2.2 W/cm2) to the lateral aspect of the left knee joint for 20 min per day and a total of six weeks. The right knee joint served as a control. The goal of this study is to see if the therapeutic dose and high dose (approximately 45-fold therapeutic dose) will have toxic effects on the physis. Histological review appeared normal growth plate in the therapeutic group. In the high dosage group three of six cases displayed flattening of the distal femoral epiphysis and wedging of the proximal tibial plateau and indistinct growth plate lines. It is of interest to note that there are opening radiolucent area in the lateral aspect on the femoral metaphysis in five of six cases, where bone resorption has taken place. Histological results showed that there are disordered arrays of the cartilaginous cells in the proliferative zone. The height of the lateral physis in the high dose group is not only greater than that in the therapeutic dose (1083.8 vs. 500.3 pm), but also greater than that in their contralateral control (530.7 pm)( P < 0.05). This shortterm study demonstrates that high dose ultrasound has profound pathologic effects in growing bone. Therapeutic doses of ultrasound do not have an adverse effect on bone growth in the short-term follow-up. 0 2003 Orthopaedic Research Society. Published by Elsevier Ltd. All rights reserved. Kejwwrls: Ultrasound; Histology; Growth plate; Toxic effect; X-ray
Introduction
Ultrasound as a therapeutic treatment has often been used by physical therapists to decrease joint stiffness, to reduce pain and muscle spasm, and to improve muscle mobility [4,8]. Most of the studies related to musculoskeletal applications of very low-intensity doses of ultrasound have been focused on outcomes in fracture healing [6,7,13]. A low-intensity ultrasound (1.5 MHz and 30 mW/cm2) was investigated as an accelerator of cortical and cancellous bone fracture healing in smokers and nonsmokers [2]. They found that the healing time for a tibial and distal radius fracture was reduced 41% in smokers and 26% in nonsmokers, and for distal radius fracture was reduced by 51% in smokers and 34% in nonsmokers following the ultrasound treatment. In a placebo-controlled study of mid-shaft tibial osteotomies in rabbits Pilla et al. [ l l ] found that brief periods (20 *Corresponding author. Tel.: +1-414-456-4685; fax: +1-414-4566541. E - m d nddress: [email protected] (R. Lyon).
midday) of pulsed ultrasound accelerated the recovery of torsional strength and stiffness, which was also confirmed by a later study [ 151. Physical therapists use a therapeutic dosage of ultrasound (approximately 0.5 W/cm2) to treat patients with overuse of soft tissue injuries and muscle spasm around the thigh, knee, and back for 5-10 min per spot. The most current study in the literature was done by Spadaro and Albanese in 1998 [13]. They studied the effects of low-intensity ultrasound on bone growth in rats. An ultrasound dose of 30 mW/cm2, which is the routine dose to stimulate fracture repair, was directed at the knee joints of the rats. The researchers found no length differences compared to the control group as well as no significant differences in bone mineral densities of the tibia and femur. The authors concluded that physeal growth is much less affected by this level of ultrasound than is fracture repair [13]. However, doses of ultrasound vary among the different clinical application. Therapeutic ultrasound is still not widely recommended in pediatric patients due to possible disturbances and damage to the physeal plate. In reference to high-intensity of
0736-0266/$ - see front matter 0 2003 Orthopaedic Research Society. Published by Elsevier Ltd. All rights reserved. doi: 10.I0 16/S07~6-0266(03)00047-0
866
R. Lyon et ul. I Journal of’ Orthopuedic Reseurch 21 (2003) 865-871
ultrasound (about 2.0 W/cm2), a study in 1958 reported severely damaged growing bone and joint destruction, including sclerosis of the bone, widening, and displacement of the physis [3]. The use of ultrasound as a surgical instrument involves even higher intensity (5-300 W/cm2), including to fragment calculi, to initiate the healing of nonunion, to ablate disease tissues, and even to remove cement during revision of prosthetic joints [12]. So far there have been few published papers documenting the effects of the different dosages of ultrasound on the growth plate in animal models, and none comparing the biological discrepancies between therapeutic and higher-intensity ultrasound on the growth plate. A more through understanding of the morphologic and histologic response of growing bone to ultrasound is needed. Is there a dose response curve of skeletal growth to ultrasound therapies? Currently the therapeutic application of ultrasound in our institution is 0.5 W/cm2. The maximum output allowed on the current equipment is 2.2 W/cm2. What would the effect of this high dose be on growing bone if an inadvertant exposure of this maximal dose occurred? Establishing a safe range of ultrasound treatment in the pediatric population is important to assure patient safety. We hypothesize that current therapeutic doses of ultrasound (0.5 W/cm2)will not cause degeneration of the cartilaginous cells in the growth plate, while highintensity doses (2.2 W/cm2) will result in degeneration of the cartilaginous cells of the growth plate, and further leads to bony deformities. The aim of this study is to determine if therapeutic and high-intensity doses of ultrasound will have toxic effects on the physeal plates in the rabbit knee joint via radiography and histology as following: (1) Longitudinal lengths of the physis in the microscope, the length and width of the femur or tibia in the X-ray, will be compared between the therapeutic and high dose groups, as well as between the treated vs. untreated contralateral group; (2) comparison of three cartilaginous cell zone between therapeutic and higher-intensity ultrasound will be performed.
inette covering, and two hind limb holes. The hind limbs were through the hole and fastened with straps. N o anesthesia was required. After a coupling gel was applied to the skin (Aquasonic 100, Parker Laboratories, Fairfield, NJ), the Ultrasound transducer was then held against the lateral side of the left hind knee centered at the joint line. The right hind knee is a control group. The target area included the distal femoral and proximal tibial growth plates. The transducer was held in place throughout the treatment period. The ultrasound output was set on pulse outflow, which is the preferred method in the clinical setting. Plain X-rays including anterior-posterior (AP) and lateral views were taken of both the right and left hind legs prior to treatment and at three weeks, and six weeks of the treatment. Plain radiographs were assessed by the following measurements: (1) Femoral length, measured from the superior aspect of the femoral head to the most distal aspect of the femoral condyle; (2) tibial length, measured from the most proximal aspect of the tibial plateau to the most distal aspect of the tibia; (3) femoral width, measured at the widest part of the distal femoral epiphysis; (4) tibial width, measured at the widest part of the proximal tibial epiphysis. Plain and Faxitron X-rays were also examined for morphological changes, such as changes in bone density and general integrity of the bony structure. Following the six weeks treatment period, the animals were sacrificed. The left and right hind legs of the rabbit were dissected out and immediately frozen at -35 “F to preserve tissue integrity. Faxitron X-rays (AP and lateral views) were then taken of both left and right hind legs. Faxitron X-ray provides clearer image than standard X-ray, because it is executed in 50 kV and 1 min. The standard X-ray uses 50 kV, 6 mA and 1 s for exposure. The right and left legs were then cut down and bisected into anterior and posterior halves on a bandsaw and placed in formalin. Following preservation, the joints were placed into an ethanol solution and then decalcified in EDTA. After being decalcified, the joints were embedded in paraffin, cut, and placed on glass slides. Six slides were made out of both the anterior and posterior halves of each knee joint. Three were stained with a standard H & E stain, the other three stained with toluidine blue. The slides were then examined by light microscopy, using an image analyzer (Image Pro Plus, Cybermedics, Siverspring, M D ) to obtain computerized micrographs. These micrographs were then measured using the image analyzer. The following measurements were performed (Fig. l): ( I ) Measurement of the total physeal height, measured from the most proximal to the most distal points of intact cartilage, in both the distal femoral and proximal tibial growth plates; (2) measurement of the reserve zone of cartilage, measured from the most proximal point of intact cartilage to the beginning of the first cell column of proliferating cells in the plane of measurement; ( 3 ) measurement of the proliferative zone of cartilage, measured from the first cell in a pro-
Methods 12 New Zealand white rabbits were divided into two equal groups of six rabbits each. The first group (rabbits 1-6) received the therapeutic dose (0.5 Wlcm’) of ultrasound, while the second group (rabbits 7-12) received the high-intensity dose (2.2 W/cm*). All 12 rabbits were treated with the ultrasound for 20 min a day on the fixed point of lateral knee for a total period of six weeks. A standard medical ultrasound generator and transducer head (Sonicator 730, Mettler Electronics. Anaheim, CA) was used in the ultrasound treatments. This unit has a frequency of 1.0 MHz, with an effective radiating area of 1.8-8.5 crn’. A 2 cm’ crystal applicator was used, with a range of pulsed or continuous output power density from 0.1 to 2.2 W/cm’. The ultrasound was applied with the rabbits held in the prone position. The rabbits were kept in a custom-made wooden frame, comprised of a platform, a body Velcro strap, a head and neckstock-
Fig. 1. Measurements of the height of the physeal plate using the image analyzer: the reserve zone (R), proliferative zone (P), and hypertrophic zone (H) ( x 10, pm).
R. Lyon et al. I Journal
of
Orthopedic Research 21 (2003) 865-871
867
liferating cell column to the first hypertrophic cell in the plane of measurement: (4) measurement of the hypertrophic zone of cartilage, measured from the first hypertrophic cell to the last point of intact cartilage in the plane of measurement. These measurements in three cartilaginous cell zones were done in three areas of each growth plate: the most lateral part of the physis, the mid-line of the physis, and the most medial part of the physis. Mean value and its standard deviation of physeal length in both three cartilaginous zones and X-ray are computed. Using one-way ANOVA. we are able to compare differences of the length or width in the femur and tibia measured by X-ray and physeal height calculated by the image analysis system between the therapeutic and high dosage group. Similarly, using Wilcoxon matched pairs test, we compare them between treated side and untreated side. A possibility of less than 0.05 is considered as significant.
Results
Appearance: At the end of the treatment period, the animals remained in good health, and no gross deformities of limb growth were noted. All six high-dose rabbits received second-degree full thickness skin burns around the area of ultrasound application. These burns were seen between 5 and 15 days after the start of treatment. At the beginning they were presented with slight erytheme, and epidermal detachment, then they were appeared as necrosis of the epidermal. These sores uneventfully covered by dry skin eschar in black and the skin eschar gradually receded at study completion, indicative of being in healing process. There was no gross skin pathology noted on any of the six therapeutic dose rabbits. Faxitroii X-ray analysis: Faxitron X-ray analysis showed marked differences between therapeutic (0.5 Wlcm2) and high dose groups (2.2 Wlcm2). None of six cases in the therapeutic group showed morphologic changes. Both the distal femoral and proximal tibial growth plates of the left (treated) and the right (untreated) legs remained open, demonstrating active longitudinal bone growth. The left physis showed no evidence of either early or delayed closure when compared to the right. However, in 1 of 6 cases the left knee showed an indistinct area of the growth plate in the distal femur and proximal tibia when compared to the right side (Fig. 2). The high-dose group (2.2 Wlcm2), however, showed significant bony pathology in 5 distal femurs, 3 proximal tibias, and 1 proximal fibula. Those five rabbits had a radiolucent area in the lateral aspect of the femoral metaphysis adjacent to the growth plate in the AP view, probably resulting from bone resorption due to ultrasound effects (Fig. 3). Changes of contour in the lateralposterior metaphysis suggest marked bone resorption of the distal femoral metaphysis. Radiographic flattening of the lateral side of the distal femoral epiphysis was observed in three of the high-dose group (Fig. 4). Varus angulation of the distal femur due to bone resorption was seen in one high-dosed group as well (Fig. 4). Three
Fig. 2. In therapeutic group Faxitron X-ray shows normal bony anatomy in both the left and right knee, while only one illustrated unclear physeal line in the lateral aspect of the femur and tibia.
growth plate in the knee joint Fig. 3 . Comparison of morpho~ogica~ Faxitron X-ray in higher dose group. The left knee (left) demonstrated radiolucent area on the lateral metaphysis of the distal femur, which may result from bony resorption, while right knee kept intact.
of six rabbits showed indistinct physeal line in the lateral epiphysis of the distal femur. Three of six cases showed indistinct radiographic growth plate lines in the proximal tibia and wedging of the tibial plateau. The reminder demonstrated normal radiographic physeal characteristics. Routine X-ray: There were no significant differences of the length or width between left femurltibia and right femurltibia within either dosage groups, and no differences of the length and width in the femur or tibia between therapeutic and high dosage groups at the beginning, three weeks and six weeks of treatment ( P > 0.05). Histological measurements: A comparison of height in the three cartilaginous cell zone in the femoral and tibial physis in the therapeutic treated rabbits showed no
R . Lyon et 01. I Journal qf Orthopclrdic Reserirc11 21 (2003) 865-871
868
Fig. 4. Comparison of morphological growth plate in the knee joint X-ray in high dose group. Left knee (treated) shows a flatten of condyle and varus of the distal femur. The tibia became wedging and partial sclerosis on the lateral side with indistinct physeal line. The right knee (right) shows normal contour of the knee joint with open growth plate.
statistically significant differences between six left (treated in 0.5 W/cm2)and right (untreated) knees ( P > 0.05). Table 1 shows that overall physeal height in lateral,
medial, and middle part of the femur and tibia pose no significant differences. Significant differences were found in comparing the treated knees of the therapeutic and high-dose groups (Table 2) ( P < 0.05). The mean height of the lateral part of the left femur growth plate in the therapeutic treated rabbits was 500.3 pm, while the mean in the high dose group was 1083.8 pm. Likewise, the mean height of the middle of the femoral reserve zone in the therapeutic treated group was 92.2 pm, while this mean value in the high dose group was 168.3 pm ( P < 0.05). Measurements were also markedly increased in the high-dose group compared to the therapeutic group with respect to the left femoral physeal height in lateral reserve zone. Significant differences were also found comparing the six left (treated) and six right (untreated) knees in the high-dose group (Table 3) ( P < 0.05). All four measurements of the lateral part of the left femoral physis (total height, reserve zone, proliferative zone, and hypertrophic zone) showed significant increases with respect to the same measurements of the right femoral physis ( P < 0.05). Of particular note was the large difference between the mean total physeal height of the lateral part of the left femoral growth plate (1083.8 pm) and the right (530.7 pm). The means of the left middle femoral reserve zone (168.3 pm) and left medial femoral
Table 1 Differences of physeal height in the lateral, middle, and medial part of the growth plate between left and right limb in the therapeutic group (Wilcoxon test, n = 6) Physeal height
Right ( m e a n f SD, pm)
Left (mean fSD, pm)
Probability
Femur lateral Femur middle Femur medial Tibia lateral Tibia middle Tibia medial
468.9 f26.7 546.9 i97.9 549.2 f 106.1 513.6i57.5 548.4f 125 619.51 108
500.3 f 153.3 552.6 f91.9 530.3 f 55.1 540.8 f66.7 540.9 f 53.8 537.8 f66.3
0.62 0.92 0.65 0.36 0.89 0.05
Table 2 Significant differences of physeal height in three cartilaginous zones between therapeutic and high-dose on the left limb ( n = 6. P < 0.05) Cartilaginous zone
Low dose (0.5 Wlcm?), mean fSD (pm)
High dose (2.2 W/cm’), mean
Femur physeal heighr (lateral) Femur reserve zone (lateral) Femur reserve zone (middle)
500.3 f 153.3 123.4 f53.9 92.2 f 24. I
1083.8 f365.7 419.6f225.8 168.3 f 27.9
+ SD (pm)
Table 3 Comparison of differences ofphyseal height in three cartilaginous zones between treated and untreated knees in the high-dose group (n = 6. P < 0.05) Cartilaginous zone
Right knee (untreated), mean iSD (pm)
Left knee (treated), mean
Femoral Femoral Femoral Femoral Femoral Femoral
530.7+ 83.5 152.4 i66.3 193.9 81.1 184.0 f 79.4 90.0 i36.5 128.9 27.4
1083.8 i365.7 419.6 =k 225.8 336.3 156.3 361.9 322.9 168.3& 27.9 1 8 7 . 9 i 18.7
physeal height (lateral) reserve zone (lateral) proliferative zone (lateral) hypertrophic zone (lateral) reserve zone (middle) reserve zone (medial)
* *
* *
* SD (pm)
R. Lyon et a/. I Journal of Orthopuedic Researclz 21 (2003) 865-871
869
Fig. 5. Comparison of the change of the physeal plate between high dosage treatment (left picture) and control group (right picture) ( x I 0 ) : a disordered array of the cartilaginous cell in the proliferative zone on the treated side.
reserve zone (1 87.9 pm) of the high-dose group were also significantly increased over the right middle femoral reserve zone (90.0 pm) and right medial femoral reserve zone (128.9 pm) (P < 0.05). Morphological review: In the legs treated with therapeutic ultrasound, most cases showed normal microstructure, while very slight differences in the growth plates were seen in the femoral reserve zone. In two of the distal femoral physes, a small island of cartilage was observed infiltrating the ossifying bone of the epiphysis. This was not seen in any of proximal tibial growth plates. Variations of this type are not necessarily pathologic. Definitive differences were observed, most notably in the zone of proliferation for the rabbits with higher-dose ultrasound treatment. In five of six animals, the cell columns were disorganized rather than ordered (Fig. 5). These changes were more severe laterally than medially. In the hypertrophic zone, the major change was also on the lateral side of the physis. Five of six rabbits had a substantial increase in the thickness and number of cell layers, and one rabbit exhibited a cartilaginous infiltration of the ossifying trabecular bone of the epiphysis. The proximal tibial physes showed similar changes as above, but to a much less degree. We did not find bony fusion in the physis for either the femur or tibia as compared to contralateral side.
Discussion The original studies of the effects of ultrasound on growing bone only utilized high-intensity ultrasound (22.5 W/cm2) [3] and recent studies are mainly concerned with using ultrasound doses in the milliwatt range [6,7,13,14]. However, as this study does not follow these animals to physeal closure, it is unclear as to whether the
histological changes reported would cause significant growth disturbances following growth plate closure. Comparisons of the radiograph and histology between therapeutic dose of 0.5 W/cm2 and a high dose of 2.2 W/cm* are distinct. Our data suggest that therapeutic doses of ultrasound do not cause microscopic damage to the physis of the knee joint, although there was one case showing an unclear growth plate lines radiographically. There are mild histological differences in the distal femoral reserve zone as seen in both therapeutic and high dose group, where cartilaginous cells infiltrated into the ossifying bone of the epiphysis. This occasional protrusion of cartilaginous cells can be a normal finding [ 101. Small interdigitations of cartilage termed mammillary processes extend into the metaphyseal bone caused by shearing forces. And they also extend into the epiphyseal bone, which subjected primarily to tensile forces. Radiographically, there were no significant differences in length or metaphyseal width between the low and high dose groups, nor were there significant differences in these measurements between the treated and control legs in either dosage group. Conversely, the high-dose ultrasound caused marked changes to the knee joint reviewed in radiographs, including some of morphologic changes in the lateral metaphysis, flattening of the femoral epiphysis, wedging of the tibial plateau, and radiolucent area on the metaphysis. These findings duplicate the studies by Forest et al. [3] who found that exposing rabbit and dog knees to (2 Wlcm') of ultrasound for 5 min a day caused marked sclerosis and physeal widening. Although we detected an indistinct physeal line in three of six physeal plates in radiographs, our histological review did not find any early bony fusion across the growth plate and only presented severe disarray of the cell columns in the proliferative zone. Changes of the microstructure in the lateral aspect of the distal femur are worse than those in
870
R. Lyon et al. I Journal
of
Orthopaediic Research 21 (2003) 865-871
the middle and medial aspect. The lateral aspect of the femur will have more absorption than on the medial side, due to a direct contact of the ultrasound applicator on the fixed lateral aspect of the distal femur or tibia. The majority of the current literature has focused on the effects of much lower intensities of ultrasound on bone growth and fracture repair [6,7,13,15]. Our study uses a therapeutic dose roughly 10 times the intensity of the ultrasound currently being investigated to accelerate fracture repair [6,7,15]. A study by Spadaro et al. [13] demonstrated that application of 30 mW/cm2 ultrasound to growing knee joints in rats showed no significant changes in long bone growth or physeal histology. Studies by three different institutes [6,7,15] used this same 30 mW/cm2 dose to measure its effects on fracture healing. Heckman et al. [6] found that this dose accelerated healing of tibial diaphyseal fractures in humans. Kristiansen et al. [7] and Wang et al. [15] also demonstrated accelerated healing times of distal radius and femoral fractures using the 30 mW/cm2 dose. A study by Wiltink et al. [16] found an increase in proliferative zone length as well as overall metatarsal length in fetal mice, using an ultrasound dose of 0.77 W/cm2. Our study does not demonstrate increases in any of the three physeal zones at the therapeutic dose of 0.5 W/cm2. However, knees exposed to the high dose ultrasound did show significant height changes in all three zones, most notably the proliferative and the reserve zone of cartilage, in which there was a nearly 1.7fold increase, and 2.8-fold increase. respectively. Meanwhile radiograph confirmed that there was an increased radiolucent area on the lateral metaphysis of the distal femur, which may result from bone resorption by ultrasound effects. However, it is hard to accurately predict the growth pattern of the growth plate within six weeks of the ultrasound treatment. It is unclear as to whether the histological changes reported will lead to significant growth disturbances and ultimately to growth plate closure. Further studies are needed to establish the longer-term follow-up of ultrasound on the rabbit physis. The exact mechanisms underlying these changes by ultrasound are still unclear. A study by Mortimer et al. [9] found that ultrasound doses of 0.5 W/cm2 increased Ca uptake in fibroblasts. Whether this mechanism exists in chrondrocytes and/or results in the pathological changes in the growth plate remains a question. It is possible that this mechanism or a physiologically similar one could contribute to both the increase in length and/ or the marked disarray of the cell columns seen in the high dose group. In our high-dose ultrasound group (1 MHz and 2.2 W/cm2), we found skin burning on the contact area by the applicator after one or two weeks of ultrasound treatment, which could be caused by increased temperature and repeated placement of the transducer. A treatment time of 20 min may not be the
main factor for skin burning, because we used a pulsed ultrasound mode. Most literature recommend either 10 min for continuous mode or 20 min for a pulsed mode. However, Boucaud et al. [l] exposed lower frequency of ultrasound (20 kHz) with 2.5 W/cm’ and continuous mode on hairless rats for 10 min. They found slight erythema and the delayed necrosis of the dermis and epidermis 24 h after sonication. They also placed a plastic film between animal skin and transducer; biologic effects on the rat skin are similar to those induced without the presence of plastic film. These findings indicated that the ultrasound-induced necrosis mechanism is a nonthermal reaction (radiation forces), and may typically arises from cavitations defined as the oscillation of gas bubbles under acoustic streaming [1,5]. Fresh full-thickness human skin was also used to examine histological changes in vitro after 2.5 W/cm2 exposure [I]. Scanning electron microscopy revealed no modification of the skin surface, whereas the use of 5.2 Wkm2 in pulsed mode on the human skin resulted in epidermal detachment and edema of the upper dermis. As to whether the histological changes on the growth plate were caused by heat generated by the ultrasound, we think the bone or cartilage changes are mainly due to ultrasound mechanism. Heating would mostly affect the superficial soft tissue. We cannot rule out that at the beginning of the treatment period the heat effect may increase vasculature and lead to some stimulation of physeal growth. In the late treatment period repeated heating will damage the arterial supply to the epiphysis. Stark et al. [14] found that the 7 and 12 h of ischemia had severe growth disturbances and produced apparent growth plate infarction. Further evaluation into heat produced in tissues in the region of ultrasound treatment is needed to resolve this issue.
Conclusion Ultrasound has been used to treat different pathological conditions. Its use around the joints of children with open physes remains controversial. It is our impression that high dose ultrasound appears to have a marked toxic effect on the rabbit femoral and tibial growth plate and pathological changes to the metaphysis and epiphysis. Therapeutic ultrasound (0.5 W/cm2) does not appear to significantly affect the physeal growth of long bones within the period of six weeks. Further study will be needed to determining longer term effects of ultrasound on growing bone.
Acknowledgements The authors thank Dr. J. Toth, and Ms. Inta Lacitis, Biomaterial Lab., for technique support. The authors
R. Lyon et al. I Journal of Orthopardic Research 21 (2003) 8 6 5 4 7 1
also express the gratitude to Mettler Electronics Inc., Anaheim, CA, for a loan of Sonicator 730. References [l] Boucaud A, Montharu J, Machet L, Arbeille B, Machet M, Patat F, Vaillant L. Clinical, histological, and electron microscopy study of skin exposed to low-frequency ultrasound. Anatomical Rec 2001 :264:114-9. [2] Cook SD, Ryaby JP, McCabe J, Frey JJ, Heckman JD, Kristiansen TK. Acceleration of tibia and distal radius fracture healing in patients who smoke. Clin Orthop Re1 Res 1997;337:198-207. [3] DeForest RE, Herrick JF, Janes JM, Krusen FH. Effects of ultrasound on growing bone. Arch Phys Med 1953;34:21-31. [4] Dyson M. Therapeutic applications of ultrasound. In: Nyborg WL, Ziskin MC, editors. Biologic effects of ultrasound. New York: Churchill Livingstone: 1985. p. 157-67. [5] Dyson M. Mechanisms involved in therapeutic ultrasound. Physiotherapy 1987;73(3):116-20. [6] Heckman JD, Ryaby JP, McCabe J, et al. Acceleration of tibia1 fracture healing by non-invasive, low-intensity pulsed ultrasound. J Bone Joint Surg 1994;76A(1):26-34. [7] Kristiansen TK, Ryaby JP, McCabe J, et al. Accelerated healing of distal radius fractures with the use of specific, low-intensity ultrasound. J Bone Joint Surg 1997;79A(7):961-73.
871
[8] Maylia E, Nokes LD. The use of ultrasound in orthopaedics. Techno1 Health care 1999;7:1-28. [9] Mortimer AJ, Dyson M. The effect of therapeutic ultrasound on calcium uptake in fibroblasts. Ultrasound Med Biol 1998;l4:499506. [lo] Ogden JA. In: Ogden J, editor. Skeletal injury in the child. 3rd ed. New York: Springer-Verlag; 2000. p. 10-3. [ l l ] Pilla AA, Mont MA, Nasser PR, Khan SA, Figueiredo M, Kaufman JJ, Siffert RS. Non-invasive low-intensity pulsed ultrasound accelerates bone healing in the rabbit. J Orthop Trauma 1990;4:24653. [12] Rubin C, Bolander M, Ryaby J, Handjiargyrou M. The use of low-intensity ultrasound to accelerate the healing of fractures. J Bone Joint Surg 2001;83A(2):259-70. [13] Spadaro JA, Albanese SA. Application of low-intensity ultrasound to growing bone in rats. Ultrasound Med Biol 1998; 24(4):567-73. [14] Stark RH, Matloub HS, Sanger JR, Cohen EB, Lynch K. Warm ischemic damage to the epiphyseal growth plate: A rabbit model. J Hand Surg 1987;12(1):5&61. [15] Wang SJ, Lewallen DG, Bolander ME, Chao EY, Ilstrup DM, Greenleat JF. Low intensity ultrasound treatment increases strength in a rat femoral fracture model. J Orthop Res 1994: 12:40-7. [16] Wiltink A, Nijweide PJ, Ooseterbaan WA, et al. Effect of therapeutic ultrasound on endochondral ossification. Ultrasound Med Biol 1995;21:561-8.
Journal of Orthopaedic Research
Journal of Orthopaedic Research 21 (2003) 865-871
www.elsevier.com/locate/ort hres
The effects of therapeutic vs. high-intensity ultrasound on the rabbit growth plate Roger Lyon *, Xue C. Liu, Joshua Meier Dqiurtnwnt of Orthopedic Surgery, Children’s Hospital of Wisconsin, Medical College of Wisconsin, 8701 Wutertowm Plunk R o d , Suite 3018. Milwuukee, WI 53226, U S A Received 14 June 2002; accepted 6 February 2003
Abstract
Six of 6-week-old NZ rabbits underwent ultrasound treatment using a therapeutic dose (0.5 W/cmZ)and other six were treated with a higher dose (2.2 W/cm2) to the lateral aspect of the left knee joint for 20 min per day and a total of six weeks. The right knee joint served as a control. The goal of this study is to see if the therapeutic dose and high dose (approximately 45-fold therapeutic dose) will have toxic effects on the physis. Histological review appeared normal growth plate in the therapeutic group. In the high dosage group three of six cases displayed flattening of the distal femoral epiphysis and wedging of the proximal tibial plateau and indistinct growth plate lines. It is of interest to note that there are opening radiolucent area in the lateral aspect on the femoral metaphysis in five of six cases, where bone resorption has taken place. Histological results showed that there are disordered arrays of the cartilaginous cells in the proliferative zone. The height of the lateral physis in the high dose group is not only greater than that in the therapeutic dose (1083.8 vs. 500.3 pm), but also greater than that in their contralateral control (530.7 pm)( P < 0.05). This shortterm study demonstrates that high dose ultrasound has profound pathologic effects in growing bone. Therapeutic doses of ultrasound do not have an adverse effect on bone growth in the short-term follow-up. 0 2003 Orthopaedic Research Society. Published by Elsevier Ltd. All rights reserved. Kejwwrls: Ultrasound; Histology; Growth plate; Toxic effect; X-ray
Introduction
Ultrasound as a therapeutic treatment has often been used by physical therapists to decrease joint stiffness, to reduce pain and muscle spasm, and to improve muscle mobility [4,8]. Most of the studies related to musculoskeletal applications of very low-intensity doses of ultrasound have been focused on outcomes in fracture healing [6,7,13]. A low-intensity ultrasound (1.5 MHz and 30 mW/cm2) was investigated as an accelerator of cortical and cancellous bone fracture healing in smokers and nonsmokers [2]. They found that the healing time for a tibial and distal radius fracture was reduced 41% in smokers and 26% in nonsmokers, and for distal radius fracture was reduced by 51% in smokers and 34% in nonsmokers following the ultrasound treatment. In a placebo-controlled study of mid-shaft tibial osteotomies in rabbits Pilla et al. [ l l ] found that brief periods (20 *Corresponding author. Tel.: +1-414-456-4685; fax: +1-414-4566541. E - m d nddress: [email protected] (R. Lyon).
midday) of pulsed ultrasound accelerated the recovery of torsional strength and stiffness, which was also confirmed by a later study [ 151. Physical therapists use a therapeutic dosage of ultrasound (approximately 0.5 W/cm2) to treat patients with overuse of soft tissue injuries and muscle spasm around the thigh, knee, and back for 5-10 min per spot. The most current study in the literature was done by Spadaro and Albanese in 1998 [13]. They studied the effects of low-intensity ultrasound on bone growth in rats. An ultrasound dose of 30 mW/cm2, which is the routine dose to stimulate fracture repair, was directed at the knee joints of the rats. The researchers found no length differences compared to the control group as well as no significant differences in bone mineral densities of the tibia and femur. The authors concluded that physeal growth is much less affected by this level of ultrasound than is fracture repair [13]. However, doses of ultrasound vary among the different clinical application. Therapeutic ultrasound is still not widely recommended in pediatric patients due to possible disturbances and damage to the physeal plate. In reference to high-intensity of
0736-0266/$ - see front matter 0 2003 Orthopaedic Research Society. Published by Elsevier Ltd. All rights reserved. doi: 10.I0 16/S07~6-0266(03)00047-0
866
R. Lyon et ul. I Journal of’ Orthopuedic Reseurch 21 (2003) 865-871
ultrasound (about 2.0 W/cm2), a study in 1958 reported severely damaged growing bone and joint destruction, including sclerosis of the bone, widening, and displacement of the physis [3]. The use of ultrasound as a surgical instrument involves even higher intensity (5-300 W/cm2), including to fragment calculi, to initiate the healing of nonunion, to ablate disease tissues, and even to remove cement during revision of prosthetic joints [12]. So far there have been few published papers documenting the effects of the different dosages of ultrasound on the growth plate in animal models, and none comparing the biological discrepancies between therapeutic and higher-intensity ultrasound on the growth plate. A more through understanding of the morphologic and histologic response of growing bone to ultrasound is needed. Is there a dose response curve of skeletal growth to ultrasound therapies? Currently the therapeutic application of ultrasound in our institution is 0.5 W/cm2. The maximum output allowed on the current equipment is 2.2 W/cm2. What would the effect of this high dose be on growing bone if an inadvertant exposure of this maximal dose occurred? Establishing a safe range of ultrasound treatment in the pediatric population is important to assure patient safety. We hypothesize that current therapeutic doses of ultrasound (0.5 W/cm2)will not cause degeneration of the cartilaginous cells in the growth plate, while highintensity doses (2.2 W/cm2) will result in degeneration of the cartilaginous cells of the growth plate, and further leads to bony deformities. The aim of this study is to determine if therapeutic and high-intensity doses of ultrasound will have toxic effects on the physeal plates in the rabbit knee joint via radiography and histology as following: (1) Longitudinal lengths of the physis in the microscope, the length and width of the femur or tibia in the X-ray, will be compared between the therapeutic and high dose groups, as well as between the treated vs. untreated contralateral group; (2) comparison of three cartilaginous cell zone between therapeutic and higher-intensity ultrasound will be performed.
inette covering, and two hind limb holes. The hind limbs were through the hole and fastened with straps. N o anesthesia was required. After a coupling gel was applied to the skin (Aquasonic 100, Parker Laboratories, Fairfield, NJ), the Ultrasound transducer was then held against the lateral side of the left hind knee centered at the joint line. The right hind knee is a control group. The target area included the distal femoral and proximal tibial growth plates. The transducer was held in place throughout the treatment period. The ultrasound output was set on pulse outflow, which is the preferred method in the clinical setting. Plain X-rays including anterior-posterior (AP) and lateral views were taken of both the right and left hind legs prior to treatment and at three weeks, and six weeks of the treatment. Plain radiographs were assessed by the following measurements: (1) Femoral length, measured from the superior aspect of the femoral head to the most distal aspect of the femoral condyle; (2) tibial length, measured from the most proximal aspect of the tibial plateau to the most distal aspect of the tibia; (3) femoral width, measured at the widest part of the distal femoral epiphysis; (4) tibial width, measured at the widest part of the proximal tibial epiphysis. Plain and Faxitron X-rays were also examined for morphological changes, such as changes in bone density and general integrity of the bony structure. Following the six weeks treatment period, the animals were sacrificed. The left and right hind legs of the rabbit were dissected out and immediately frozen at -35 “F to preserve tissue integrity. Faxitron X-rays (AP and lateral views) were then taken of both left and right hind legs. Faxitron X-ray provides clearer image than standard X-ray, because it is executed in 50 kV and 1 min. The standard X-ray uses 50 kV, 6 mA and 1 s for exposure. The right and left legs were then cut down and bisected into anterior and posterior halves on a bandsaw and placed in formalin. Following preservation, the joints were placed into an ethanol solution and then decalcified in EDTA. After being decalcified, the joints were embedded in paraffin, cut, and placed on glass slides. Six slides were made out of both the anterior and posterior halves of each knee joint. Three were stained with a standard H & E stain, the other three stained with toluidine blue. The slides were then examined by light microscopy, using an image analyzer (Image Pro Plus, Cybermedics, Siverspring, M D ) to obtain computerized micrographs. These micrographs were then measured using the image analyzer. The following measurements were performed (Fig. l): ( I ) Measurement of the total physeal height, measured from the most proximal to the most distal points of intact cartilage, in both the distal femoral and proximal tibial growth plates; (2) measurement of the reserve zone of cartilage, measured from the most proximal point of intact cartilage to the beginning of the first cell column of proliferating cells in the plane of measurement; ( 3 ) measurement of the proliferative zone of cartilage, measured from the first cell in a pro-
Methods 12 New Zealand white rabbits were divided into two equal groups of six rabbits each. The first group (rabbits 1-6) received the therapeutic dose (0.5 Wlcm’) of ultrasound, while the second group (rabbits 7-12) received the high-intensity dose (2.2 W/cm*). All 12 rabbits were treated with the ultrasound for 20 min a day on the fixed point of lateral knee for a total period of six weeks. A standard medical ultrasound generator and transducer head (Sonicator 730, Mettler Electronics. Anaheim, CA) was used in the ultrasound treatments. This unit has a frequency of 1.0 MHz, with an effective radiating area of 1.8-8.5 crn’. A 2 cm’ crystal applicator was used, with a range of pulsed or continuous output power density from 0.1 to 2.2 W/cm’. The ultrasound was applied with the rabbits held in the prone position. The rabbits were kept in a custom-made wooden frame, comprised of a platform, a body Velcro strap, a head and neckstock-
Fig. 1. Measurements of the height of the physeal plate using the image analyzer: the reserve zone (R), proliferative zone (P), and hypertrophic zone (H) ( x 10, pm).
R. Lyon et al. I Journal
of
Orthopedic Research 21 (2003) 865-871
867
liferating cell column to the first hypertrophic cell in the plane of measurement: (4) measurement of the hypertrophic zone of cartilage, measured from the first hypertrophic cell to the last point of intact cartilage in the plane of measurement. These measurements in three cartilaginous cell zones were done in three areas of each growth plate: the most lateral part of the physis, the mid-line of the physis, and the most medial part of the physis. Mean value and its standard deviation of physeal length in both three cartilaginous zones and X-ray are computed. Using one-way ANOVA. we are able to compare differences of the length or width in the femur and tibia measured by X-ray and physeal height calculated by the image analysis system between the therapeutic and high dosage group. Similarly, using Wilcoxon matched pairs test, we compare them between treated side and untreated side. A possibility of less than 0.05 is considered as significant.
Results
Appearance: At the end of the treatment period, the animals remained in good health, and no gross deformities of limb growth were noted. All six high-dose rabbits received second-degree full thickness skin burns around the area of ultrasound application. These burns were seen between 5 and 15 days after the start of treatment. At the beginning they were presented with slight erytheme, and epidermal detachment, then they were appeared as necrosis of the epidermal. These sores uneventfully covered by dry skin eschar in black and the skin eschar gradually receded at study completion, indicative of being in healing process. There was no gross skin pathology noted on any of the six therapeutic dose rabbits. Faxitroii X-ray analysis: Faxitron X-ray analysis showed marked differences between therapeutic (0.5 Wlcm2) and high dose groups (2.2 Wlcm2). None of six cases in the therapeutic group showed morphologic changes. Both the distal femoral and proximal tibial growth plates of the left (treated) and the right (untreated) legs remained open, demonstrating active longitudinal bone growth. The left physis showed no evidence of either early or delayed closure when compared to the right. However, in 1 of 6 cases the left knee showed an indistinct area of the growth plate in the distal femur and proximal tibia when compared to the right side (Fig. 2). The high-dose group (2.2 Wlcm2), however, showed significant bony pathology in 5 distal femurs, 3 proximal tibias, and 1 proximal fibula. Those five rabbits had a radiolucent area in the lateral aspect of the femoral metaphysis adjacent to the growth plate in the AP view, probably resulting from bone resorption due to ultrasound effects (Fig. 3). Changes of contour in the lateralposterior metaphysis suggest marked bone resorption of the distal femoral metaphysis. Radiographic flattening of the lateral side of the distal femoral epiphysis was observed in three of the high-dose group (Fig. 4). Varus angulation of the distal femur due to bone resorption was seen in one high-dosed group as well (Fig. 4). Three
Fig. 2. In therapeutic group Faxitron X-ray shows normal bony anatomy in both the left and right knee, while only one illustrated unclear physeal line in the lateral aspect of the femur and tibia.
growth plate in the knee joint Fig. 3 . Comparison of morpho~ogica~ Faxitron X-ray in higher dose group. The left knee (left) demonstrated radiolucent area on the lateral metaphysis of the distal femur, which may result from bony resorption, while right knee kept intact.
of six rabbits showed indistinct physeal line in the lateral epiphysis of the distal femur. Three of six cases showed indistinct radiographic growth plate lines in the proximal tibia and wedging of the tibial plateau. The reminder demonstrated normal radiographic physeal characteristics. Routine X-ray: There were no significant differences of the length or width between left femurltibia and right femurltibia within either dosage groups, and no differences of the length and width in the femur or tibia between therapeutic and high dosage groups at the beginning, three weeks and six weeks of treatment ( P > 0.05). Histological measurements: A comparison of height in the three cartilaginous cell zone in the femoral and tibial physis in the therapeutic treated rabbits showed no
R . Lyon et 01. I Journal qf Orthopclrdic Reserirc11 21 (2003) 865-871
868
Fig. 4. Comparison of morphological growth plate in the knee joint X-ray in high dose group. Left knee (treated) shows a flatten of condyle and varus of the distal femur. The tibia became wedging and partial sclerosis on the lateral side with indistinct physeal line. The right knee (right) shows normal contour of the knee joint with open growth plate.
statistically significant differences between six left (treated in 0.5 W/cm2)and right (untreated) knees ( P > 0.05). Table 1 shows that overall physeal height in lateral,
medial, and middle part of the femur and tibia pose no significant differences. Significant differences were found in comparing the treated knees of the therapeutic and high-dose groups (Table 2) ( P < 0.05). The mean height of the lateral part of the left femur growth plate in the therapeutic treated rabbits was 500.3 pm, while the mean in the high dose group was 1083.8 pm. Likewise, the mean height of the middle of the femoral reserve zone in the therapeutic treated group was 92.2 pm, while this mean value in the high dose group was 168.3 pm ( P < 0.05). Measurements were also markedly increased in the high-dose group compared to the therapeutic group with respect to the left femoral physeal height in lateral reserve zone. Significant differences were also found comparing the six left (treated) and six right (untreated) knees in the high-dose group (Table 3) ( P < 0.05). All four measurements of the lateral part of the left femoral physis (total height, reserve zone, proliferative zone, and hypertrophic zone) showed significant increases with respect to the same measurements of the right femoral physis ( P < 0.05). Of particular note was the large difference between the mean total physeal height of the lateral part of the left femoral growth plate (1083.8 pm) and the right (530.7 pm). The means of the left middle femoral reserve zone (168.3 pm) and left medial femoral
Table 1 Differences of physeal height in the lateral, middle, and medial part of the growth plate between left and right limb in the therapeutic group (Wilcoxon test, n = 6) Physeal height
Right ( m e a n f SD, pm)
Left (mean fSD, pm)
Probability
Femur lateral Femur middle Femur medial Tibia lateral Tibia middle Tibia medial
468.9 f26.7 546.9 i97.9 549.2 f 106.1 513.6i57.5 548.4f 125 619.51 108
500.3 f 153.3 552.6 f91.9 530.3 f 55.1 540.8 f66.7 540.9 f 53.8 537.8 f66.3
0.62 0.92 0.65 0.36 0.89 0.05
Table 2 Significant differences of physeal height in three cartilaginous zones between therapeutic and high-dose on the left limb ( n = 6. P < 0.05) Cartilaginous zone
Low dose (0.5 Wlcm?), mean fSD (pm)
High dose (2.2 W/cm’), mean
Femur physeal heighr (lateral) Femur reserve zone (lateral) Femur reserve zone (middle)
500.3 f 153.3 123.4 f53.9 92.2 f 24. I
1083.8 f365.7 419.6f225.8 168.3 f 27.9
+ SD (pm)
Table 3 Comparison of differences ofphyseal height in three cartilaginous zones between treated and untreated knees in the high-dose group (n = 6. P < 0.05) Cartilaginous zone
Right knee (untreated), mean iSD (pm)
Left knee (treated), mean
Femoral Femoral Femoral Femoral Femoral Femoral
530.7+ 83.5 152.4 i66.3 193.9 81.1 184.0 f 79.4 90.0 i36.5 128.9 27.4
1083.8 i365.7 419.6 =k 225.8 336.3 156.3 361.9 322.9 168.3& 27.9 1 8 7 . 9 i 18.7
physeal height (lateral) reserve zone (lateral) proliferative zone (lateral) hypertrophic zone (lateral) reserve zone (middle) reserve zone (medial)
* *
* *
* SD (pm)
R. Lyon et a/. I Journal of Orthopuedic Researclz 21 (2003) 865-871
869
Fig. 5. Comparison of the change of the physeal plate between high dosage treatment (left picture) and control group (right picture) ( x I 0 ) : a disordered array of the cartilaginous cell in the proliferative zone on the treated side.
reserve zone (1 87.9 pm) of the high-dose group were also significantly increased over the right middle femoral reserve zone (90.0 pm) and right medial femoral reserve zone (128.9 pm) (P < 0.05). Morphological review: In the legs treated with therapeutic ultrasound, most cases showed normal microstructure, while very slight differences in the growth plates were seen in the femoral reserve zone. In two of the distal femoral physes, a small island of cartilage was observed infiltrating the ossifying bone of the epiphysis. This was not seen in any of proximal tibial growth plates. Variations of this type are not necessarily pathologic. Definitive differences were observed, most notably in the zone of proliferation for the rabbits with higher-dose ultrasound treatment. In five of six animals, the cell columns were disorganized rather than ordered (Fig. 5). These changes were more severe laterally than medially. In the hypertrophic zone, the major change was also on the lateral side of the physis. Five of six rabbits had a substantial increase in the thickness and number of cell layers, and one rabbit exhibited a cartilaginous infiltration of the ossifying trabecular bone of the epiphysis. The proximal tibial physes showed similar changes as above, but to a much less degree. We did not find bony fusion in the physis for either the femur or tibia as compared to contralateral side.
Discussion The original studies of the effects of ultrasound on growing bone only utilized high-intensity ultrasound (22.5 W/cm2) [3] and recent studies are mainly concerned with using ultrasound doses in the milliwatt range [6,7,13,14]. However, as this study does not follow these animals to physeal closure, it is unclear as to whether the
histological changes reported would cause significant growth disturbances following growth plate closure. Comparisons of the radiograph and histology between therapeutic dose of 0.5 W/cm2 and a high dose of 2.2 W/cm* are distinct. Our data suggest that therapeutic doses of ultrasound do not cause microscopic damage to the physis of the knee joint, although there was one case showing an unclear growth plate lines radiographically. There are mild histological differences in the distal femoral reserve zone as seen in both therapeutic and high dose group, where cartilaginous cells infiltrated into the ossifying bone of the epiphysis. This occasional protrusion of cartilaginous cells can be a normal finding [ 101. Small interdigitations of cartilage termed mammillary processes extend into the metaphyseal bone caused by shearing forces. And they also extend into the epiphyseal bone, which subjected primarily to tensile forces. Radiographically, there were no significant differences in length or metaphyseal width between the low and high dose groups, nor were there significant differences in these measurements between the treated and control legs in either dosage group. Conversely, the high-dose ultrasound caused marked changes to the knee joint reviewed in radiographs, including some of morphologic changes in the lateral metaphysis, flattening of the femoral epiphysis, wedging of the tibial plateau, and radiolucent area on the metaphysis. These findings duplicate the studies by Forest et al. [3] who found that exposing rabbit and dog knees to (2 Wlcm') of ultrasound for 5 min a day caused marked sclerosis and physeal widening. Although we detected an indistinct physeal line in three of six physeal plates in radiographs, our histological review did not find any early bony fusion across the growth plate and only presented severe disarray of the cell columns in the proliferative zone. Changes of the microstructure in the lateral aspect of the distal femur are worse than those in
870
R. Lyon et al. I Journal
of
Orthopaediic Research 21 (2003) 865-871
the middle and medial aspect. The lateral aspect of the femur will have more absorption than on the medial side, due to a direct contact of the ultrasound applicator on the fixed lateral aspect of the distal femur or tibia. The majority of the current literature has focused on the effects of much lower intensities of ultrasound on bone growth and fracture repair [6,7,13,15]. Our study uses a therapeutic dose roughly 10 times the intensity of the ultrasound currently being investigated to accelerate fracture repair [6,7,15]. A study by Spadaro et al. [13] demonstrated that application of 30 mW/cm2 ultrasound to growing knee joints in rats showed no significant changes in long bone growth or physeal histology. Studies by three different institutes [6,7,15] used this same 30 mW/cm2 dose to measure its effects on fracture healing. Heckman et al. [6] found that this dose accelerated healing of tibial diaphyseal fractures in humans. Kristiansen et al. [7] and Wang et al. [15] also demonstrated accelerated healing times of distal radius and femoral fractures using the 30 mW/cm2 dose. A study by Wiltink et al. [16] found an increase in proliferative zone length as well as overall metatarsal length in fetal mice, using an ultrasound dose of 0.77 W/cm2. Our study does not demonstrate increases in any of the three physeal zones at the therapeutic dose of 0.5 W/cm2. However, knees exposed to the high dose ultrasound did show significant height changes in all three zones, most notably the proliferative and the reserve zone of cartilage, in which there was a nearly 1.7fold increase, and 2.8-fold increase. respectively. Meanwhile radiograph confirmed that there was an increased radiolucent area on the lateral metaphysis of the distal femur, which may result from bone resorption by ultrasound effects. However, it is hard to accurately predict the growth pattern of the growth plate within six weeks of the ultrasound treatment. It is unclear as to whether the histological changes reported will lead to significant growth disturbances and ultimately to growth plate closure. Further studies are needed to establish the longer-term follow-up of ultrasound on the rabbit physis. The exact mechanisms underlying these changes by ultrasound are still unclear. A study by Mortimer et al. [9] found that ultrasound doses of 0.5 W/cm2 increased Ca uptake in fibroblasts. Whether this mechanism exists in chrondrocytes and/or results in the pathological changes in the growth plate remains a question. It is possible that this mechanism or a physiologically similar one could contribute to both the increase in length and/ or the marked disarray of the cell columns seen in the high dose group. In our high-dose ultrasound group (1 MHz and 2.2 W/cm2), we found skin burning on the contact area by the applicator after one or two weeks of ultrasound treatment, which could be caused by increased temperature and repeated placement of the transducer. A treatment time of 20 min may not be the
main factor for skin burning, because we used a pulsed ultrasound mode. Most literature recommend either 10 min for continuous mode or 20 min for a pulsed mode. However, Boucaud et al. [l] exposed lower frequency of ultrasound (20 kHz) with 2.5 W/cm’ and continuous mode on hairless rats for 10 min. They found slight erythema and the delayed necrosis of the dermis and epidermis 24 h after sonication. They also placed a plastic film between animal skin and transducer; biologic effects on the rat skin are similar to those induced without the presence of plastic film. These findings indicated that the ultrasound-induced necrosis mechanism is a nonthermal reaction (radiation forces), and may typically arises from cavitations defined as the oscillation of gas bubbles under acoustic streaming [1,5]. Fresh full-thickness human skin was also used to examine histological changes in vitro after 2.5 W/cm2 exposure [I]. Scanning electron microscopy revealed no modification of the skin surface, whereas the use of 5.2 Wkm2 in pulsed mode on the human skin resulted in epidermal detachment and edema of the upper dermis. As to whether the histological changes on the growth plate were caused by heat generated by the ultrasound, we think the bone or cartilage changes are mainly due to ultrasound mechanism. Heating would mostly affect the superficial soft tissue. We cannot rule out that at the beginning of the treatment period the heat effect may increase vasculature and lead to some stimulation of physeal growth. In the late treatment period repeated heating will damage the arterial supply to the epiphysis. Stark et al. [14] found that the 7 and 12 h of ischemia had severe growth disturbances and produced apparent growth plate infarction. Further evaluation into heat produced in tissues in the region of ultrasound treatment is needed to resolve this issue.
Conclusion Ultrasound has been used to treat different pathological conditions. Its use around the joints of children with open physes remains controversial. It is our impression that high dose ultrasound appears to have a marked toxic effect on the rabbit femoral and tibial growth plate and pathological changes to the metaphysis and epiphysis. Therapeutic ultrasound (0.5 W/cm2) does not appear to significantly affect the physeal growth of long bones within the period of six weeks. Further study will be needed to determining longer term effects of ultrasound on growing bone.
Acknowledgements The authors thank Dr. J. Toth, and Ms. Inta Lacitis, Biomaterial Lab., for technique support. The authors
R. Lyon et al. I Journal of Orthopardic Research 21 (2003) 8 6 5 4 7 1
also express the gratitude to Mettler Electronics Inc., Anaheim, CA, for a loan of Sonicator 730. References [l] Boucaud A, Montharu J, Machet L, Arbeille B, Machet M, Patat F, Vaillant L. Clinical, histological, and electron microscopy study of skin exposed to low-frequency ultrasound. Anatomical Rec 2001 :264:114-9. [2] Cook SD, Ryaby JP, McCabe J, Frey JJ, Heckman JD, Kristiansen TK. Acceleration of tibia and distal radius fracture healing in patients who smoke. Clin Orthop Re1 Res 1997;337:198-207. [3] DeForest RE, Herrick JF, Janes JM, Krusen FH. Effects of ultrasound on growing bone. Arch Phys Med 1953;34:21-31. [4] Dyson M. Therapeutic applications of ultrasound. In: Nyborg WL, Ziskin MC, editors. Biologic effects of ultrasound. New York: Churchill Livingstone: 1985. p. 157-67. [5] Dyson M. Mechanisms involved in therapeutic ultrasound. Physiotherapy 1987;73(3):116-20. [6] Heckman JD, Ryaby JP, McCabe J, et al. Acceleration of tibia1 fracture healing by non-invasive, low-intensity pulsed ultrasound. J Bone Joint Surg 1994;76A(1):26-34. [7] Kristiansen TK, Ryaby JP, McCabe J, et al. Accelerated healing of distal radius fractures with the use of specific, low-intensity ultrasound. J Bone Joint Surg 1997;79A(7):961-73.
871
[8] Maylia E, Nokes LD. The use of ultrasound in orthopaedics. Techno1 Health care 1999;7:1-28. [9] Mortimer AJ, Dyson M. The effect of therapeutic ultrasound on calcium uptake in fibroblasts. Ultrasound Med Biol 1998;l4:499506. [lo] Ogden JA. In: Ogden J, editor. Skeletal injury in the child. 3rd ed. New York: Springer-Verlag; 2000. p. 10-3. [ l l ] Pilla AA, Mont MA, Nasser PR, Khan SA, Figueiredo M, Kaufman JJ, Siffert RS. Non-invasive low-intensity pulsed ultrasound accelerates bone healing in the rabbit. J Orthop Trauma 1990;4:24653. [12] Rubin C, Bolander M, Ryaby J, Handjiargyrou M. The use of low-intensity ultrasound to accelerate the healing of fractures. J Bone Joint Surg 2001;83A(2):259-70. [13] Spadaro JA, Albanese SA. Application of low-intensity ultrasound to growing bone in rats. Ultrasound Med Biol 1998; 24(4):567-73. [14] Stark RH, Matloub HS, Sanger JR, Cohen EB, Lynch K. Warm ischemic damage to the epiphyseal growth plate: A rabbit model. J Hand Surg 1987;12(1):5&61. [15] Wang SJ, Lewallen DG, Bolander ME, Chao EY, Ilstrup DM, Greenleat JF. Low intensity ultrasound treatment increases strength in a rat femoral fracture model. J Orthop Res 1994: 12:40-7. [16] Wiltink A, Nijweide PJ, Ooseterbaan WA, et al. Effect of therapeutic ultrasound on endochondral ossification. Ultrasound Med Biol 1995;21:561-8.