Journal of Ort hopaedic ELSEVIER
Research
Journal of Orthopaedic Research 22 (2004) 1086-1093
www.elsevier.com/locate/orthres
Bone mass is preserved in a critical-sized osteotomy by low energy pulsed electromagnetic fields as quantitated by in vivo micro-computed tomography Michael 0. Ibiwoye a, Kimerly A. Powell a, Mark D. Grabiner a,b, Thomas E. Patterson ', Yoshitada Sakai a, Maciej Zborowski a, Alan Wolfman c,l, Ronald J. Midura Department of Biomedical Engineering, Lerner Research Institute of The Cleveland Clinic Foundation, ND20, 9500 Euclid Avenue, Cleveland, O H 44195, USA School of KinesioloKy, University of Illinois at Chicago, Chicago, IL 60608, USA Depurtment of Cell Biology, Lernrr Reseurch Institute of The Cleveland Clinic Foundation, NDZO, Cleveland, OH 44195, U S A
Received 9 October 2003; accepted 30 December 2003
Abstract
The effectiveness of non-invasive pulsed electromagnetic fields (PEMF) on stimulating bone formation in vivo to augment fracture healing is still controversial, largely because of technical ambiguities in data interpretation within several previous studies. To address this uncertainty, we implemented a rigorously controlled, blinded protocol using a bilateral, mid-diaphyseal fibular osteotomy model in aged rats that achieved a non-union status within 3 4 weeks post-surgery. Bilateral osteotomies allowed delivery of a PEMF treatment protocol on one hind limb, with the contralateral limb representing a within-animal sham-treatment. Bone volumes in both PEMF-treated and sham-treated fibulae were assessed simultaneously in vivo using highly sensitive, high-resolution micro-computed tomography (pCT) over the course of treatment. We found a significant reduction in the amount of time-dependent bone volume loss in PEMF-treated, distal fibular segments as compared to their contralateral sham-treated bones. Osteotomy gap size was significantly smaller in hind limbs exposed to PEMF over sham-treatment. Therefore, our data demonstrate measurable biological consequences of PEMF exposure on in vivo bone tissue. 0 2004 Orthopaedic Research Society. Published by Elsevier Ltd. All rights reserved. Keywords; Fracture healing; Fractures non-united; Imaging three-dimensional; Bone resorption; Osteotomy
Introduction PEMF is often used as a non-invasive, post-operative treatment for augmenting vertebral fusion following posterolateral lumbar arthrodesis [ 17,201. PEMF is also used to treat recalcitrant, non-united long bone fractures [I ,3,9,26]. However, despite an overall positive clinical consensus, there still exists lingering doubt in the scientific literature regarding the effectiveness of PEMF to
Ahhreuiuiions; PEMF, pulsed electromagnetic field; pCT, microcomputed tomography; 11, image intensifier; ROI, region-of-interest; ICP, iterative closest point. *Corresponding author. Tel.: +I-216-445-3212; fax: +1-216-4454383. E-muil address:
[email protected] (R.J. Midura). These authors contributed equally to this work.
'
enhance bone formation [1,13,14]. Much of this uncertainty is ascribed to contradictory conclusions resulting from subjective data interpretation. For example, Kahanovitz et al. [I41 investigated the effect of PEMF on lumbar fusions in the dog based on a semi-quantitative assessment of two-dimensional (2D) planar radiographs and histological sections of the fusion mass. They found no PEMF-dependent acceleration of bone healing in this model system. In view of the fact that bone tissue changes due to PEMF stimulation may be very small and patchy, such changes may be difficult to evaluate accurately by relying only on conventional 2D planar radiography and histological sections. Major drawbacks of many previous studies include a poor assessment of PEMF treatment dose and subject compliance, an inadequate control for the large variations in natural physiology within a subject population,
0736-02666 - see front matter 0 2004 Orthopaedic Research Society. Published by Elsevier Ltd. All rights reserved. doi: 10.1016/j.orthres.2003.12.017
M . 0. Ibtwoye ei ul. I Journal of' Orihopaerlic Reseurrh 22 12004) 1086-IV93
the use of qualitative outcome measures for assessing bone volume accrual, and a lack of blinded and objective data assessment. Our experimental design corrects for these drawbacks in the following ways. First, it utilized bilateral fibular osteotomies, thereby providing a contralateral sham-treated control limb within each rat to reduce the influences of between-animal variations in natural physiology. Second, our study carefully monitored PEMF dosage and subject compliance. Third, it used a non-invasive imaging modality, in vivo microcomputed tomography (pCT), to simultaneously assess in both PEMF-treated and sham-treated limbs for each animal the longitudinal changes in bone volume at each individual non-union fracture site over a 10-week test period. Finally, all data analyses were done in a blinded fashion, thereby maintaining objectivity. In the current study, we sought to experimentally simulate the situation of patients who begin PEMF therapy after a diagnosis of a long bone non-union [3,9]. Accordingly, our objective in the present study was to longitudinally measure bone volume changes in nonunited osteotomies in vivo, either exposed to daily PEMF or sham-treatments. We hypothesized that PEMF treatment would enhance the quantity of bone at a nonunion osteotomy site over that of sham-treated counterparts.
Materials and methods Animals and operutive techniques All animal procedures used in this study were reviewed and approved by the Institutional Animal Care and Use Committee. Most experimental mid-diaphyseal fracture models usually involve stabilization of the osteotomized bones with external or intramedullary fixation [21]. Two principal considerations influenced our choice of the fibula in this study: the unique anatomical arrangement of the tibia/ fibula unit enabled us to create a large fibular defect without applying external splinting or internal fixation, while also ensuring non-union without morbidity or mortality. Furthermore, because of its unique anatomical location and biomechanical role, the rat tibia provides stability against the mechanical effects of the animal's weight and cage activities on the healing of fibular osteotomies. A total of 15 adult male Sprague Dawley rats (Harlan, USA) weighing 500 g were used in the two experiments of this study. In the first study, we established a standard protocol for a fibular non-union model by prospectively analyzing radiographic changes in bone healing using nine bilaterally osteotomized rats. Digital planar radiographs ( 2 0 4 8 ~ 2048 12-bit projections; Phoenix X-ray, Germany) in the lateral plane were recorded for both hind limbs 1 week before surgery to exclude any pre-existing anatomical abnormalities. Similar images were recorded 1 day after surgery t o verify the position and accuracy of bone resection, and thereafter on every third day to assess changes in osteotomy gap size. After establishing this necessary database to prove a non-union status at these osteotomy sites, we employed three-dimensional pCT imaging in a second study to increase the sensitivity and accuracy of quantifying subtle changes in bone volume and gap size following treatment of osteotomized fibulae with PEMF. The rats were housed in individual cages in the Central Animal Facilities, handled under identical conditions and allowed access to food and water ad libitum. Under general anesthesia with Nembutal (Abbott Laboratories, USA; 60 mg/kg, i.p.), the entire hind limb including the lower anterior abdominal wall and groin was shaved with fine-toothed hair clippers and disinfected with Betadine solution. Cefazolin (Abbott Laborato-
1087
ries. USA; 30 mg/kg, i.p.) was administered as prophylaxis immediately prior to surgery. The osteotomies were performed using a modification of a previously described procedure [I 51. Under sterile conditions, a lateral incision was made that extended vertically from immediately below the knee joint to a point just above the ankle joint. The deep fascia was incised and the lateral intermuscular septum was divided by blunt dissection t o separate the anterior and posterior compartment muscles and to gain access to the lateral surface of the fibula and its attached musculus peronei longus cv hr With the anterior and posterior muscle groups fully retracted and entire lateral fibula exposed, a bone segment (6.0 f 0.5 mm) was excised, complete with the periosteal covering, from the fibula mid-diaphysis using single, sharp cuts with a high-speed rotary saw-toothed blade (0.1 mm thick stainless steel; Fine Science Tools, USA). T o prevent heatinduced bone damage at the osteotomy site, the blade was pre-cooled to 4 "C with sterile physiological saline containing gentamicin (Sigma, USA; 50 pg/ml) and fungizone (Life Technologies, USA; 2.5 pg/ml). After irrigation with a jet of cold saline containing antibiotics as above, two to three drops of 0.5% Marcaine in isotonic saline (Abbott Laboratories, USA) were applied to the osteotomy site as a local anesthetic. and the skin was closed with 2-0 silk interrupted sutures (Ethicon Inc., USA). Finally, Bacitracin ointment (Clay Park Laboratories, USA: 500 units) was applied to the closed surgical wound. Planar radiographs as described above were used to confirm the accuracy and location of the osteotomy defect post-operatively. Osteotomized rats were allowed unrestricted cage activities but were examined daily and weighed twice weekly. Every rat in this study exhibited a slight decrease in body weight (-10%) during the initial 3 weeks in our animal facilities prior to surgery and P E M F treatment, from 502 k 37 to 459 2 30 g. Thereafter, body weights remained stable for the remaining 12 weeks of the protocol. This weight pattern is comparable with the growth curve for healthy, normal out-bred rat populations matched for strain, age and sex (Harlan, Indianapolis, IN). As expected for using antibiotic prophylaxis, there were no clinical signs of infection. Additionally. there were n o signs of malnutrition or debilitating pain.
PEMF ireutment of' the ruts For the second experiment but prior to osteotomy, six rats were acclimatized to the harnessing conditions used in this study. One week before surgery, each rat was placed into a sling-suit harness (Model RSFl extra large size, Lomir Biomedical, Inc., Malone, NY) after anesthesia induction with 1%)vol./vol. Isoflurane, and allowed to awake in the harness. Isoflurane was chosen, as it is a fast acting inhalation anesthetic that rapidly clears from the systemic system when its delivery is stopped. The rats were held in these harnesses for 3 h daily (see below) without P E M F treatment. Typically, after 2-3 days of harnessing, the rats were observed to have become accustomed to this procedure and no longer exhibited stressful behavior. The acclimatization protocol continued u p until the P E M F treatments were initiated at 28 days after surgery. Overall, this approach significantly enhanced compliance with the PEMF protocol throughout the treatment period. While under inhalation anesthesia with Isoflurane, both the left and right hind limbs from each rat were inserted through separate openings on each side of the sling-suit harness. Correct positioning of the legs was carried out by gentle skin traction over the back of the thigh and by nudging the knee joint through an aperture until the legs were fully extended outside the harness. The legs were prevented from being retracted back through these openings by using an adjustable tie-off fit snugly at the upper thigh just below the hip. This harnessing system situated each hind limb dangling downward o n either side of a centrally positioned barrier made of metal alloy having a low magnetic reluctance. Thus, the right hind limb from mid-thigh downward was placed within the electromagnetic field generated by the custom P E M F coil, while the left hind limb was dangling downward shielded from PEMF stimulation (Fig. I). Throughout the harnessing procedure, traction on leg muscles that could stimulate extraneous bone formation at the osteotomy site was avoided [4,10]. Moreover, the amount of limb manipulation during harnessing was the same for both the PEMF-treated (right) and sham (left) hind limbs. The rats were further secured in the harness by a Velcro zip over the back from the root of the neck to the root of the tail. The rats were exposed t o an FDA-approved PEMF waveform, Physio-Stim@ (Orthofix Inc., McKinney, TX), generated by a control
1088
M. 0. Ibiwoye et al. I Journal of' Orthopedic Research 22 (2004) 1086-1093
Fig. I . Exact placement of the hind limbs in the sling-suit harness to achieve sham (left)- and PEMF (right)-treatments in the same rat. unit powered by a 9 V direct current power source. The magnetic field waveform consists of bursts of triangular (saw-tooth) pulses having a pulse frequency of 3.8 kHz, a burst duration of 5.56 ms, and a burst on-off period of 67 ms. The resulting burst on-off frequency is 15 Hz. The magnetic field amplitude, approximately 2 mT, equivalent to 20 G, was distributed uniformly over the fracture volume by virtue of a specially shaped coil, The coil enclosed the volume of approximately 7.5 cm (height) by 5 cm (width) by 2.5 cm (depth). The rats were exposed to the PEMF waveform for 3 h a day, 7 days a week for 10 weeks. The waveform of each coil was routinely checked by a search coil [27] to ensure correct functioning of the individual coils and the efficacy of the barrier blocking PEMF penetration was also checked regularly (Infiniium Oscilloscope Model 54810A, Agilent Technologies, USA). Inspecting the positions of both hind limbs every 20 min assured protocol compliance. In viva pCT imaging of the rats While under pentobarbital anesthesia the animals were secured on their backs to the rotating platform of a custom designed pCT imaging system (K.A. Powell, unpublished data) and the left and right hind limbs were elevated vertically into the X-ray beam. The hind limbs were rigidly supported to eliminate motion artifacts during scanning. The animal's torso, head, and hind quarters were shielded with a 2 mm thick lead covering to limit the amount of X-ray exposure to other parts of the body. The average radiation dose as measured by several thin-layer dosimeters was -65 rads per imaging session, and each rat's hind limbs received a total of 1950 rads (1.95 Gy) at the fibular site for the entire period of investigation. The dose used in our aged rats is comparable to that employed in the study by Engstroni et al. [S] which showed that targeted exposure of 4-week-old Sprague Dawley rat tibiae to 2 Gy irradiation had no effect on normal epiphyseai growth of their bones. X-ray projection data (100 pm isotropic voxel resolution) of both the left and right fibulae and tibiae were obtained by collecting 360 2048x 2048 12-bit projection radiographs at 1" intervals around the entire specimen (circular data acquisition). These images were collected at 34 kV, 450 PA, and 1 s exposure time with the image intensifier (11) operating in 7 in. mode. A lead shuttering system was used to shield the animal from the X-ray beam during stage rotation thereby minimizing their X-ray dosage. The images were sub-sampled by a factor of 4 in both dimensions and pre-processed prior to 3D reconstruction. X-ray projection pre-processing consists of: (1) darkand white-field normalization, (2) ring artifact reduction, and (3) 2D unwarping to correct for 11 distortion. Three-dimensional reconstructions of the left and right hind limbs were obtained using a tent-FDK cone-beam reconstruction algorithm (K.A. Powell, unpublkhed data). Bone was segmented from the surrounding background signal using a global threshold. Identification of the outer boundaries of the segmented regions was automated and the boundary points from every
other frame of data in the 3D regions-of-interest (ROIs) were used for the spatial registration. An iterative closest point (ICP) method was used for registering the outer boundaries of the segmented regions (K.A. Powell, unpublished data). The ICP algorithm is a least-squares approach that minimizes the distances between sets of points. Singlevalue decomposition was used to calculate the transformation for minimizing the distances between paired points. The iterative procedure was stopped when the mean distances between all successive points did not change more than a fixed amount ( i t . , 0.5). Once the data transformation matrices were obtained, trilinear interpolation was used for the rigid body transformations of the original data sets. A ROI that included 113 of the distal segment of the fibula and the region of the tibia where the distal end joins the tibia was used for spatially registering the distal segment of the fibula. The fibular bone segments were further segmented by identifying a clipping plane in a 3D volume visualization program (Volsuite XX) and cropping the 3D images along this clipping plane. The clipping plane for the distal segment was positioned at the insertion point of the fibula into the tibia. These cropped data sets were then spatially oriented relative to their principal axis and minor manual adjustments were made to the clipping planes based on review of the principally oriented data sets. The nominal lengths of the proximal and distal segments were determined along the principal axis of the segment. The slice volumes from the fibular segments were calculated as a function of length by summing the segmented voxels in each cross-sectional slice orthogonal to the principal axis of the bone segment. The bone volume of the segment was calculated by summing all the cross-sectional slice volumes along the length of the bone [11,12]. The change in cross-sectional slice volume along the bone segment was measured relative to the first time point in the longitudinal imaging sequence. The precision and accuracy of this non-invasive imaging modality was confirmed by repeated imaging of some rats. The results from this repetitive test indicated a 1% error in registration, which would result in only a 5'%, error in calculation of normalized bone volume (K.A. Powell, personal communication). Histological assessments At the end of the 10-week treatment, the rats were euthanized with compressed carbon dioxide gas. The tibialfibula units from both hind limbs were removed and fixed overnight at 4 "C with 2% paraformaldehyde1Tris buffered saline (TBS), pH 7.5. The proximal and distal fibula segments were decalcified for 14 days at 4 "C in 400 mM EDTAl TBS. After decalcification, the specimens were cryoprotected overnight at 4 "C in 20% sucrose/TBS, flash frozen and then embedded In cryomedium (Tissue Freezing Medium, Triangle Biomedical Sciences, Durham, NC). Frozen sections were cut at 10 pm thickness and either stained immediately with hematoxylin and eosin and then mounted with Cytoseal XL for histological assessment, or stored at -80 "C unt~l required. Images were taken using a Nikon Microphot Fx microscope outfitted with a SPOT-RTCCD digital camera. Biostatistical analysis Wilcoxon signed rank tests were used to determine if the difl'erences observed for the dependent variables between the untreated control and PEMF treated limbs were significant. A maximum P-value of 0.05 was accepted to denote a significant difference.
Results A criticul-sized fibular osteotomy yields a long bone nonunion status
A non-union status in the 6 mm fibular osteotomies was confirmed by identifying the elapsed time, in days, after surgery when the average linear distance across the osteotomy gap was no longer shortening (Fig. 2A).
M.O. Ibiwoyr et al. I Journal of Orthopardic Research 22 (2004) 1086-1093
Fig. 2. A critical-sized osteotomy yields a long bone non-union status: (A) an average gap size was assessed in each bilaterally osteotomized rat by measuring from time-sequential planar radiographs the linear distance connecting four corresponding points on the periosteal and endosteal borders of the apposing proximal and distal osteotomy segments via two independent assessments (Bar = 10 mm), (B) cumulative gap closure first-order decay curve for both right and left fibula osteotomies together. Day 28 is highlighted as the time when PEMF treatment would begin.
There were similar exponential decreases in average gap size for both left and right fibulae over the first 14 days post-operatively (Fig. 2B). This suggests that the initial response of the fibula to this acute bone loss is an attempt to fill the osteotomy defect with new bone tissue. From day 21 onward, there were no further apparent changes in average gap size, suggesting the establishment of a long bone non-union (Fig. 2B). Radiographic assessment of two rats 6 months post-surgery indicated no further increase in gap closure (data not shown). Based on these observations, we chose 28 days postsurgery as an appropriate time frame to initiate PEMF treatment, since we are confident of a non-union at this time. PEMF reduces jurther osteotomy gap widening in vivo
Our pCT imaging system was of sufficient spatial resolution to examine the time-dependent effects of daily PEMF exposure on the size of the osteotomy gap over a 10-week period (Fig. 3). Consistent with the findings in previous studies [5,24], our radiographic results indicated a partial and uneven bone formation response after osteotomy that became a non-union situation by 21 days (Fig. 2B); this response pattern resembled “stalactites projecting down from a cave ceiling”. Therefore, the maximal length to which the osteotomy gap was filled in at each osteotomy site could only be objectively compared by measuring the minimal distance between the furthermost extended bony ends of the proximal and distal fibular segments (Fig. 4A). The change in minimum gap distance observed in the PEMF-treated limbs was better than that of their respective sham controls (Fig. 4C). In half of the animals, a narrowing in this distance across the osteotomy gap was detected in PEMF-treated fibulae over the 10-
1089
Fig. 3. In vivo KCT imaging generates high quality bone volume reconstruction and segmentation from the surrounding non-osseous tissues. Shown is a posterior view of the entire tibia and fibular bones from both left (L) and right (R) hind limbs of rat F at 1 and 10 weeks.
week test period (Fig. 4B). In these same animals, the sham-treated fibulae showed a sustained widening of this distance across the osteotomy gap (Fig. 4B). Of the six rats investigated, four (67%) clearly resisted widening of the minimal gap size in the PEMF-treated fibula osteotomies as compared to their sham controls over the 10-week test period (Fig. 4B). After 10 weeks, the histological appearance of the bone tissue, the degree of vascularity, and the amounts of loose connective tissue and skeletal muscle surrounding the distal fibular non-union site were indistinguishable between the sham-treated (Fig. 5A) and PEMF-treated (Fig. 5B) specimens. In particular, both the sham- and PEMF-treated specimens exhibited signs of pitting or scalloping on the periosteal surface by 10 weeks suggestive of some resorptive activity.
PEMF reduces bone volume loss The pCT bone volume images were used to plot the cross-sectional area of each bone segment as compared to its nominal length, which revealed that, first, bone volume losses for both sham- and PEMF-treated specimens were greater within a 2 mm distance from the original osteotomy site as compared to those for regions of bone farther away (Fig. 6A and B). Second, the shortening in nominal length of the sham-treated distal fibular segment was typically greater than its matched PEMF-treated counterpart. All six rats showed varying amounts of bone loss in the distal segments of both the PEMF-treated and control distal fibula segments (Fig. 7A). On average, the bone loss observed at the distal fibula in the PEMFtreated limbs was significantly smaller than that of the sham controls (Fig. 7B). Bone loss in the control group varied from 16% to 76% of the original volume, whereas
1090
M . 0. Ibhvoye et ul. I Journul qf Orthopuedic Reseurch 22 (2004) 1086-1093
Fig. 4. P E M F treatment reduces the widening of osteotomy gap size in vivo: (A) fibula volume reconstructions (in vivo pCT) were precision registered and aligned to permit highly accurate measurements of subtle changes in minimal osteotomy gap size over a 10-week P E M F treatment period. Imaging resolution = 0.1 mm voxel, (B) PEMF-dependent narrowing of minimal osteotomy gap size is apparent in rats C, E and F. The change in minimal gap size over the 10-week period was computed using the formula: (L,,) - L I ) / L I 100'%~; where L , and L,o are the minimal gap lengths for weeks I and 10, respectively. A negative value indicates a narrowing, while a positive value indicates a widening of the minimal gap size, (C) PEMFexposed fibula osteotomy gaps are significantly narrower than their contralateral sham-treated counterparts.
fibular segments. The reasons for this failure are discussed below. However, user intervention to readjust the proximal fibular volume alignments nearer the osteotomy sites, though subjective, could be used to compare PEMF- and sham-treated specimens. On average, this manual analysis revealed that the sham-treated proximal fibular segments exhibited a bone volume loss of 18% k 12%, while the PEMF-treated specimens exhibited a loss of 8%+ 10%. The difference in these mean values was not significant by Wilcoxon signed rank test.
Discussion
Fig. 5. Histological evaluation of bone and surrounding non-osseous tissues in the distal fibulae. Shown is a representative mid-sagittal section of a large portion of the distal fibulae from rat F stained with hematoxylin and eosin: (A) sham-treated; (B) PEMF-treated; b, bone; p, periosteal soft tissue cap; m, muscle.
only 7-22'K loss was detected in the treated group over the 10-week period of PEMF exposure. Consistent with the data regarding the minimum osteotomy gap size in these same animals (Fig. 4C), every PEMF-treated distal fibula segment responded with superior preservation of bone volume as compared to their respective control counterparts (Fig. 7A). The program that was successful in the automated registration of the distal fibular segments described above was not successful in registering the proximal
Direct current electrical potentials and their associated electromagnetic fields are generated in bone tissue by mechanical stress, and are thought to influence the physiology of de novo bone formation and mineralization [2,7]. This concept provides the rationale for treating bone traumas or pathologies with exogenous PEMF [3,19]. In these clinical situations, a bone fracture site is reduced and stabilized to promote a good bone fusion response. However, this immobilization procedure would abrogate any natural mechanical stressinduced electromagnetic fields that the bone would normally experience. It has been argued that in some cases this PEMF-depleted situation may lead to delayed or non-union fractures [8,18,23,25]. This logic forms the basis for the concept that exogenously supplied PEMF can substitute for suppressed native electromagnetic fields, and augment the healing of immobilized bone fractures [26], or vertebrae fusion after lumbar arthrodesk [1,2]. Despite this logic, the beneficial effects of PEMF on fracture healing remain controversial [ 1,13,14]. Many studies claim to demonstrate significant effects of PEMF on bone growth, but have failed to produce unequivocal
M. 0. Ihiiooyr et ul. / Journul of’Orthopuedic Reseurck 22 (2004) 1086-1093
1091
Fig. 6. Analysis of bone volume changes in the distal fibula segment. Shown are the plots of cross-sectional area slices (0.1 mm voxel thickness) against the nominal length for the sham (A) and PEMF-treated (B) distal fibula segments of rat F over the 10-week test period.The gray portion between the two curves represents the time-dependent change in fibula volume across the entire bone length over the 10-week test period. The doubleheaded arrow in (A) indicates a substantial loss in the nominal length of the sham-treated control, which is not observed for the corresponding PEMF-treated distal fibula segment. The change in bone volume over the 10-week period was computed using the formula: (I/i,, - I/i)/K x 100’%1; where for each bone segment V, and V,, are the summed cross-sectional areas (i.e., volume) for weeks I and 10, respectively. Thus, for rat F, the PEMF-treated distal fibula lost 13.1%) of initial bone volume [ = (91.41 mm’ - 105.16 mm2)/105.16 mm’x lOO’%].while the contralateral sham-treated distal fibula lost 45.3% of initial bone volume [ = (31.49 mm’ - 57.61 mm’)/57.61 mm2 x loo%].
,
,A
,
c
0
A
B
C D E Rat Identification
1
m
-60
F
Fig. 7. PEMF Treatment preserves bone mass in vivo. The change in distal fibula bone volume for each rat over the 10-week period was computed using the formula described in Fig. 6: (A) varying degrees of bone volume reduction are seen in both PEMF-treated and sham-treated distal fibulae, (B) bone volume loss is significantly less in PEMF-exposed distal fibulae than in corresponding sham-treated bone segments.
conclusions because of one or more experimental flaws. Our study provides evidence that PEMF treatment resulted in a 75% reduction in the loss of bone volume at the distal fibular end of a critical-sized osteotomy in live animals. However, the volume changes reported here do not necessarily reflect changes in bone mineral density or mineral content. Further, in half of the animals, detectable bone growth was observed at the osteotomy sites in the PEMF-exposed fibulae as compared to their contralateral sham-treated fibulae, which on average widened their minimum gap size by 10’% Our study measured an enhancement of bone quantity at the proximal fibular segment of PEMF-treated versus shamtreated counterparts, though an imaging artifact related to the 1CT analysis of the proximal fibula segments marred our confidence in these results. The large size and proximity of the tibia to the fibula (Fig. lA, inset and Fig. 2B) resulted in some variable
X-ray beam-hardening artifacts in pCT images of the funnel-like region immediately subjacent to the fibula’s subchondral plate. These beam-hardening artifacts subsequently resulted in a variable streaking through the image’s voxels in this area that diminished accurate automated registration and alignment of longitudinal images of the proximal fibula segments. Specifically, the funnel-shaped proximal fibula’s metaphyseal region represents a much larger proportion of the entire proximal fibula mass than the bone region close to the osteotomy site. Given the streaking artifacts that variably occur in this funnel-shaped region of the proximal fibula and its pronounced effect on this bone segment’s volume, accurate automated alignments of this bone segment were not feasible. Another issue presented in the proximal, but not the distal fibular segment was an element of extra motion resulting from the ligament connections between the
1092
M . 0. Ibiwoye et al. I Journal of Orthopaedic Research 22 (2004) 1086-1093
proximal epiphysis of the fibula and the proximal epiphysis of the tibia. Thus, the alignment between the proximal and the distal fibular segments over time variably changed for each individual hind limb. Therefore, we could not accurately measure the proximal fibular segment volume based off of the distal fibular registration. The only feasible solution for analyzing the proximal fibular segment was to manually readjust the proximal fibular volume alignments nearer their osteotomy sites. However, this user intervention contributed an element of undesirable subjectivity compared with the entirely automated and objective data analysis for the distal fibular segments. Our confidence in the accuracy, precision and objectivity of the data obtained from the distal fibular segments and minimal gap size analyses leads us to conclude that PEMF enhances bone preservation at a critical-sized osteotomy site. These data provide a quantitative demonstration that PEMF treatments are biologically active in live animals. Based on these results, plans are under way to study the functional role of electromagnetic energy on bone physiology in vivo. In the model system used in this study, the observed bone volume changes in PhysioStim@-treatedbone nonunions as compared to their sham-treated counterparts are relatively small in magnitude. Indeed, the findings in this study have little clinical value. However, they have a significant impact with regard to the effects of PEMF on the biology and physiology of healing bone tissue. Our findings may also explain why there has been controversy in the interpretations of past scientific findings regarding PEMF’s efficacy to heal bone trauma, and certainly illustrates the need to use advanced pCT imaging to accurately quantify these differences in bone structure. Our interpretation of these relatively small effects is that more work needs to be done to enhance the waveform variables of exogenous PEMF signals, which could eventually lead to an improved effect on bone tissue formation in vivo. In the absence of quantitative data that validate the effectiveness of PEMF on bone formation in vivo, the utility of data from in vitro studies on the effects of PEMF on cultured cells [16,22] is limited. In this light, our study provides a strong rationale for the continued exploration of the cellular and molecular mechanisms in bone tissue elicited by PEMF treatment. A more widespread acceptance of PEMF as an anabolic treatment for bone repair will occur when its mechanistic properties are understood.
Acknowledgements This study was supported by a grant from the Orthopaedic Research and Education Foundation (OREF) with funding provided by Orthofix Inc. We
want to acknowledge the expert assistance of Larry Latson, M.S. and Michael Becker with pCT imaging and data processing. References Barker AT, Dixon RA, Sharrard WJW, Sutcliffe ML. Pulsed magnetic therapy for tibial non-union. Lancet 1984;1:9946. Bassett CAL, Becker RO. Generation of electric potentials by bone in response to mechanical stress. Science 1962;137: 10634. Brighton CT, Pollack SR. Treatment of recalcitrant non-union with capacitively coupled electrical field. J Bone Joint Surg Am 1985;67:577-85. Drew JS, Schwartz MH, Sachs SA. Distraction osteogenesis. NY State Dent J 1999;65:26-9. Engstrom H, Janson JO, Engstrom C. Effect of local irradiation on longitudinal bone growth in the rat. Acta Radio1 Oncol 1983; 22:129-33. Fredericks DC, Nepola JV, Baker JT, Abbott J, Simon B. Effects of pulsed electromagnetic fields on bone healing in a rabbit tibial osteotomy model. J Orthop Traumatol 2000;14:93-100. Fukada E, Yasuda I. On the piezoelectric effect of bone. J Phys SOCJap 1957;12:1158-62. Garland DE, Moses B, Salyer W. Long follow-up of fracture nonunions treated with PEMFs. Contemp Orthops 1991;22:295302. Goodship AE, Kenwight J. The influence of induced micromovement upon the healing of experimental tibial fractures. J Bone Joint Surg Brit 198.5:67:650-5. Gundersen HJG, Bendtsen TF, Korbo L, Marcussen N, Moller A, Nielsen K, et al. Some new, simple and efficient stereological methods and their use in pathological research and diagnosis. APMIS 1988;96:379-94. Gundersen HJG, Jensen EB. The efficiency of systematic sampling in stereology and its prediction. J Microsc 1987;147:22963. Jenis LG, An HS, Stein R, Young B. Prospective comparison of the effect of direct current electrical stimulation and pulsed electromagnetic fields on instrumented posterolateral lumbar arthrodesis. J Spinal Disord 2000;13:290-6. Kahanovitz N, Arnoczky SP, Nemzek J, Shores A. The effect of electromagnetic pulsing on posterior lumbar spinal fusions in dogs. Spine 1994;19:705-9. Kirchen ME, O’Connor KM, Gruber HE, Sweeney JR, Fras IA, Stover SJ, et al. Effects of microgravity on bone healing in a rat fibular osteotomy model. Clin Orthop Relat Res 1995;318:23142. Luben RA, Cain CD, Chen MC, Rosen DM, Adey WR. Effects of electromagnetic stimuli on bone and bone cells in vitro: inhibition of responses to parathyroid hormone by low-energy low-frequency fields. Proc Natl Acad Sci 1982;79:41804. Mark RA. Spine fusion for discogenic low back pain: outcomes in patients treated with or without pulsed electromagnetic field stimulation. Adv Therap 2000;17:57-67. McKibbin B. The biology of fracture healing in long bones. J Bone Joint Surg Brit 1978;60:150-62. McLeod K, Rubin CT. The effect of low-frequency electrical fields on osteogenesis. J Bone Joint Surg Am 1992;74:920-9. Mooney VMA. A randomized double-blind prospective study of the efficacy of pulsed electromagnetic fields for interbody lumbar fusions. Spine 1990;15:708-12. Petersson J, Johnell 0. Electrical stimulation of osteogenesis in delayed union of the rabbit fibula. Arch Orthop Trauma Surg 1983;101:247-SO.
M.O. Ihiwoye et al. I Journal of Orthopaedic Research 22 (2004) 1086-1093
[22] Pezzetti F, de Mattei M, Cadossi A, Zucchini P, Carinci F, Traina GC, et al. Effects of pulsed electromagnetic fields on human chondrocytes: an in vitro study. Calcif Tissue Inti 1999;65:396401. [23] Sarmiento A, Scheffer JF, Beckerman L, Latta LL, Enis JE. Fracture healing in rat femora as affected by functional weight bearing. J Bone Joint Surg Am 1977;59:369-75. [24] Scott G, King JB. A prospective double-blind trial of electrical capacitive coupling in the treatment of non-union of long bones. J Bone Joint Surg Am 1994;76:820-6.
I093
[25] Sharrard WJW. A double-blind trial of pulsed electromagnetic fields for delayed union of tibia1 fractures. J Bone Joint Surg Brit 1990;72:347-55. [26] Simonis RB, Good C, Cowell TK. The treatment of non-union by pulsed electromagnetic fields combined with a Denham external fixator. Injury 1983;15:255-60. [27] Zborowski M, Midura RJ. Wolfman A, Patterson TE, Ibiwoye MO, Sakai Y, et al. Magnetic field visualization in applications to pulsed electromagnetic field stimulation of tissues. Ann Biomed Eng 2003;31(2):195-206.
