Journal of

Orthopaedic
Research
ELSEVIER Journal of Orthopaedic Research 23 (2005) 899-908
www.elsevier.com/locate/orthres

Pulsed electromagnetic fields reduce knee 0st eoart hritic

lesion progression in the aged Dunkin Hartley guinea pig
M. Fini G. Giavaresi a, P. Torricelli a, F. Cavani b, S. Setti ',
V. Can6 b, R. Giardino
Department of Experimental Surgery, Codivilla-Putti Research Institute, Rizroli Institute of Orthopaedics,
a
Via di Barbiano, 1/10, 40136 Bologna, Italy
Department of Anatomy and Histology, University of Modena and Reggio Emilia, Largo del Pozzo. 71, 41100 Modena. Italy
IGEA SRL., via Parmenide 10IA. 41012 Carpi Modena, Italy
Chair of Surgical Pathophysiology, University of Bologna. via Massarenti 9, 40100 Bologna, ltaly
Accepted 10 January 2005

Abstract

An experimental in vivo study was performed to test if the effect of Pulsed Electromagnetic Fields (PEMFs) on chondrocyte
metabolism and adenosine A2a agonist activity could have a chondroprotective effect on the knee of Dunkin Hartley guinea-pigs
of 12 months with spontaneously developed osteoarthritis (OA). After a pilot study, 10 animals were randomly divided into two
groups: PEMF-treated group (6 hlday for 3 months) and Sham-treated group. Microradiography and histomorphometry were per-
formed on the entire articular surface of knee joints used in evaluating chondropathy severity, cartilage thickness (CT), cartilage
surface Fibrillation Index (FI), subchondral bone plate thickness (SBT) and histomorphometric characteristics of trabecular epiph-
yseal bone. The PEMF-treated animals showed a significant reduction of chondropathy progression in all knee examined areas
(p < 0.05). CT was significantly higher (p < 0.001) in the medial tibia plateaus of the PEMF-treated group when compared to the
Sham-treated group. The highest value of FI was observed in the medial tibia plateau of the Sham-treated group (p < 0.05). Signi-
ficant lower values were observed in SBT of PEMF-treated group in comparison to Sham-treated group in all knee examined areas
(p < 0.05). The present study results show that PEMFs preserve the morphology of articular cartilage and slower the progression of
OA lesions in the knee of aged osteoarthritic guinea pigs. The chondroprotective effect of PEMFs was demonstrated not only in the
medial tibia1 plateau but also on the entire articular surface of the knee.
0 2005 Orthopaedic Research Society. Published by Elsevier Ltd. All rights reserved.

Keywords: Osteoarthritis; Pulsed electromagnetic fields; Inflammation; Adenosine; Chondroprotection

Introduction the age of 75 have at least one joint affected by OA and,

Knee osteoarthritis (OA) management has become a
great challenge due to increased life expectancy o f a n
aging population more and more in demand of an active
and pain-free life [6]. Seventy to 90% of Americans above
* Corresponding author. Tel.: +39 051 6366557; fax: +39 051
6366580.
E-mail address: milena.fini@ior.it (M. Fini).
furthermore, it is estimated that by the year 2020, 18.2%
(59.4 million of Americans) will be affected by some form
o f OA [24,28]. Specifically, adequate treatment in early
phase pathology should be adopted in order to prevent
further degeneration especially when predictive factors
(symptoms, age, metabolic, mechanical a n d genetic influ-
ences) for OA development can be identified.
Ideally, therapy for OA should be directed towards
early pathogenetic events occurring during the disease

0736-0266/%- see front matter 0 2005 Orthopaedic Research Society. Published by Elsevier Ltd. All rights reserved.
doi: 10.1016/j.orthres.2005.01.008
900 M. Fini et al. I Journal of Orthopaedic Research 23 (2005) 899-908
process. Since the articular cartilage, subchondral bone
plate and trabecular epiphyseal bone are all targets in
OA, the possibility to investigate the causes which inter-
fere with both cartilage and bone tissue metabolism ap-
pears to be of special interest [49]. Current therapy for
patients with OA is primary palliative and aimed at
reducing pain, controlling inflammation and preserving
joint function. The majority of treatments have not been
shown to modify the complex pathological processes
that occur in these tissues, unable to balance undergoing
catabolic and anabolic pathways [42]. In addition, such
therapy for OA is unlikely to have a direct effect at the
subchondral bone level [34].
An approach to which treatment was aimed at stim-
ulating anabolic cartilage metabolism and, at the same
time, inhibiting cartilage breakdown was termed “Chon-
droprotection”. Chondroprotective disease modifying
drugs for OA should slow down or stabilize the progres-
sion of OA by altering underlying pathologic processes
~51.
For instance, gradual damage to cartilaginous tissue
is due to an imbalance in normal extracellular matrix
turnover where tissue degradation exceeds synthesis
due to alterations of normal functional activities of
chondrocytes (proliferation, matrix synthesis, apoptotic
death). Also unpaired response to different stimuli
[ 14,22,27]enables catabolic processes to prevail, thereby
leading to progressive loss of normal cartilage function.
Cartilage matrix breakdown leads to the development of
fibrillation, fissures and gross ulcerations leading to loss
of cartilage tissue. As far as bone tissue is concerned, it
is not completely clear how subchondral bone plate and
trabecular epiphyseal bone behave during the various
phases of OA progression. With increased disease sever-
ity, loss of cartilage function leads to hypertrophic bone
changes, osteophyte formation and subchondral bone
thickening due to mechanical stress applied to weight
bearing joints and existing microfractures [18,271.
Inflammatory mediated factors play a fundamental
role in a great variety of diseases including joint damage
consequent to septic, autoimmune and/or degenerative
processes. In recent years, there has been increasing evi-
dence regarding the role of inflammation in OA and cur-
rent knowledge suggests that proinflammatory cytokines
are responsible for the catabolic processes occurring in
pathological tissues [5,19,41,42]. Recent therapy in clin-
ical use today bases its effectiveness on the selective inhi-
bition of cyclooxygenase-2, an enzyme that is expressed
under the stimulus of varying inflammatory factors [33].
Adenosine has been recognized as a potent endogenous
anti-inflammatory agent that mediates its effect by inter-
acting with specific membrane receptors (most interest-
ingly A2a receptors) which are considered to be
specifically involved in the natural inhibitory mechanism
and/or termination of inflammation [4,12,37,45,48]. The
expression of adenosine receptor gene transcripts by
articular chondrocytes has been documented and further
reports indicate that adenosine A2a receptor agonists re-
duce cartilage damage when used in the treatment of
septic OA and rat adjuvant-induced arthritis [4,12,45].
Recent in vitro studies have demonstrated that PEMFs
act as A2a agonists by significantly increasing the
expression and functionality of A2a adenosine receptors
in human neutrophils [48].
Furthermore, different experimental cell culture and
in vivo models of endochondral ossification have dem-
onstrated the effect of PEMFs on increasing chondro-
cyte proliferation, synthetic activity and phenotypic
maturation [1,11,13,14,21,43].
All these considerations strongly support the ratio-
nale for further in vivo study of the possibility to inter-
fere with early phase mechanisms or progression of OA
through the use of PEMFs. Aged Dunkin Hartley gui-
nea pigs were used to assess PEMF effectiveness in
reducing OA knee lesion progression. In this study, knee
joints were explanted in toto. Whole joint (medial and
lateral tibia1 plateaus including medial and lateral femo-
ral condyles) cartilage, subchondral bone and epyphy-
seal trabecular bone underwent further evaluation.
Histology and histomorphometry were used to test if
the effect of PEMFs on chondrocyte metabolism in
addition to adenosine A2a agonist activity (previously
observed in vitro) could lead to a chondroprotective
effect in vivo.
Materials and methods
Animals
The study was approved by the Italian Ministry of Health and was
performed following Italian Law on animal experimentation. Dunkin
Hartley guinea pigs (Charles River, Calco, Lecco, Italy) aged 12
months were used. Animals were housed individually in Plexiglas cages
(40 x 25 x I 8 cm) and nourished with standard food (Piccioni, Settimo
Milanese, Italy) and given tap water ad libitum. Experimental condi-
tions were set up at a temperature 20 t 1 “C with a relative humidity
of 55% and 12 h of illumination alternated with 12 h of darkness. At
the end of the study, animals were euthanized via intravenous injection
of Tanax (Hoechst Roussel Vet, Milan, Italy) under general anaesthe-
sia (ketamine 87 mg/kg and xylazine 13 mg/kg). Right and left knee
joints were cut 1 cm above and below the joint line, stripped of muscle
and postfixed.
PEMF stimulator and experimental study
Electromagnetic stimulators generated a Pulsed Electromagnetic
Field with the following characteristics: frequency = 75 Hz, intensity
of electromagnetic field = 1.6 mT and duty cycle = 1.3 ms.The coil
was placed outside the cage and then connected to the pulsed genera-
tor. The stimulators were turned on for 6 h a day for 3 months. The
same conditions were applied to the five animals housed in separate
non-energized cages. This constituted the control of the experiment.
When the pulsed generator was switched off,the non-static electromag-
netic field background, measured by Emdex I1 (EnertechQZ), was of
5 ? 0.2 pT in both experimental and control cages. The instruments
used to evaluate the magnetic field and the induced voltage were a
Gaussmeter DG50 (Teslameter, Electrophysical Laboratory, Nervi-
 M. Fini rt al. I Journal o j Orthopaedic Research 23 (2005) 899-908
ano, Milano, Italy) and a Tektronix 720A oscilloscope (Tektronix,
Inc., Beaverton). Temperature was measured in both active and Sham
conditions with a probe having a sensitivity of 0.1 "C (Hygrometer HD
8501H, Deltaohm, Padova, Italy) and no differences were observed be-
tween the two groups. In the exposure conditions, coils and cages were
not in direct contact but separated by a 1 mm distance. The PEMF
generator system was kindly provided by IGEA (ONE, IGEA Srl,
Carpi, Italy).
In a pilot study 6 animals were used to select the minimal number
of animals and the daily exposure time to be used in the experimental
study. Two different exposure times, 4 and 12 h/day for 3 months, were
compared to a control group. A post-hoc power analysis (Sample
Power, SPSS Inc., Chicago, Illinois, USA) was done to calculate the
effect size of this study (1.22) and the sample size for the experimental
study (n = 5 for each group; p < 0.05). Results of the pilot study
showed that both exposure times were effective. Therapeutic effect
improvement was best seen in the 12 h/day when compared to the 4
h/day stimulation. In the experimental study, a minimal stimulation
time of 6 h/day was selected. This decision was also based on previous
animal and clinical studies [7,8,20,31].
Ten Dunkin Hartley guinea pigs aged 12 months were used. The
animals were randomly divided into two groups of five: PEMF-treated
group underwent PEMF stimulation for 6 h/day for 3 months while
the Sham-treated group underwent treatment simulation. At the end
of the study, animals (15 month-aged) were euthanized with the same
materials and methods previously described. A total of 10 PEMF-trea-
ted and 10 Sham-treated knees were examined.
Histology and histomorphornetry
Specimens were fixed in 4% buffered paraformaldehyde and dehy-
drated in a graded series of alcohols for undecalcified bone processing
in polymethylmethacrylate. As shown in Fig. 1, blocks were sectioned
along a sagittal plane with a Leica 1600 diamond saw microtome
(Leica SPA, Milan, Italy). A series of 10 sections, 300 pm in thickness
and spaced 300 pm apart, were obtained and microradiographed
(Micrography Italstructure, Como, Italy). After microradiography,
sections were thinly sliced to about 5 pm and stained with toluidine
blue and fast green and then processed for histological and histomor-
phometric analysis with a transmission and polarized light Axioskop
microscope (Carl Zeiss GmbH, Jena, Germany) and image analysis
Kontron KS 300 software (Kontron Electronic GmbH, Eiching bei
Munchen, Germany).
A total of 6 central slices were analyzed for each knee and the semi-
quantitative histological grading criteria of Mankin [32] modified by
I iLi
histology
6 central slices
Fig. I . Schematic representation of the entire knee joint sectioning. Ten sections 300-pm thick were obtained. Six central sections were studied by
means of microradiography. The same central sections were thinly sliced to 5 pm and used for histology and histomorphometry.
90 1
Carlsson was used for chondropathy evaluation at a magnification
of lOOx (Table 1) [9,25]. The histological score represented the sum
of articular cartilage structure, proteoglycan loss, cellularity and tide-
mark. Briefly, histological evidence of chondropathy included evalua-
tion of cartilage structure (surface irregularities such as clefts graded
from 0 in normal smooth uninterrupted cartilage to 8 in the presence
of clefts extending to zone of calcified cartilage); proteoglycan loss
(as seen by loss of toluidine staining graded from 0 in uniform staining
throughout articular cartilage to 6 in case of loss of staining in all 3
zones for half the length or greater of the condyle or plateau); cartilage
cellularity (from 0 in normal to 3 in hypocellularity); tidemark integrity
(0 in intactlsingle tidemark and 1 when tidemark is crossed by vessels
and reduplicated). This modified Mankin score has a possible maxi-
mum score of 18 points where a 0 score translates to a completely nor-
mal cartilage present throughout the entire knee. Lastly, a quantitative
evaluation of cartilage was determined. Cartilage thickness (CT) was
measured in pm at a magnification of 20x and the cartilage surface
fibrillation index (FI) was calculated according to the method devel-
oped by Pastoureau et al. [40] by dividing the length of the cartilage
surface border by the length of a standardized measured area x 100
(expressed in %) at a magnification of 80x.
The subchondral bone plate thickness (SBT) was measured in pm
from the cartilage-bone interface to the top of the epiphyseal mar-
row space. The mean of 10 measurements for each section perpendic-
ular to the articular surface is calculated as reported by Huebner et al.
PI.
Histomorphometric evaluation of tibia1 and femoral epiphyseal tra-
becular bone underlying subchondral bone plate was measured on the
same samples where the cartilage was evaluated. Measurements were
performed on the superior half of the epiphysis (from the end of the
subchondral bone plate to a distance of 450 f 50 pm from it) as sug-
gested by Pastoreau et al. [40]. The following parameters were calcu-
lated according to the American Society of Bone and Mineral
Research: trabecular bone volume (BV/TV, %), trabecular thickness
(Tb.Th, pm), trabecular number (Tb.N, /mm), trabecular separation
(Tb.Sp, pm) at a magnification of 50x [39].
All parameters were scored separately in the medial and lateral tib-
ial plateaus and femoral condyles. All sections were read blindly.
Statistics
Statistical analysis was performed using the SPSS v. 10.1 software
(SPSS Inc., Chicago, Illinois, USA). Data are reported as the
mean f SD at a significance level of p < 0.05. After testing data
for normal distribution, homogeneity of variance and sphericity of
I
1
slice thinning (5
I
902 M. Fini et al. / Journul of Orlhopaedic Research 23 (2005) 899-908

Table 1
Semiquantitative histologic grading score for knee joints chondropathy evaluation in guinea pigs [I-31
Parameter Grade Description
Articular cartilage structure 0 Normal, smooth, uninterrupted
1 Mild surface irregularities
2 Irregular surface, 1-3 superficial clefts
3 >3 clefts and/or loss of cartilage to superficial zone
4 1-3 clefts extending into the middle zone
5 r 3 clefts and/or loss of cartilage extending into the middle zone
6 1-3 clefts extending into the deep zone
7 2 3 clefts extending into the deep zone and/or loss of cartilage to deep zone
8 Clefts extending to zone of calcified zone
Toluidine blue staining 0 Uniform staining through articular cartilage
1 Loss of staining in superficial zone only and for <half the length of the condyle or plateau
2 Loss of staining in superficial zone for half the length or greater of the condyle or plateau
3 Loss of staining in superficial and middle zones for <half the length of the condyle or plateau
4 Loss of staining in superficial and middle zones for half the length or greater of the condyle or plateau
5 Loss of staining in all 3 zones for <half the length of the condyle or plateau
6 Loss of staining in all 3 zones for half the length or greater of the condyle or plateau
Cells 0 Normal (112 cells/lacuna)
1 Diffuse/slight hypercellularity
2 Regions of hypercellularity and cloning
3 H ypocellularity
Tidemark integrity 0 Intact/single tidemark
1 Crossed bv vessek/reduDhcdtion of tidemark

variance-covariance matrix, a repeated measures univariate ANOVA
was done to assess significant interactions between the between-subject
factor, treatment (Sham-treated; PEMF-treated), and the within-sub-
jects factor, measurement site (medial femoral condyle, medial tibial
plateau, lateral femoral condyle, lateral tibial plateau), for the histo-
morphometric data. When such interactions were found, a univariate
ANOVA was done to investigate the effects of the factors on the data
by means of hypotheses expressed as linear matrix according to the
SPSS syntax.
Results
During the study, all animals remained healthy as
indicated by continuous normal food consumption and
maintained body weight (mean body weight: 1220 f 78
g and 1228 k 70 g for Sham and PEMF-treated animals
respectively).
Figs. 2 and 3(a) and (b) feature the histological
appearance of PEMF- and Sham-treated animal knee
joints. Cartilage loss of staining and irregularities were
observed in Sham-treated animals together with osteo-
phytes (Figs. 2(b) and 3(b)) as observable also by micro-
radiography (Fig. 4(b) and (d)).
Statistical analysis performed by considering the
interaction of ‘treatment’ and ‘measurement site’ factors
on histomorphometric data showed that significant
interactions existed for the cartilage histological grading
score ( F = 5.60, p < 0.05), CT ( F = 15.31, p < 0.0005),
and FI ( F = 8.00, p < 0.001), indicating that all parame-
ters differed significantly for both treatment conditions
as well as the investigated articular surfaces. On the con-
trary, no significant interactions were observed for SBT
and epiphyseal trabecular bone histomorphometric
parameters. When the effect of each factor (treatment
and measurement site) was independently analysed for
SBT and epiphyseal trabecular bone histomorphometric
parameters, significant effects of ‘treatment’ were found
for SBT (p < 0.0005) and of the ‘measurement site’ for
epiphyseal trabecular bone histomorphometric para-
meters (p < 0.05).
Histological grading score results are reported in
Table 2. The PEMF-treated animals scored low, indicat-
ing a reduction in the progression of chondropathy in
comparison with Sham-treated animals (p < 0.05). The
results of medial tibia plateau had the highest histolo-
gical grading scores in both experimental and control
groups indicating that this site was more susceptible to
chondropathy. Regarding CT, a significantly higher
(p < 0.00 1) value was found in the medial tibia plateaus
of PEMF-treated group when compared to other mea-
surement sites of the same group and to the CT value
of the medial tibia plateaus of the Sham-treated group
(Table 3). The highest FI value was observed in the med-
ial tibia plateau of the Sham-treated group (p < 0.05)
(Table 4). Significantly lower values were observed in
SBT for the PEMF-treated group in comparison to
the Sham-treated group when considering each measure-
ment site 0, < 0.05) (Table 5). Epiphyseal trabecular
bone histomorphometric parameters did not show any
significant differences between Sham- and PEMF-trea-
ted animals (Table 6 ) .
 M. Fini et al. I Journul of’ Ortliopuedic Rrseurch 23 (2005) 899-908 903

Fig. 2. Histological sections of the same knee joint frontal section (level no. 7) of PEMF-treated (a) and Sham-treated (b) animals. Medial
compartment. Surface irregularities and loss of staining are visible on the tibial (T) articular cartilage of Sham-treated animals. The arrow indicates
an osteophyte on the femoral (F) surface of the Sham-treated animal. Toluidine blue staining, magnification lox.

Fig. 3. Histological sections of the same knee joint frontal section (level no. 6) of PEMF-treated (a) and Sham-treated (b) groups. Medial
compartment. Surface irregularities, loss of staining, hypocellularity are visible on the tibial (T) articular cartilage of Sham-treated animals. Loss of
staining and hypocellularity are also visible on the femoral (F) articular surface of Sham-treated animals. m: meniscus. Toluidine blue staining,
magnification 4 0 ~ .

Discussion the medial tibial plateau is the area most affected

The present experimental in vivo study was per-
formed in order to investigate the effect of PEMFs on
asymptomatic and spontaneously developed OA lesions
of the knee of the aged Dunkin Hartley guinea-pig
model. The Dunkin Hartley guinea pig is the most used
animal model for naturally occurring OA where early
histological changes appear with initial development
on the medial tibial plateau [23,25,26,38,47]. Reported
model reliability permits one to obtain quantitative data
from a relatively small group of animals [25].
As far as OA development is concerned, present re-
sults confirm previous literature data showing that
[25,38,47]. However, current evaluation of total knee tis-
sue verified the presence of OA lesions in all examined
areas with varying levels of severity.
The adopted statistical test allowed us to obtain
information on the differences among the various knee
joint areas within each group. On all articular cartilagi-
neous knee joint surfaces, PEMF treatment significantly
reduced cartilaginous alterations in structure, proteogly-
can content, tidemark and cellularity as demonstrated
with the adapted semiquantitative histological grading
score of knee joint chondropathy. In particular, in the
medial tibial plateaus (the most affected area of the
knee) the mean histological grade in the Sham-treated
904 M . Fini et al. I Journal of Orthopaedic Research 23 (2005)899-908

Fig. 4. Microradiography of the same knee joint frontal sections (levels nos. 6 and 7) of PEMF-treated (a and c) and Sham-treated (b and d) animals.
T Tibia; F Femour; M: medial compartment; L: Lateral compartment. The arrows indicate osteophytes that are particularly evident on the femoral
and tibial surface of Sham-treated animals.

Table 2 Table 3
Cartilage histological score results for Sham-treated and PEMF- Cartilage thickness (pm) results for Sham-treated and PEMF-treated
treated animals (Mean f SD, n = 5) animals (Mean f SD, n = 5)
Measurement site Treatment Measurement site Treatment
Sham-treated PEMF-treated Sham-treated PEMF-treated
~ ~~ ~~

Medial tibia plateau 10.9 i: 0.9" 3.9 f 0.7'2b Medial tibia plateau 167.4f 11.1 265.6 f 30.2*."
Medial femoral condyle 7.4 f 0.8 2.5 f 0.7' Medial femoral condyle 161.4 f 20.7 155.4 f 17.7
Lateral tibia plateau 6.4 f 1.0 1.6 f 0.4' Lateral tibia plateau 153.5 f 15.8 162.9 f 17.2
Lateral femoral condyle 5.9 f 0.9 1.8 f 0.5' Lateral femoral condyle 148.0 f 17.3 l47.7? 16.4
Univariate ANOVA test: Univariate ANOVA test:

- between Sham-treated versus PEMF-treated for each measure- - between Sham-treated versus PEMF-treated for each measure-
ment site: *p < 0.05. ment site: 'p < 0.05.
-between measurement sites within each group. ~ between measurement sites within each group.

a Medial tibia plateau versus other measurement sites (r, < 0.05).
Medial tibia plateau versus Lateral femoral condyle and Lateral
tibia plateau @ < 0.05).
group was 10.9 kO.9 compared to 3.9+ 0.7 in the
PEMF-treated group (p < 0.05). These results are of
particular interest because the knee chondropathy was
evaluated with the semi-quantitative histological grad-
ing scheme of Mankin [32] modified by Carlsson
[9,25]. Different grading scores are suggested for studies
on OA progression and the Mankin score is surely the
most widely used but the Carlson modified score is more
selective tool in evaluating cartilage structure and pro-
teoglycan loss in comparison with the Mankin score.
FI quantitative measurements confirmed data regarding
articular surface structure presenting with a greater
Medial tibia plateau versus other measurement sites (p < 0.001).
number of irregularities (cartilage fibrillation and clefts)
observed in Sham-treated animals in comparison with
treated ones. In the medial tibial plateau, the mean FI
in the Sham-treated group was 127 f 10% compared
+
with 104 2% in the PEMF-treated group (p < 0.05).
Cartilage surface irregularities, fibrillations and clefts
are late phenomena of OA lesion progression and, so,
the medial tibial plateau being the most affected area it
is normal that in Sham-treated animals the FI value is
higher than in the other knee joint measured sites. The
articular cartilage was significantly thicker in the medial
tibial plateau of PEMF-treated (265.6 f 30.2 pm) in
comparison with Sham-treated animals (1 67.4 f 1 1.1
pm) (p < 0.05). Obtained results on cartilage thickness
 M. Fini et al. I Journal of' Orthopuedic Research 23 (2005) 899-908 905

Table 4 parison with not stimulated animals (219 pm in PEMF

Fibrillation index (%) results for Sham-treated and PEMF-treated stimulated animals versus 98 pm in Sham-treated ani-
animals (Mean f SD, n = 5)
mals) and suggested that PEMF stimulation of Trans-
Measurement site Treatment forming Growth Factor-beta 1 (TGF-I31) could be
Sham-treated PEMF-treated responsible of the repair mechanism of action [lo].
Medial tibia plateau 127 f 10 104 f 2'3" As a consequence of a better functioning stimulated
Medial femoral condyle 108 f 5 102f 1 articular cartilage and in accordance with a lower histo-
Lateral tibia plateau 104 f 2 101 ? 1 logical score observed in PEMF-treated animals, a sig-
Lateral femoral condyle 103 f 1 103f 1
nificantly thinner SBP was observed in treated animals
Univariate ANOVA test: than in the control models. As far as cartilage histolo-
- between Sham-treated versus PEMF-treated for each measure- gical grading results are concerned, the PEMF stimula-
ment site: *p < 0.05. tion effect was evident in all examined knee areas. In the
- between measurement sites within each group. medial tibial plateau the mean SBT in the Sham-treated
a Medial tibia plateau versus other measurement sites (p < 0.05).
group was 284 ? 38 pm compared with 242 f 27 pm in
the PEMF-treated group (p < 0.05).
No significant differences between PEMF and Sham-
treated animals were observed in epiphyseal trabecular
Table 5
Subchondral bone thickness (pm) results for Sham-treated and PEMF- bone remodelling processes observed beneath the sub-
treated animals (Mean f SD, n = 5) chondral bone plate as demonstrated by BV/TV,
Measurement site Treatment Tb.N, Tb.Th and Tb.Sp measurements. Alterations in
Sham-treated PEMF-treated
epiphyseal trabecular bone metabolism in OA joints
seem to be quite complex, for instance, a mixture of
Medial tibia plateau 284.0 f 38.0 242.2 f 27.3'
Medial femoral condyle 325.0 f 32.8 271 .O f 27.5'
thickening and increased porosity. In human subjects af-
Lateral tibia plateau 304.3 f 31.4 265.4 f 29.2' fected with hip OA, detailed ultrastructural analyses
Lateral femoral condyle 306.5 f 34.4 261.7 k 28.3' showed that more bone was present in the subchondral
Unpaired Student's t-test be!ween Sham-treated versus PEMF-treated bone plate as well as in trabecular epiphyseal bone of the
for each measurement site: p < 0.05. femoral head. The bone appeared to be very porous with
large quantities of resorption pits with evidence of atyp-
ical bone formation [29]. It is possible that a stimulation
seem to indicate that PEMF treatment selectively pre- time of 3 months is not sufficient to condition a so com-
vents cartilage thinning due to OA at the medial tibial plex bone remodelling process.
plateau. The effect of PEMF was more evident in the The majority of OA lesion scores is limited to the
knee joint area that during the study period will undergo study of cartilaginous tissue. A limited number of grad-
spontaneously the largest degeneration. Data concur ing systems evaluate subchondral bone by means of his-
with a recent in vivo study of Ciombor et al. where tological grading scales or by measuring subchondral
PEMF stimulation (1 Wday) was applied to the knee bone thickness and epyphyseal bone trabecular volume
joints of 12 month old guinea pigs for 6 months [lo]. [30,40]. In the present paper, subchondral bone plate
After having evaluated the animal medial tibial plateau and trabecular epiphyseal bone were evaluated. It is
by means of a histologicaYhistochemica1score together widely recognized the importance of maintaining the
with immunohistochemistry studies, the above authors coexisting integrity of cartilaginous and bony tissues nec-
observed that cartilage was thicker in the medial tibial essary for articular function and that sclerosis and ebur-
plateau of guinea pigs stimulated with PEMFs in com- nation of subchondral bone set in motion mechanical

Table 6
Histomorphometric results of epiphyseal femoral and tibial bone for Sham-treated and PEMF-treated animals (Mean f SD, n = 5)
Parameters Groups Medial tibia Medial femoral Lateral tibia Lateral femoral
nlateau condvle olateau condvle
BVRV (Yo) Sham-treated 75.63 f 3.14 78.23 f 5.67 66.69 f 6.62 72.32 f 7.38
PEMF-treated 74.97 f 4.97 77.13 f 4.52 65.71 f 4.03 68.52 f 3.05
Tb.Th (pm) Sham-treated 186.75k 15.91 196.63 f 9.26 174.19 k 15.48 179.86 f 13.94
PEMF-treated 185.10 f 15.39 191.52 f 16.40 170.36 f 5.31 168.96 f 7.89
Tb.N (mm-') Sham-treated 4.12 f 0.28 4.12 f 0.16 3.87 f 0.24 4.07 f 0.23
PEMF-treated 4.06 f 0.18 3.97 f 0.16 3.90 f 0.23 4.08 f 0.25
Tb.SP (w) Sham-treated 62.59 f 8.80 53.84 ? 14.45 92.40 f 25.12 71.55 f 27.53
PEMF-treated 64.55 f 13.32 58.75 f 12.96 92.73 f 18.63 79.51 f 12.77
906 M. Fini et al. I Journal of Orthopaedic Research 23 (2005) 899-908
stress mechanisms acting on cartilage secondarily result-
ing in a further increase of bone density and cartilage
damage progression [3,16,17,27,34,36,42,49].
How PEMFs positively affect OA lesion progression
can be hypothesized thanks to basic knowledge acquired
from data available on the effect of PEMFs on cells and
tissues.
PEMFs increase chondrocyte matrix synthesis and
proliferation in vitro PEMF exposure bettered chondro-
cyte differentiation and phenotypic maturation when ap-
plied in an in vivo model of endochondral ossification
thanks to a sustained increase in TGFB-1 [1,11]. TGF-
131 is considered to be the most potent mitogen for adult
human articular chondrocytes among a variety of tested
factors and regulate cartilage and bone development
[1,22]. It was observed in ‘in vitro’ studies on bovine
articular cartilage that PEMF stimulation promoted
proteoglycan synthesis and, more interestingly, the effect
was preserved also in the presence of IL-1D (one of the
major inflammatory agents implicated in the etiology
of OA) [14]. These observations strongly support an
anabolic effect of PEMFs on cartilage. In the authors’
opinion, the anabolic effect resulting from stimulation
is also active in vivo with a direct effect on chondrocyte
metabolism.
Another possible explanation of the in vivo chondro-
protective effect of PEMFs derives from the demon-
strated effect of PEMFs on the inflammatory process.
The OA process is often associated with an inflamma-
tory response and many local inflammatory factors are
considered responsible, at least in part, for the changes
seen in cartilage, synovial membrane and subchondral
bone [42]. Current evidence suggests that soluble media-
tors such as cytokines, growth factors, inflammatory
mediators, metalloproteinases and chondrodegradative
enzymes are responsible for the catabolic process occur-
ring in joints affected by OA (i.e. IL-IB, TNF-a, IL-6,
IL-8, IL-2, LIF, IL-17, NO and others). On the other
hand, natural inhibitors (TGF-B1, IGF-1, IL-4, IL-10
and IL-13) able to counteract proinflammatory effects
were identified [35,41,42].
PEMF saturation binding experiments revealed a sig-
nificant increase of A2a adenosine receptor density in
human neutrophils treated with PEMFs accompanied
by a significant increase in adenylcyclase activity and
reduction of superoxide anion production as a result
of upregulation of A2a receptors [48]. Modulation of
extracellular adenosine concentration or modulation of
adenosine receptor activity is suggested to represent a
natural mechanism of controlling inflammation and
a novel pathway for intervention in arthritic diseases
[37,44,46]. Adenosine limits inflammatory response
through receptor-mediated regulation of different cell
types. It suppresses lipopolysaccharide-induced pro-
duction of NO and proinflammatory cytokine synthe-
sis (TNF-a, IL-8, IL-2, IL-6, etc.) and increases the
synthesis of IL-10 [2,45]. Recent studies have also
demonstrated the functional response of articular chon-
drocytes to adenosine which at the same time release
adenosine in response to tissue damage in arthritic con-
ditions [45]. Finally, new pharmacological adenosine
A2a agonist agents have been reported to have a chon-
droprotective effect in experimental in vivo studies
~4,121.
The present results have confirmed and extended the
observations of other authors by showing that PEMFs
preserve the morphology of articular cartilage and re-
tard the development of OA lesions in the knee of aged
osteoarthritic guinea pigs [lo]. Furthermore, in the pres-
ent study, the chondroprotective effect of PEMFs was
demonstrated not only in the medial tibial plateau but
also on the entire articular knee surface. An anti-inflam-
matory mechanism of action is also hypothesized by the
present authors on the basis of recent evidence regarding
the chondroprotective effect of adenosine agonists selec-
tive to the A2a receptor and of the in vitro capability of
PEMFs of increasing A2a adenosine receptor numbers.
Acknowledgement
This work was partially supported by Rizzoli Ortho-
paedic Institute (Ricerca Corrente). We gratefully ac-
knowledge the technical assistance provided by Claudio
Dal Fiume, Nicola Corrado, Franca Rambaldi, Paola
Chiavelli, Albert0 Ianni and Patrizia Nini of The Exper-
imental Surgery Department, Rizzoli Orthopaedic Insti-
tute, Bologna, Italy). No benefits in any form have been
or will be received from a commercial party directly or
indirectly related to the subject of this article.
References
[I] Aaron RK, Wang S, Ciombor DM. Upregulation of basal TGF
beta 1 levels by EMF coincident with chondrogenesis-impla-
tions for skeletal repair and tissue engineering. J Orthop Res
2002;20(2):23340.
[2] Benton HP, MacDonald MH, Tesch AM. Effects of adenosine on
bacterial lipopolysaccharide and interleukin induced nitric oxide
release from equine articular chondrocytes. Am J Vet Res 2002;
63(2):20&10.
[3] Bobinac D, Spanjol J, Zoricic S, Maric I. Changes in articular
cartilage and subchondral bone histomorphometry in osteoar-
thritic knee joints in humans. Bone 2003:284-90.
[4] Boyle DL, Moore J, Yang L, Sorkin LS, Firestein GS. Spinal
adenosine receptor activation inhibits inflammation and joint
destruction in rat adjuvant-induced arthritis. Arthritis Rheum
2002;46(11):307&82.
[5] Brooks P. Inflammation as an important feature of osteoarthritis.
Bull World Health Organ 2OO3;8 1(9):689-90.
[6] Bruyere 0, Dardenne C, Lejeune E, Zegels B, Pahaut A, Richy F,
et al. Subchondral tibial bone mineral density predicts future
joint space narrowing at the medial femorotibial compartment in
patients with knee osteoarthritis. Bone 2003;32:541-5.
 M. Fini et al. I Journal of Orthopaedic Research 23 (2005) 899-908
[7] Cane V, Botti P, Farneti D, Soana S. Electromagnetic stimulation
of bone repair: a histomorphometric study. J Orthop 1991;9(6):
908-1 7.
[8] Cane V, Botti P, Soana S. Pulsed electromagnetic fields improve
osteoblast activity during the repair of an experimental osseous
defect. J Orthop Res 1993;11(5):66&70.
[9] Carlsson CS, Loeser RF, Purser CB, Gardin JF, Jerome CP.
Osteoarthritis in cynomolgus macaques 111: effects of age, gender,
and subchondral bone thickness on the severity of disease. J Bone
Miner Res 1996;11:1209-17.
[lo] Ciombor DM, Aaron RK, Wang S, Simon B. Modification of
osteoarthritis by pulsed electromagnetic field-a morphological
study. Osteoarthr Cartilage 2003; 11(6):45542.
[I I] Ciombor DM, Lester G, Aaron RK, Neame P, Caterson B. Low
frequency EMF regulates chondrocyte differentiation and expres-
sion of matrix proteins. J Orthop Res 2002;20(1):40-50.
[12] Cohen SB, Gill SS, Baer GS, Leo BM, Scheld WM, Diduch
DR. Reducing joint destruction due to septic arthrosis using
an adenosine 2A receptor agonist. J Orthop Res 2004;22:
427-35.
[I31 De Mattei M, Caruso A, Pezzetti F, Pellati A, Stabellini G,
Sollazzo V, et al. Effects of pulsed electromagnetic fields on
human articular chondrocyte proliferation. Connect Tissue Res
200 1;42(4):269-79.
1141 De Mattei M, Pasello M, Pellati A, Stabellini G, Massari L,
Gemmati D, et al. Effects of electromagnetic fields on proteogly-
can metabolism of bovine articular cartilage explants. Connect
Tissue Res 2003;44(34): 154-9.
[15] Dean D, Muniz OE, Rodriquez I, Carreno MR, Morales S,
Agundez A, et al. Amelioration of lapine osteoarthritis by
treatment with glycosaminoglycan-peptide association complex
(Rumalon). Arthritis Rheum 1991;34(3):304-13.
[I61 Dieppe P. Subchondral bone should be the main target for the
treatment of pain and disease progression in osteoarthritis.
Osteoarthr Cartilage 1999;7:3254.
[I71 Fazzalari NL, Parkinson IH. Femoral trabecular bone of osteo-
arthritic and normal subjects in an age and sex matched group.
Osteoarthr Cartilage 1998;6:377-82.
[I81 Fenwick SA, Gregg PJ, Rooney P. Osteoarthritic cartilage loses
its ability to remain avascular. Osteoarthr Cartilage 1999;7:
44-52.
[I91 Fernandes JC, Martel-Pelletier J, Pellettier JP. The role of
cytokines in osteoarthritic pathophysiology. Biorheology 2002;
39( 1-2):23746.
[20] Fini M, Cadossi R, Cane V, Cavani F, Giavaresi G, Krajewski A,
et al. The effect of pulsed electromagnetic fields on the osteoin-
tegration of hydroxyapatite implants in cancellous bone: a
morphologic and microstructural in vivo study. J Orthop Res
2002;20:75663.
[21] Fioravanti A, Nerucci F, Collodel G, Markoll R, Marcolongo R.
Biochemical and morphological study of human articular chon-
drocytes cultivated in the presence of pulsed signal therapy. Ann
Rheum Dis 2002;61(11):1032-3.
[22] Guerne PA, Blanco F, Kaelin A, Desgeorges A, Lotz M. Growth
factor responsiveness of human articular chondrocytes in aging
and development. Arthritis Rheum 1995;38(7):960-8.
[23] Han B, Cole A, Shen Y, Brodie T, Williams J. Early alterations in
the collagen meshwork and lesions in the ankles are associated
with spontaneous osteoarthritis in guinea-pigs. Osteoarthr Carti-
lage 2002; 10(10):778.
[24] Hinton R, Moody RL, Davis AW, Thomas SF. Osteoarthritis:
diagnosis and therapeutic considerations. Am Fam Physicians
2002;65:8414.
[25] Huebner JL, Hanes MA, Beekman B, TeKoppele JM, Kraus VB.
A comparative analysis of bone and cartilage metabolism in two
strains of guinea-pig with varying degrees of naturally occurring
osteoarthritis. Osteoarthr Cartilage 2002; 1075847.
907
[26] Kapadia RD. Badger AM, Levin JM, Swift B, Bhattacharyya A,
Dodds RA, et al. Meniscal ossification in spontaneous osteoar-
thritis in the guinea-pig. Osteoarthr Cartilage 2000:83747.
[27] Lajeunesse D, Hlal G, Pellettier JP, Pellettier JM. Subchondral
bone morphological and biochemical alterations in osteoarthritis.
Osteoarthr Cartilage 1999;7:321-2.
[28] Lawrence RC, Helmick CG, Arnett FC, Deyo RA, Felson DT.
Giannini EH, et al. Estimates of the prevalence of arthritis and
selected musculoskeletal disorders in the United States. Arthritis
Rheum 1998;41(5):778-99.
[29] Li B, Marshall D, Roe M, Aspden RM. The electron microscope
appearance of the subchondral bone plate in the human femoral
head in osteoarthritis and osteoporosis. J Anat 1999;195:101-10.
[30] Mainil-Varlet P, Riesert F, Grogan S, Mueller W, Saager C,
Jakob RP. Articular cartilage repair using a tissue-engineered
cartilage-like implant: an animal study. Osteoarthr Cartilage 2001;
9(SA):S&I5.
[31] Mammi GI, Rocchi R, Cadossi R, Massari L, Traina GC. The
electrical stimulation of tibia1 osteotomies. Double blind study.
Clin Orthop 1993;288:24&53.
[32] Mankin HJ, Dorfman H, Lippiello L, Zarins A. Biochemical and
metabolic abnormalities in articular cartilage from osteo-arthritic
human hips. J Bone Joint Surg 1971;53-A:523-37.
[33] Martel-Pelletier J, Pelletier JP, Fahm IH. Cyclooxygenase-2 and
prostaglandins in articular tissues. Semin Arthritis Rheum 2003;
33(3):155-67.
[34] Mattei JP, Roux H. New potential therapeutic goals: subchondral
bone and progression of osteoarthritis. Osteoarthr Cartilage 1999:
7:329-30.
[35] Moskowitz RW, Kelly MA, Lewallen DG. Understanding
osteoarthritis of the knee-causes and effects. Am J Orthop 2004;
33(25):5-9.
[36] Moskowitz RW. Bone remodelling in osteoarthritis: subchondral
and osteophytic responses. Osteoarthr Cartlage 1999;7:3234.
[37] Ohta A, Sitkovsky M. Role of G-protein-coupled adenosine
receptors in downregulation of inflammation and protection from
tissue damage. Nature 2001;414:916-9.
[38] Osborne D, Woodhouse S, Meacock R. Early changes in the
sulfation of chondroitin in guinea-pig articular cartilage, a
possible predictor of osteoarthritis. Osteoarthr Cartilage 1994;
2(3):215-23.
[39] Parfitt AM, Drezner MK, Glorieux FH, Kanis JA, Malluche H,
Meunier PJ, et al. Bone histomorphometry: standardization of
nomenclature, symbols and units. J Bone Miner Res 1987;2:
595-610.
[40] Pastoreau P, Leduc S, Chomel A. De Ceuninck F. Quantitative
assessment of articular cartilage and subchondral bone histology
in the meniscectomized guinea pig model of osteoarthritis.
Osteoarthr Cartilage 2003;11:412-23.
[41] Pelletier JM. Pathophysiology of osteoarthritis. Osteoarthr Car-
tilage l 999;7:37 1-3.
[42] Pelletier JP. The influence of tissue cross-talking on OA progres-
sion: role of non-steroidal anti-inflammatory drugs. Osteoarthr
Cartilage 1999;7:37&6.
[43] Pezzetti F, De Mattei M, Caruso A, Cadossi R, Zucchini P,
Carinci F, et al. Effects of pulsed electromagnetic fields on human
chondrocytes: an in vitro study. Calcif Tissue Int 1999;
65(5):396401.
[44]Tesch A, Benton H, MacDonald M, Kollias-Baker C. Adenosine
activates articular chondrocytes via specific A2a receptors. In:
47th Annual Meeting of the Orthopaedic Research Society, San
Francisco, California, February 25-28, 2001.
[45] Tesch AM, MacDonald MH, Kollias-Baker C, Benton HP.
Chondrocytes respond to adenosine via A2 receptors and
activity is potentiated by an adenosine deaminase inhibitor
and a phosphodiesterase inhibitor. Osteoarthr Cartilage 2002;
103443.
908 M. Fini et al. I Journal of Orthopaedic Research 23 (2005) 899-908

[46] Tesch AM, MacDonald MH, Kollias-Baker C, Benton HP.
Effects of an adenosine kinase inhibitor and an adenosine
deaminase inhibitor on accumulation of extracellular adenosine
deaminase inhibitor on accumulation of extracellular of extracel-
Mar adenosine by equine articular chondrocytes. Am J Vet Res
2002;63(11):1512-9.
[47] Tessier JJ, Bowyer J, Brownrigg NJ, Peers S, Westwood FR,
Waterton JC, et al. Characterisation of the guinea pig model of
osteoarthritis by in vivo three-dimensional magnetic resonance
imaging. Osteoarthr Cartilage 2003;11:1-9.
[48] Varani K, Gessi S, Merighi S, Iannotta V, Cattabriga E, Spisani S,
et al. Effects of low frequency electromagnetic fields on A2A
adenosine receptors in human neutrophils. Br J Pharmacol
2002;136:57-66.
[49] Wollheim FA. Bone sialoprotein-a new marker for subchondral
bone. Osteoarthr Cartilage 1999;7:331-2.