Orthopaedic Surgery (2011), Volume 3, No.

3, 181–187

CLINICAL ARTICLE

Efficacy of ultrasound in the treatment of osteoarthritis

of the knee os4_144 181..187

Peng-fei Yang MM1, Dong Li MM2, Shi-mo Zhang MM2, Qing Wu MM3, Jin Tang MM1,
Liang-ku Huang MM1, Wei Liu MM1, Xi-dong Xu MM1, Shi-rong Chen MD1
1
Department of Orthopaedics, The Second Affiliated Hospital of Chongqing Medical University, 2National Engineering Research Center of
Ultrasound Medicine, and 3People’s Hospital of Shapingba District, Chongqing, China

Objective: To investigate the effect of ultrasound in the treatment of osteoarthritis of the knee.
Methods: Eighty-seven out- and in-patients with osteoarthritis of the knee (15 men and 72 women) underwent
ultrasonic therapy from February to October 2010. The patients were randomly assigned to an ultrasound group
(Group A) and a placebo group (Group B). Group A was offered ultrasonic therapy while Group B underwent mock
treatment. The visual analogue scale (VAS) and Lequesne scores of the two groups before and after treatment were
compared. The data were analyzed by normal distribution, Student’s t-test and the rank sum test. The means during
and after treatment of the VAS and Lequesne efficacy index of the treatment group were calculated and plotted on a
bar graph.
Results: Single sample analysis of Groups A and B showed VAS efficacy index: mean = 0.3640, P = 0.000; Lequesne
efficacy index, median = 0.3080, P = 0.000, and mean = 0.1000, P = 0.000; Lequesne efficacy index, mean = 0.0364, P =
0.024, respectively. Independent samples t test and rank sum test showed significant differences between the two groups,
namely P = 0.000 for both the VAS and Lequesne curative effect indexes. The means of the VAS efficacy index of the
treatment group increased during and after treatment. The means of the Lequesne efficacy index of the treatment group
showed no apparent decrease by 28 days after treatment.
Conclusion: Ultrasound treatment significantly alleviates joint symptoms, relieving joint swelling, increasing joint
mobility and reducing inflammation, in osteoarthritis patients.
Key words: Knee; Osteoarthritis; Sonication

Introduction os4_144 181..187

years, has been proved to promote repair of full-thickness
articular cartilage defects, result in formation of hyaline
Osteoarthritis (OA) is a joint disease caused by long-
cartilage-like repair tissue at the site of the defect1, soften
term wear, which damages articular cartilage and causes
and dissipate condensed fibrous connective tissue and
intra-articular fracture and intra-articular loose bodies,
delay the progress of early OA2. This study aimed to
resulting in joint fibrosis, pain and dysfunction. It is char-
explore the clinical effects of ultrasound in the treatment
acterized by cartilaginous changes, bone hypertrophy and
of osteoarthritis of the knee (KOA).
osteophyte formation. It is the commonest chronic pro-
gressive joint disease, having a serious impact on the
Materials and methods
quality of life of patients. Current clinical methods for the
treatment of OA are diverse, but both drug therapy and General data
surgical treatment have shortcomings and limitations. Eighty-seven out- and in-patients with KOA were
Ultrasound, a new treatment method developed in recent studied, including 100 knees in total. All diagnoses were
made in accordance with the KOA criteria recommended
Address for correspondence Shi-rong Chen, MD, Department by the 1995 American College of Rheumatology (ACR)3.
of Orthopaedics, The Second Affiliated Hospital of Chongqing There were 15 male and 72 female patients with an average
Medical University, 76 Riverside Road, Chongqing, China 400010
of 58.3 years (range, 38–81 years). The course of disease
Tel: 0086-23-63693558; Fax: 008-23-63693558; Email:
[email protected] was from 2 months to 20 years with an average of 2.8
Received 29 December 2010; accepted 25 March 2011 years. All patients gave informed consent and attended for
DOI: 10.1111/j.1757-7861.2011.00144.x regular follow-up. The research was approved by the

© 2011 Tianjin Hospital and Blackwell Publishing Asia Pty Ltd 181
182 P Yang et al, Ultrasound in osteoarthritis

Figure 1 Procedure for use of therapeutic

ultrasound equipment: the knee eyes, inte-
rior and lateral knee joint spaces have been
marked and the heads fixed to the knee
eyes and the knee joint space.

Ethics Committee of our institution. Fifty knees were ran- marked; appropriate parameters set on the ultrasound
domly assigned to the ultrasound group (Group A) and equipment; about 2 mL of coupling agent smeared onto
the other 50 to a placebo group (Group B). The ultra- each head of the ultrasound treatment equipment; the
sound therapeutic equipment, also known as ultrasound heads fixed to the eyes of the knee and the knee joint
therapeutic equipment for arthritis, was provided by the space, as shown in Fig. 1; the elastic band adjusted; and
National Engineering Research Center of Ultrasound ensuring that the equipment head was firmly applied to
Medicine (NERCUM, Chongqing, China). the skin and would not slip. Patients in group A were
treated with the treatment model for 15 minutes and then
Inclusion and exclusion criteria switched over to the rehabilitation model for 20 minutes.
The inclusion criteria were as follows: (i) meeting the For patients in group B only the mode was changed and
diagnostic criteria for KOA, the chief features being pain they received no treatment.
and movement disorder; (ii) local pain with a visual
analog score (VAS) of more than three points; (ii) more Clinical assessment
than 38 years old; and (iv) stage I-III according to the The VAS and Lequesne scores were recorded before
X-ray films. treatment, during treatment and at four weeks after treat-
The exclusion criteria were as follows: (i) failure to ment to obtain the curative effect index according to the
meet any one of the above inclusion criteria; (ii) skin following equation.
ulceration, infection, bullous lesions, and other skin con-
ditions; (iii) major scarring as a result of previous surgery VAS curative effect index = (VAS score before
or trauma; (iv) disorders of speech or hearing or those treatment-VAS score after treatment)/(VAS score
who did not comply with the therapy; (v) “halo-pin”; before treatment) ¥ 100%
(vi) other cases that clinicians considered unsuitable for The Lequesne curative effect index was calculated as
the study. Patients were excluded if they met any one of follows.
these criteria.
Lequesne curative effect index = (total score before
treatment-total score after treatment)/pre-treatment
Treatment
total score ¥ 100%
Ultrasound or mock treatment was performed as
follows. The patient’s history was taken, a medical report Both indexes were used to assess the effectiveness of the
form completed; the eyes of the knee (on the level of the treatment. After the treatment had been completed, the
lower border of the patella, lateral and medial to the patients were closely followed up to monitor the progress
patellar ligament), interior and lateral knee joint space of the disease and evaluate the effects of the treatment.

© 2011 Tianjin Hospital and Blackwell Publishing Asia Pty Ltd

Orthopaedic Surgery (2011), Volume 3, No. 3, 181–187 183

Statistical analysis 0.3080, SD = 0.42076, P = 0.000) showed that both the

SPSS 17.0 statistical software was used. Statistical sig- VAS and Lequesne efficacy indexes of group A after treat-
nificance was set at P < 0.05. Normal analysis of the two ment were statistically significantly different from the
groups, single sample t test of the efficacy index of the pre-treatment scores, indicating that the treatment was
control group and completely independent samples t test effective.
and rank sum test were used to compare and analyze any
differences between the treatment and control groups. Single sample t test of the efficacy index in
the control group
Results Single sample t test of the efficacy index of group B
(VAS efficacy index, M = 0.1000, SD = 0.18729, P = 0.000;
Treatment Lequesne efficacy index, M = 0.0364, SD = 0.11082, P =
Over the course of the treatment, no patients had 0.024) showed that both the VAS and Lequesne efficacy
adverse reactions such as skin irritation, blisters and so on. indexes of the control group after the mock treatment
Three patients complained of mental stress, dizziness, pal- were statistically significantly different from the pre-
pitation, or fatigue. Their treatment was stopped imme- treatment scores, indicating that the treatment was
diately, they were asked to lie down, given oxygen, oral effective.
glucose water and so on and their symptoms resolved
quickly.
Completely independent samples t test and rank
sum test comparing effectiveness between
Normal analysis of the treatment group (group A) the treatment and placebo groups
After the fifth treatment session the VAS and Lesquesne The VAS efficacy indexes of the treatment and control
effect indexes for group A were analyzed for their Wishart groups after treatment were compared. As is shown in
(W) distributions. According to this method of checking Table 1, Levene’s test showed heterogeneity of variance
distribution, the VAS curative effect index yielded W = (P = 0.003) and t test showed there was a statistically
0.958, P = 0.071, and the Lequesne curative effect index significant difference between the two groups (P = 0.000).
W = 0.959, P = 0.080, verifying that both of them had The rank sum test also showed a statistically significant
a normal distribution. difference (P = 0.000). This is also evident from the mean
of the treatment group being 0.364 whereas that of the
Normal analysis of the placebo group (group B) control group was 0.1000. Accordingly it was concluded
After the fifth treatment session the curative and that the efficacy of treatment was better than that of
Lequesne effect indexes for group B were analyzed for placebo (mock treatment) when the VAS efficacy indexes
their W distributions. According to this method of check- weres used as the evaluation criterion.
ing distribution, the VAS curative effect index yielded W = The Lequesne efficacy indexes of the treatment and
0.955, P = 0.053, and the Lequesne curative effect index placebo groups after treatment were also compared. As is
W = 0.954, P = 0.051, verifying that both of them had shown in Table 2, Levene’s test showed heterogeneity of
a normal distribution. variance (P = 0.000) and t test showed a statistically sig-
nificant difference between the means of the two groups
Single sample t test of the efficacy index in (P = 0.000). The rank sum test also showed a statistically
the treatment group significant difference (P = 0.001). This was also evident
Single sample t test of the efficacy index of the treat- from the mean of the treatment group being 0.3080 while
ment group (VAS efficacy index, mean = 0.3640, SD = that of the control group was 0.0364. Accordingly it was
0.28062, P = 0.000; Lequesne efficacy index, mean = concluded that the efficacy of treatment was better than

Table 1 Analysis of differences in the VAS curative effect index between the two groups after treatment

Levene’s test t test Mean Rank sum test

Items P P Group A Group B P

Equal variance 0.003 0.000 0.3640 0.1000 0.000

Unequal variance — 0.000

© 2011 Tianjin Hospital and Blackwell Publishing Asia Pty Ltd

184 P Yang et al, Ultrasound in osteoarthritis

Table 2 Analysis of differences in the Lequesne curative effect index between the two groups after treatment

Levene’s test t test Mean Rank sum test

Items P P Group A Group B P

Equal variance 0.000 0.000 0.3080 0.0364 0.001

Unequal variance — 0.000

that of the mock treatment when the Lequesne efficacy treatment (1, 3, 7, 14, and 28 days) to record their VAS and
indexes were used as the evaluation criteria. Lequesne scores. The VAS (visual analogue scale) score
represents the degree of pain, zero being no pain and ten
Change in means of the efficacy index in the most severe pain. The patients indicated their scores
the treatment group on the VAS themselves according to the pain they were
The means of the VAS efficacy index of the treatment experiencing. The Lequesne score criterion includes six
group during (1–5 days) and after the treatment (3, 7, 14, indicators, namely rest pain, pain on movement, tender-
and 28 days) were calculated and plotted on a bar graph. ness, swelling, morning stiffness, and walking ability. We
As shown in Fig. 2, the symptom of pain was significantly assessed the scores (0–3) for each indicator, zero repre-
relieved during the treatment. The efficacy was still senting normal and three the worst. One patient from the
evident 28 days after treatment. treatment group and two from the control group took oral
The means of the Lequesne efficacy index of the treat- analgesics during follow-up, but the amount of medica-
ment group after the treatment (1, 3, 7, 14, and 28 days) tion they took was small, and this was not statistically
were calculated and plotted on a bar graph. As shown in significant.
Fig. 3, the efficacy did not appear to decline within 28 days
after treatment. Discussion
Follow-up Osteoarthritis of the knee
All patients were followed up for about one month. We Osteoarthritis of the knee is a disorder in which there is
went to their homes or they came to our hospital after the bone and cartilage degeneration, articular cartilage degen-

0.5
0.45
Mean VAS curative

0.4
effect index

0.35
0.3
0.25
0.2 Figure 2 Average VAS curative effect
0.15
0.1 index during and after treatment. The
0.05 symptom of pain was significantly relieved
0
during treatment and the therapeutic
the first day 2 days 3 days 4 days 5 days 3 days 7 days 14 days 28 days
of the treatment effect continued to gradually improve up
Time after treatment until 28 days after treatment.
Mean Lequesne curative effect index

0.4
0.35
0.3
0.25
0.2
0.15
0.1
0.05
Figure 3 Average Lequesne curative effect
0
index after treatment. There was no
24 hours 3 days 7 days 14 days 28 days
obvious decline in efficacy over the 28 day
Time after treatment follow-up period.

© 2011 Tianjin Hospital and Blackwell Publishing Asia Pty Ltd

Orthopaedic Surgery (2011), Volume 3, No. 3, 181–187 185

eration including fibrosis, cracks, ulcers and even loss of degradation of the cartilage matrix. It can therefore
the full thickness of the articular cartilage1. KOA not only promote cartilage repair, improve joint mobility, relieve
affects the articular cartilage, but also involves the entire joint pain and slow the pathological changes of osteoar-
joint including the subchondral bone, ligaments, joint thritis. Studies have shown that ultrasound may increase
capsule, synovial membrane and periarticular muscles. intra-articular type II collagen synthesis6 and stimulate
The major clinical symptoms are knee joint pain and cartilage cell proliferation7.
instability2. Most people of 65 years or older have varying The knee joint fluid of patients with OA contains large
symptoms of OA, such as pain, joint dysfunction, loss of amounts of cytokines, among which interleukin-1 (IL-1)
working ability or total disability. At present, the main is representative. IL-1 is mainly involved in increasing the
treatment is non-surgical and the goal of rehabilitation is expression of endothelial cell adhesion molecule receptor
to relieve pain, and reduce further strain and the degree of and promoting synthesis of collagenase and other neutral
the disability. proteases. It thereby enhances the inflammatory response
within the synovium and synovial fluid, and promotes the
The mechanism of ultrasound treatment proliferation of synovial cells. In the normal state, the
Ultrasound is a newly-emerging rehabilitation treat- amount of IL-1 maintains a balance. When that balance is
ment. Its effects include the following: mechanical effects, disturbed, it plays an important role in causing destruc-
cavitation, thermal effects and chemical effects. Ultra- tion of bone and cartilage and inflammation of synovial
sound can promote or accelerate certain chemical reac- fluid. In KOA patients, the amount of IL-1 is greater than
tions. When applied to articular cartilage, joint capsules the normal, the amount being positively correlated with
and synovial membranes, it can change the membrane the severity of KOA. Increased amounts of IL-1 are closely
potential, enhancing the permeability of ions and col- correlated with synovitis and articular cartilage matrix
loids; promote blood circulation; soften tissue; stimulate degradation and interfere with the function of chondro-
cells; accelerate chemical reactions; enhance metabolism; cytes11. Ultrasound promotes the diffusion of small mol-
affect enzyme function and content of biologically active ecules in the blood into the synovial fluid, and can
substances; change the pH; decrease the excitability of probably increase the exudation of synovial fluid12,
sensory nerves; increase the pain threshold, and thus degrade or eliminate inflammatory mediators and other
achieve a therapeutic effect. It has been proved that ultra- transmitters, and reduce the production and release of
sound can promote the repair of defects in full-thickness inflammatory mediators13.
articular cartilage and result in formation of hyaline
cartilage-like tissue at the the sites of defects3,4. In elderly
patients, ultrasound applied during the consolidation Clinical research on ultrasound
phase of distraction osteogenesis accelerates callus matu- Ultrasound is a safe and effective treatment modality
ration after opening-wedge high tibial osteotomy by for achieving pain relief and improvement of function in
hemicallotasis5. patients with KOA14,15. Further research is required to
Research has suggested that ultrasound is likely to investigate its long-term efficacy16. Köybaşi et al. recom-
increase type II collagen synthesis in articular cartilage via mended using therapeutic ultrasound in the management
activation of chondrocytes and induction of type II col- of hip osteoarthritis, and believed that large clinical trials
lagen mRNA expression6. Ultrasound stimulates chondro- on ultrasound were necessary in order to standardize the
cyte proliferation and matrix production in chondrocytes treatment modality in this patient group17. Because the
of human articular cartilage in vitro. Therefore, ultra- extent to which ultrasound produces pain relief and func-
sound might provide a feasible tool for cartilage tissue tional improvement in KOA patients is still uncertain,
repair in osteoarthritic patients7. further study is needed18.
Matrix metalloproteinase-9 (MMP-9), a glycosylated Ultrasound is not effective in the absence of sufficient
protease secreted by stromal cells8, has the specific capa- functional chondrocytes. In the later stages of OA, chon-
bilities of degrading and causing degeneration in type I, drocytes gradually degenerate and undergo apoptosis,
II and III collagen proteins. MMP-9, the amount of resulting in a reduction in the number of number of func-
which increases with worsening of osteoarthritis, is an tional cartilage cells. Two studies of osteoarthritis induced
important biological indicator of osteoarthritis. Ultra- in animals have suggested that ultrasound can enhance
sound can enhance cell metabolism9, promote cartilage cartilage repair in early osteoarthritis19,20. Therefore the
matrix components synthesis such as proteoglycan, effect of ultrasound would be significantly greater in
chondroitin sulfate10, inhibit the production of cartilage patients with early osteoarthritis than in those with severe
degrading enzymes, and reduce cartilage damage and osteoarthritis21,22.

© 2011 Tianjin Hospital and Blackwell Publishing Asia Pty Ltd

186 P Yang et al, Ultrasound in osteoarthritis

Conclusion 7. Korstjens CM, van der Rijt RH, Albers GH, et al. Low-
intensity pulsed ultrasound affects human articular
This study suggests that ultrasound is beneficial for chondrocytes in vitro. Med Biol Eng Comput, 2008,
KOA, as assessed by evaluation indexes, such as those 46: 1263–1270.
derived from VAS and Lequesne scores. The effect of mock 8. Fujiki M, Shineha J, Yamanokuchi K, et al. Effects of
treatment on the placebo group was statistically signifi- treatment with polysulfated glycosaminoglycan on
cant, suggesting that psychological factors may play a role. serum cartilage oligomeric matrix protein and
The treatment group had better results than the control C-reactive protein concentrations, serum matrix
group and this difference was statistically different, metalloproteinase-2 and -9 activities, and lameness in
showing that ultrasound has a clear therapeutic effect on dogs with osteoarthritis. Am J Vet Res, 2007, 68:
KOA. However, the target and mechanism of the beneficial 827–833.
effects of ultrasound on OA are still not entirely clear, and 9. Warden SJ, Favaloro JM, Bennell KL, et al. Low-
require further study. An important direction for future intensity pulsed ultrasound stimulates a bone-forming
response in UMR-106 cells. Biochem Biophys Res
research is elucidation of the biologic impact of various
Commun, 2001, 286: 443–450.
ultrasound waveform variables, such as the peak pulse
10. Nishikori T, Ochi M, Uchio Y, et al. Effects of low-
intensity23,24.
intensity pulsed ultrasound on proliferation and chon-
droitin sulfate synthesis of cultured chondrocytes
embedded in Atelocollagen gel. J Biomed Mater Res,
Acknowledgments 2002, 59: 201–206.
Support from the Second Affiliated Hospital of Chong- 11. Kaneko S, Satoh T, Chiba J, et al. Interleukin-6 and
qing Medical University and of the National Engineering interleukin-8 levels in serum and synovial fluid of
patients with osteoarthritis. Cytokines Cell Mol Ther,
Research Center of Ultrasound Medicine is gratefully
2000, 6: 71–79.
acknowledged.
12. Weishaupt D, Schweitzer ME, Rawool NM, et al. Indi-
rect MR arthrography of the knee: effects of low-
intensity ultrasound on the diffusion rate of
Disclosure intravenously administered Gd-DTPA in healthy volun-
No benefits in any form have been, or will be, received teers. Invest Radiol, 2001, 36: 493–499.
from a commercial party related directly or indirectly to 13. Tang J, Huang LK, Li D, et al. Efficacy of low-intensity
the subject of this manuscript. pulsed ultrasound in the treatment of knee osteoar-
thritis. J Third Mil Med Univ, 2010, 32: 2562–2564.
14. Tascioglu F, Kuzgun S, Armagan O, et al. Short-term
effectiveness of ultrasound therapy in knee osteoar-
References thritis. J Int Med Res, 2010, 38: 1233–1242.
1. Zhang HB. Gene treatment of osteoarthritis. Zhong- 15. Ozgönenel L, Aytekin E, Durmuşoglu G. A double-
guo Lin Chuang Kang Fu, 2003, 7: 4390–4391. blind trial of clinical effects of therapeutic ultrasound
2. Cook SD, Salkeld SL, Popich-Patron LS, et al. Improved in knee osteoarthritis. Ultrasound Med Biol, 2009, 35:
cartilage repair after treatment with low-intensity 44–49.
pulsed ultrasound. Clin Orthop Relat Res, 2001, 391 16. Loyola-Sánchez A, Richardson J, MacIntyre NJ. Efficacy
(Suppl. 1): S231–S243. of ultrasound therapy for the management of knee
3. Altman RD. The classification of osteoarthritis. J Rheu- osteoarthritis: a systematic review with meta-analysis.
matol Suppl, 1995, 43: 42–43. Osteoarthritis Cartilage, 2010, 18: 1117–1126.
4. Jia XL, Chen WZ, Zhou K, et al. Effects of low-intensity 17. Köybaşi M, Borman P, Kocaoğlu S, et al. The effect of
pulsed ultrasound in repairing injured articular carti- additional therapeutic ultrasound in patients with
lage. Chin J Traumatol, 2005, 8: 175–178. primary hip osteoarthritis: a randomized placebo-
5. Tsumaki N, Kakiuchi M, Sasaki J, et al. Low-intensity controlled study. Clin Rheumatol, 2010, 29: 1387–
pulsed ultrasound accelerates maturation of callus in 1394.
patients treated with opening-wedge high tibial 18. Rutjes AW, Nüesch E, Sterchi R, et al. Therapeutic
osteotomy by hemicallotasis. J Bone Joint Surg Am, ultrasound for osteoarthritis of the knee or hip.
2004, 86: 2399–2405. Cochrane Database Syst Rev, 2010, (1): CD003132.
6. Naito K, Watari T, Muta T, et al. Low-intensity pulsed 19. Choi BH, Woo JI, Min BH, et al. Low-intensity ultra-
ultrasound (LIPUS) increases the articular cartilage type sound stimulates the viability and matrix gene expres-
II collagen in a rat osteoarthritis model. J Orthop Res, sion of human articular chondrocytes in alginate bead
2010, 28: 361–369. culture. J Biomed Mater Res A, 2006, 15: 858–864.

© 2011 Tianjin Hospital and Blackwell Publishing Asia Pty Ltd

Orthopaedic Surgery (2011), Volume 3, No. 3, 181–187 187

20. Min BH, Woo JI, Cho HS, et al. Effects of low-intensity low-intensity ultrasound treatment. Osteoarthritis Car-
ultrasound (LIUS) stimulation on human cartilage tilage, 2010, 18: 724–733.
explants. Scand J Rheumatol, 2006, 35: 305–311. 23. Srbely JZ. Ultrasound in the management of osteoar-
21. Wang XH, Wang H, Yang SH, et al. Effect of low thritis: part I: a review of the current literature. J Can
intensity pulsed ultrasound in repairing rabbits carti- Chiropr Assoc, 2008, 52: 30–37.
lage injury caused by experimental osteoarthritis. Int J 24. Nyborg WL. Optimization of exposure conditions for
Orthop, 2009, 30: 191–193. medical ultrasound. Ultrasound Med Biol, 1985, 11:
22. Gurkan I, Ranganathan A, Yang X, et al. Modification 245–260.
of osteoarthritis in the guinea pig with pulsed

© 2011 Tianjin Hospital and Blackwell Publishing Asia Pty Ltd