Acupuncture as a complementary therapy to the pharmacological treatment of osteoarthritis of the knee: randomised controlled trial
BMJ 2004; 329 doi: https://doi.org/10.1136/bmj.38238.601447.3A (Published 18 November 2004) Cite this as: BMJ 2004;329:1216All rapid responses
Rapid responses are electronic comments to the editor. They enable our users to debate issues raised in articles published on bmj.com. A rapid response is first posted online. If you need the URL (web address) of an individual response, simply click on the response headline and copy the URL from the browser window. A proportion of responses will, after editing, be published online and in the print journal as letters, which are indexed in PubMed. Rapid responses are not indexed in PubMed and they are not journal articles. The BMJ reserves the right to remove responses which are being wilfully misrepresented as published articles or when it is brought to our attention that a response spreads misinformation.
From March 2022, the word limit for rapid responses will be 600 words not including references and author details. We will no longer post responses that exceed this limit.
The word limit for letters selected from posted responses remains 300 words.
Re: Problems with Sham Acupuncture
The evaluation of the result variables was independent and blind, and
thus the evaluators were unaware of the distribution to the different
groups. Among the inclusion criteria was the absence of previous
experience of acupuncture, in order to avoid patients recognizing the
group to which they had been assigned.
All the participants, in the experimental and in the control groups,
completed every one of the treatment sessions. Specifically, in the
control group nobody abandoned a session before it had ended (everybody
“took the bait”). The patients who left the control group, whether because
their symptoms did not improve or for other, personal reasons, did not
state at any time that they were included within the placebo group.
Furthermore, these six persons who left due to a lack of improvement lend
force to the argument that the placebo really was inert.
Re: Questionable methods
In the basal analysis, no statistically significant or clinically relevant differences were found between the two groups. Moreover, we carried out a multivariate analysis, adjusted for the baseline variables, which revealed no significant changes between the experimental and the control groups concerning the assessment of improvement. The complementary radiographic tests were not repeated due to the degenerative, chronic and irreversible characteristics of the process (Ahlbäck score).
Re: Re: Acupunture for OA of the knee. Vas J et al BMJ 2004; 329;1216-9
In our case, as acupuncture is included as a healthcare provision within the Andalusian Health Service, studies were carried out to show improvements in the cost/benefit ratio. This resulted mainly from a reduction in the consumption of AINES and analgesics, and reduced side effects resulting from the latter [1,2]. Any stimulus provoked by a needle piercing the skin may be capable of activating the spinal nociceptive neurones and send an excitatory message to the nerve centres, thus causing a non-specific analgesic effect, due among other factors to the activation of the pain-suppression system in the spinal cords (Diffuse noxious inhibitory controls) [3-6]. By definition, the placebo must be inert, which is why we believe the control method decided upon (with no piercing of the skin) is the most appropriate.
Reference list
1. Vas J, Panadero Ruz J, Pérez Blanco JL, Méndez Martínez C, Gaspar Navarro O, Aguilar Domínguez I: Acupuntura. Una alternativa en el tratamiento del dolor. Atención Farmacéutica 2003, 5: 18-24.
2. Vas J, Méndez Martínez C, Vega E: Tratamiento con acupuntura del dolor crónico benigno. Medicina de Familia (Andalucia) 2002, 3: 239-244.
3. Le BD, Dickenson AH, Besson JM: Diffuse noxious inhibitory controls (DNIC). II. Lack of effect on non-convergent neurones, supraspinal involvement and theoretical implications. Pain 1979, 6: 305- 327.
4. Le BD, Villanueva L, Bouhassira D, Willer JC: Diffuse noxious inhibitory controls (DNIC) in animals and in man. Patol Fiziol Eksp Ter 1992, 55-65.
5. Bing Z, Cesselin F, Bourgoin S, Clot AM, Hamon M, Le BD: Acupuncture-like stimulation induces a heterosegmental release of Met- enkephalin-like material in the rat spinal cord. Pain 1991, 47: 71-77.
6. Bing Z, Villanueva L, Le BD: Acupuncture-evoked responses of subnucleus reticularis dorsalis neurons in the rat medulla. Neuroscience 1991, 44: 693-703.
Re: Acupunture in knee arthroplasty?--No role
Any stimulus provoked by a needle piercing the skin may be capable of activating the spinal nociceptive neurones and send an excitatory message to the nerve centres, thus causing a non-specific analgesic effect, due among other factors to the activation of the pain-suppression system in the spinal cords (Diffuse noxious inhibitory controls) [1-4]. By definition, the placebo must be inert, which is why we believe the control method decided upon (with no piercing of the skin) is the most appropriate. The acupuncture performed in our healthcare unit was in line with all the recommendations for good clinical practice. Only aseptic, single-use materials were employed.
Reference List
1. Le BD, Dickenson AH, Besson JM: Diffuse noxious inhibitory controls (DNIC). II. Lack of effect on non-convergent neurones, supraspinal involvement and theoretical implications. Pain 1979, 6: 305- 327.
2. Le BD, Villanueva L, Bouhassira D, Willer JC: Diffuse noxious inhibitory controls (DNIC) in animals and in man. Patol Fiziol Eksp Ter 1992, 55-65.
3. Bing Z, Cesselin F, Bourgoin S, Clot AM, Hamon M, Le BD: Acupuncture-like stimulation induces a heterosegmental release of Met- enkephalin-like material in the rat spinal cord. Pain 1991, 47: 71-77.
4. Bing Z, Villanueva L, Le BD: Acupuncture-evoked responses of subnucleus reticularis dorsalis neurons in the rat medulla. Neuroscience 1991, 44: 693-703.
Best Regards
Competing interests:
None declared
Competing interests: No competing interests
The evaluation of the result variables was independent and blind, and
thus the evaluators were unaware of the distribution to the different
groups. Among the inclusion criteria was the absence of previous
experience of acupuncture, in order to avoid patients recognizing the
group to which they had been assigned.
Any stimulus provoked by a needle piercing the skin may be capable of
activating the spinal nociceptive neurones and send an excitatory message
to the nerve centres, thus causing a non-specific analgesic effect, due
among other factors to the activation of the pain-suppression system in
the spinal cords (Diffuse noxious inhibitory controls) [1-4]. By
definition, the placebo must be inert, which is why we believe the control
method decided upon (with no piercing of the skin) is the most
appropriate.
Credibility and confidence in the treatment should have been measured;
this is a question that is recognized in the Discussion and will be taken
into account in subsequent research.
Best Regards
References
1. Le BD, Dickenson AH, Besson JM: Diffuse noxious inhibitory
controls (DNIC). II. Lack of effect on non-convergent neurones,
supraspinal involvement and theoretical implications. Pain 1979, 6: 305-
327.
2. Le BD, Villanueva L, Bouhassira D, Willer JC: Diffuse noxious
inhibitory controls (DNIC) in animals and in man. Patol Fiziol Eksp Ter
1992, 55-65.
3. Bing Z, Cesselin F, Bourgoin S, Clot AM, Hamon M, Le BD:
Acupuncture-like stimulation induces a heterosegmental release of Met-
enkephalin-like material in the rat spinal cord. Pain 1991, 47: 71-77.
4. Bing Z, Villanueva L, Le BD: Acupuncture-evoked responses of
subnucleus reticularis dorsalis neurons in the rat medulla. Neuroscience
1991, 44: 693-703.
Competing interests:
None declared
Competing interests: No competing interests
According to the categories of syndromes in traditional Chinese
medicine, arthritis of the knee is normally classified as a Cold/Humidity-
type Bi syndrome. In clinical practice, this condition would obviously
benefit from moxibustion, both direct and indirect. Nevertheless, this
technique was not adopted for the clinical trial, for two main reasons:
firstly, the study partly coincided with the summer months, in a city
where maximum temperatures of 45º may be reached. The excessive air
temperature made it unadvisable, then, to apply this particular technique,
while at the same time it would not have been practical in a clinical
study to apply a given technique only during certain periods of the year.
The second reason was derived from our wish to ensure that the real
technique and the placebo were indistinguishable, it being impossible to
apply a placebo for moxibustion due to the striking visibility of the
combustion of the artemisia cones.
Best Regards
Competing interests:
None declared
Competing interests: No competing interests
The data referring to the protocols for electrical stimulation were
excluded for reasons of space, and are given below. All the needles
inserted into the local points were stimulated electrically in pairs
according to the following protocol: pair 1) Yanglingquan GB34 –
Yinlingquan SP9; pair 2) Zusanli ST36 – Neixiyan EXLE 5. An alternate
frequency (2/15 Hz) electrostimulation was applied for 20 minutes each
session [1]. The stimulation intensity was as high as possible, just under
the pain threshold [1]. The stimulation intensity of the needles was
adjusted continuously during the procedure
Best Regards
References
1. Chen XH, Han JS: All three types of opioid receptors in the spinal
cord are important for 2/15 Hz electroacupuncture analgesia. Eur J
Pharmacol 1992, 211: 203-210.
Competing interests:
None declared
Competing interests: No competing interests
The evaluation of the result variables was independent and blind, and
thus the evaluators were unaware of the distribution to the different
groups. Among the inclusion criteria was the absence of previous
experience of acupuncture, in order to avoid patients recognizing the
group to which they had been assigned.
Best Regards
Competing interests:
None declared
Competing interests: No competing interests
Dear Sir/ Madam;
I have watched with interest the debate following the publication of these
very positive findings of the effects of Acupuncture on osteoarthritis.
The work confirms known clincial anecdotal evidence as well as supporting
the earlier work of the authors referred to by Bjordal and colleagues.
This paper provides significant support to the non pharmacological
treatments of osteoarthritis and thogh there are some small methodological
flaws in the research process the overall result is definitive. The
proposition on whether the electro stimulative effect exceeded the direct
acupuncture effect is an area worthy of further clincial research. It
could be that the concurrent dual method was cummulative and responsible
for the high effect Size that was noted.
In reading the comments of the rapid responders I am concerned that
several of the authors make their statements of 'No Competing interests'
yet themselves are emminent authors but with noted reserach grants from
pharaceutical companies.
Can a clarification be made that there are no competing interests that
would influence their negative comments on these significant clincial non
pharmacological treatments of osteoarthritis.
Competing interests:
None declared
Competing interests: No competing interests
A highly significant improvement in patient perceived health outcomes
is reported in this randomised controlled trial of acupuncture for knee
osteoarthritis, with a high participation rate (100%) and a low study
withdrawal rate (<10%).The WOMAC index(1), widely accepted as the
instrument of choice in the evaluation of treatments for knee
osteoarthritis, enables comparison with the effect sizes reported in large
numbers of other randomised controlled trials assessing a range of
interventions (2).
The participant baseline WOMAC scores represent severe group
disability and pain (1). In this study, acupuncture produced unusually
high standardised treatment effect sizes of 1.35, 0.61 and 1.27 for pain,
stiffness and function domains of the WOMAC respectively (2). The reported
mean difference in WOMAC scores after the intervention is overall greater
than after total knee arthroplasty (TKA)at 12 weeks (standardised
treatment effect sizes of 1.03, 0.63 and 1.06 for pain, stiffness and
function domains respectively) which is remarkable (3). In lay terms, a
group of people who could hardly manage a flight of stairs and were
experiencing severe daily pain before treatment had minimal disability and
pain within 12 weeks of combined diclofenac and acupuncture.
Treatment allocation was concealed from ‘independent evaluators’ in
this study, however there is no description of how and where the outcome
measures were administered. WOMAC was designed as a self-completion
questionnaire. Campbell and Dieppe have reported discrepancies between
patients’ assessment of outcome, when the WOMAC index is administered by a
clinician in a trial clinic (4).
So should we demand that our primary care trusts commission
acupuncture treatment for knee osteoarthtitis? An economic evaluation,
alongside multi-centre studies demonstrating generalisability and
reproducibility of theses results would be necessary to change health
policy. If cost-effective, then acupuncture may be the primary care
treatment of choice for severely disabled people with knee osteoarthritis,
especially if TKA is contraindicated.
References
1. Jinks C, Jordan K, Croft P. Measuring the population impact of
knee pain and disability with the Western Ontario and McMaster
Universities Osteoarthritis Index (WOMAC). Pain 2002;100:55-64
2. Jordan KM, Arden NK, Doherty M, Bannwarth B, Bijlsma JWJ, Dieppe P
et al. EULAR Recommendations 2003: an evidence based approach to the
management of knee osteoarthritis: Report of a Task Force of the Standing
Committee for International Clinical Studies Including Therapeutic Trials
(ESCISIT). Ann Rheum Dis 2003;62:1145-55.
3. Brazier JE, Harper R, Munro J, Walters SJ, Snaith ML. Generic and
condition-specific outcome measures for people with osteoarthritis of the
knee. Rheumatology 1999;38:870-7.
4. Campbell R, Quilty B, Dieppe P. Discrepancies between patients'
assessments of outcome: qualitative study nested within a randomised
controlled trial. BMJ 2003;326:252-3.
Competing interests:
None declared
Competing interests: No competing interests
In their rapid response, Ravaud & Dougados comment on what they
regard as “a very large effect size” that has ”never been observed with
chronic painful conditions”. They claim that the effect size obtained by
Vas et al. exceeds 2 (although they did not specify the absolute value)
and that this is higher than the highest ES (1.74) of any treatment
evaluated when the EULAR recommendations on knee osteoarthritis were
updated1. Interestingly, the highest ES of 1.74 in the EULAR update was
for acupuncture.
We have great respect for the important work of Ravaud & Dougados
in osteoarthritis research, but would like to make two comments to their
rapid response.
Firstly, it is unclear to us how they have arrived at an effect size
above 2 for pain. We thought that the ES calculations used in the EULAR
update were a composite of the mean differences in change for pain and
function rather than pain on its own. Still, it seems that differences
from this interpretation probably has little importance because results
for pain and function are closely correlated for many interventions.
To find the ES of the study by VAS et al., we have input the mean
difference in change for each group from the study by Vas et al. to meta-
analyses software calculations. The data for the Acupuncture group was
48.3 (i.e. change from baseline, SD = 15.6) and the Placebo group an
effect of 23.1 (i.e. change from baseline, SD = 29.7). With the software
package of Prof. Schwarzer used in the EULAR update, this gives us an ES
(i.e. difference in change between the groups) of 1.06 (Cohens d) and 1.05
(Hedges g). With the Comprehensive Meta-analysis software package from
Biostat Inc. we get an ES of 1.06. Therefore, we believe that the alleged
ES of > 2 forwarded by Ravaud&Dougados, probably is too high.
Secondly, Ravaud & Dougados claim that the results of Vas et al.
need confirmation before they can be trusted. We would like to point out
that the results of another electroacupuncture trial by Sangdee et al.
20022 of similar sample size seems to provide confirmation of a similar ES
as for the trial by Vas et al. For pain on VAS, Sangdee et al. reported a
mean difference of change of 25.3 mm (ES = 0.99) between
electroacupuncture (mean change 48.2 SD 24.3) and placebo (mean change
22.9, SD 27.0). This compares favourably with the 25.2 mm in the study by
Vas et al. In the same study, Sangdee et al. also found that
electroacupuncture was superior to the NSAID drug diclofenac by a mean
difference of 15.2 mm on VAS.
However, we still believe that the observed effects in both studies
could be due to the effects of segmental electrical stimulation, akin to
acupuncture-like transcutaneous electrical nerve stimulation, rather than
acupuncture-specific mechanisms.
References
1. Jordan KM, Arden NK, Doherty M, Bannwarth B, Bijlsma JW, Dieppe P,
et al. EULAR Recommendations 2003: an evidence based approach to the
management of knee osteoarthritis: Report of a Task Force of the Standing
Committee for International Clinical Studies Including Therapeutic Trials
(ESCISIT). Ann Rheum Dis 2003;62(12):1145-1155.
2. Sangdee C, Teekachunhatean S, Sananpanich K, Sugandhavesa N,
Chiewchantanakit S, Pojchamarnwiputh S, et al. Electroacupuncture versus
diclofenac in symptomatic treatment of osteoarthritis of the knee: a
randomized controlled trial. BMC Complement Altern Med 2002;2(1):3.
Competing interests:
None declared
Competing interests: No competing interests
Vas et al report on the impressive symptomatic treatment effect of
acupuncture in patients with knee osteoarthritis (OA) observed from the
results of a monocentric single blind randomised controlled trial (1).
Beyond the numerous methodological difficulties (blinding, choice of
control,…) inherent in the assessment of non pharmacological treatments in
OA (2), the most disturbing point is the size of treatment effect
observed.
The visual analogue pain score decreased from 58.9 mm (SD 11.2) to 10.6 mm
(SD 10.8) in the intervention group and from 60.3 mm (SD 13.7) to 37.2 mm
(26.3) in the control group which resulted in an effect size greater than
2.
This very large effect size exceeds that observed with any treatment
assessed in knee OA. According to the EULAR knee OA recommendation, effect
sizes vary from 0 to 1.74 for different treatments tested (pharmacological
including intra-articular corticosteroids or non- pharmacological) (3). To
our knowledge, such a great decrease of pain has never been observed with
chronic painful conditions. This result is even more remarkable if we
consider the stringent statistical analysis performed for handling the
missing data of control patients. In the control group, 16% of patients
were withdrawn and their final assigned score was the best score observed
in this group. In addition, the diclofenac consumption was approximately
60 % higher in the control group than in the intervention group.
Like others (4), we think that the main criteria for the quality of trials
is the repeatability (ie generating consistent results in different trials
evaluating the same intervention). Although, we are deeply convinced of
the interest of non pharmacological treatments in OA, we think that such
results need to be confirmed by others before being trusted.
1. Vas J, Mendez C, Perea-Milla E, Vega E et al. Acupuncture as a
complementary therapy to the pharmacological treatment of osteoarthritis
of the knee: randomised controlled trial. BMJ, 2004, 329:1216-1219.
2. Boutron I, Tubach F, Giraudeau B, Ravaud P. Methodological
differences in clinical trials evaluating nonpharmacological and
pharmacological treatments of hip and knee osteoarthritis. JAMA, 2003,
290:1062-1070.
3. Jordan KM, Arden NK, Doherty M et al. EULAR recommendations 2003:
an evidence based approach to the management of knee osteoarthritis:
report of a task force of the standing committee for international
clinical studies including therapeutic trials (ESCISIT). Ann Rheum Dis,
2003, 62:1145-1155.
4. Concato J, Horwitz RI. Beyond randomised versus observational
studies. The Lancet, 2004, 363:1660-1661.
Competing interests:
None declared
Competing interests: No competing interests
Blind leading the blind
The debate about blinding seems to have missed the crucial point that
patients will sense electroacupuncture [EA].
The authors state "The stimulation intensity [of EA] was as high as
possible, just under the pain threshold." [Rapid Response]. Surely, this
would compromise blinding. Patients would immediately guess that they were
getting active treatment, providing consent was truely informed and
patients had been told that there was a possibility of receiving a sham
intervention.
This information was absent from the published article because "The
data referring to the protocols for electrical stimulation were excluded
for reasons of space,..". [Rapid Response]
It is not clear whether this information was available to BMJ
referees when judging the merits of the study. Nevertheless, the
information about the use of EA as described in the published article
should have raised concerns about blinding, especially the authors' bold
claim that ".....factors reinforce our belief that the blinding procedure
applied was successful." [published article].
The BMJ should consider the use of the STRICTA recommendations [1]
when reviewing papers on acupuncture and it should ensure that full
details of technique-based intervetions are provided in the published
articles.
1. MacPherson H. et al. Standards for Reporting Intervetions in
Controlled Trials of Acupuncture. The STRICTA recommendations. Acupuncture
in Medicine 2002, 20(1): 22-25
Competing interests:
None declared
Competing interests: No competing interests