MD, PhD, FMedSci, FSB, FRCP, FRCPEd

One of the questions I hear regularly is ‘HOW DO THE EFFECTS OF THIS ALTERNATIVE TREATMENT COMPARE TO THOSE OF CONVENTIONAL OPTIONS’? Take acupuncture in the management of osteoarthritis, for instance. There is some encouraging evidence suggesting it might help. The most recent systematic review that I know of concluded that “acupuncture provided significantly better relief from knee osteoarthritis pain and a larger improvement in function than sham acupuncture, standard care treatment, or waiting for further treatment.” However, in order to estimate its value in practice, we ought to know whether it is as good as or perhaps even better than standard treatments. In other words, what we really want to know is its relative effectiveness.

Data to evaluate the relative effectiveness of acupuncture or other alternative therapies are hard to come by. Ideally, one would require clinical trials which provide direct comparisons between the alternative and the conventional therapy. Sadly, such studies are scarce or even non-existent. Therefore we might have to rely on more indirect evidence. A new paper could be a step in the right direction.

The aim of this systematic review was to critically evaluate existing osteoarthritis (OA) management guidelines to better understand potential issues and barriers.

A systematic review of the literature in MEDLINE published from January 1, 2000 to April 1, 2013 was performed and supplemented by bibliographic reviews, following PRISMA guidelines and a written protocol. Following initial title and abstract screening, two authors independently reviewed full-text articles; a third settled disagreements. Two independent reviewers extracted data into a standardized form. Two authors independently assessed guideline quality; three generated summary recommendations based on the extracted guideline data.

Overall, 16 articles were included in the final review. There was broad agreement on recommendations by the various organizations. For non-pharmacologic modalities, education/self-management, exercise, weight loss if overweight, walking aids as indicated, and thermal modalities were widely recommended. For appropriate patients, joint replacement was recommended; arthroscopy with debridement was not recommended for symptomatic knee OA. Pharmacologic modalities most recommended included acetaminophen/paracetamol for first line treatment and oral or topical NSAIDs for second line therapy. Intra-articular corticosteroids were generally recommended for hip and knee OA. Controversy remains about the use of acupuncture, knee braces, heel wedges, intra-articular hyaluronans, and glucosamine/chondroitin.

I think that this tells us fairly clearly that, compared to other options, acupuncture is not considered to be an overwhelmingly effective treatment for osteoarthritis by those who understand that condition best. Several other therapies seem to be preferable because the evidence is clearer and stronger and their effect sizes is larger. This, I think begs the question whether it is in the best interest of patients or indeed ethical to ignore this knowledge and recommend acupuncture as a treatment of osteoarthritis.

More generally speaking, we should always bear in mind that it is not enough proving a therapy to be effective; we usually also need to consider what else is on offer. And if you think that this is rather complex, you are, of course, correct – but wait until someone mentions issues such as safety and cost of all the relevant therapeutic options.

30 Responses to The relative effectiveness of acupuncture for osteoarthritis

  • I don’t see how acupuncture can be deemed in any way effective, (nor have I seen any reliable studies that establish this) as there is no prior plausibility to explain how sticking needles into imaginary “points” could affect any disease or symptom. It is tooth fairy science, pseudo science, or just plain quackery–whichever you prefer. It is pointless to continue to fund studies of such nonsense.

    • You need to look into the evidence properly.

      • What would you call “properly” pray tell?

        See sciencebasedmedicine.org for details of my education. I understand that the popular press often puts out articles that some study has verified acupuncture for a narrow range of ailments (nausea, some types of pain), but these simply do not hold up under proper scientific scrutiny–and the things credited to acupuncture are very narrow to begin with–hardly the wide range of ailments that most practitioners claim.

  • Here’s another new study showing no significant difference between ‘real’ acupuncture and non-inserting sham acupuncture for osteoarthritis: http://www.healio.com/rheumatology/practice-management/news/online/%7Bf6d64fb0-44d2-4e47-abe4-e068b69e24b5%7D/acupuncture-needle-types-equally-effective-with-exercise-for-knee-oa

    “Puncturing and nonpuncturing acupuncture needles were equally effective along with exercise-based physical therapy as treatment for patients with knee osteoarthritis, according to study results…

    In the WOMAC response rate, there was no significant difference between standard needle (31.6%) and non-penetrating needle (30.3%) acupuncture. Both treatment groups displayed improvement from therapy (P=0.5). Most adverse events were mild, with 47 reported by the true acupuncture cohort compared with 31 in the control group (P=.08).

    Using multivariate logistic regression, there was an association between a positive expectation from acupuncture and reported improvement (adjusted OR=2.14; 95% CI, 1.13-4.10), with no differences noted by patients’ race, sex or age.”

    Note the use of the term “equally effective” when “equally ineffective” would be just as appropriate.

    • It would be useful to know what was done with the needles once they were inserted in the puncturing group. We’re they just tapped in and left? Was ‘de Qi’ elicited, and if so what kind of de Qi? We’re the acupuncturists experienced in treating these conditions? Was palpation used to select tender points? Was there diferentiation used at all between patients? None of these (crucial, in my opinion) factors are mentioned.

      • Also, moxa would very often be indicated for osteoarthritis. The Chinese word for acupuncture actually means ‘needle moxa’ – they are two strands of the same treatment and one or the other or both should be used depending on careful assessment of the patient.

      • yes, in your opinion…!!!

        • You are welcome to rest your case whenever you like. I choose to base my understanding of the world on personal experience, the experiences of others I trust, as well as the critical evaluation of scientific evidence. I actively avoid forming firm opinions based solely on scientific evidence because it is only one way to understand reality.

      • Please locate “qi” for me on an anatomical chart–or anywhere else. Perhaps you could also find the fairies in my garden?

        • Edzard – I don’t claim to know all the answers by any means, but I’d like to think I’m trustworthy. Are you saying I shouldn’t be trusted because I don’t base my view of the world entirely on the scientific method?

        • Irene – Qi can’t be ‘located’ because it is a philosophical idea rather than something measurable. If you’d like to learn more about it I’m happy to suggest some starting points. I can’t help with your fairy problem though I’m afraid.

          • If qi is only an idea and not something that is measurable, what’s the mechanism for affecting it?

          • quantum-mechanics, no doubt

          • Alan – Qi, Yin and Yang etc. and philosophical ideas which aim to describe and understand the patterns common throughout nature. In traditional medicine, these theories help guide the practitioner towards effective treatments. Because these are universal theories, the ‘mechanisms for affecting them’ are infinite. You’d have to study something like the Dao De Jing to begin to properly understand this (I’m only scratching the surface myself). If you mean ‘how does acupuncture work’ from the biomedical perspective, there is interesting work being done by people like Dr Ed Neal relating to the manipulation of blood supply and fascial communication. If you’re genuinely interested this introduction video may be useful: http://www.youtube.com/watch?v=BRHkPIO-4C4

          • Tom Kennedy said:

            Qi, Yin and Yang etc. and philosophical ideas which aim to describe and understand the patterns common throughout nature.

            What, precisely, do they describe and what are the patterns common throughout nature?

            In traditional medicine, these theories help guide the practitioner towards effective treatments.

            What’s the evidence that substantiates those notions as theories? Would it me more correct to call them hypotheses?

            Because these are universal theories

            What’s the evidence that these ‘theories’ are universal?

            the ‘mechanisms for affecting them’ are infinite.

            Name one.
            All you seem to have done is demand more questions be asked.

          • Unfortuantely the Chinese goverment decided to adopt the pinyin system to deceive our understanding but if we used the Wade-Giles we’d enlighteningly transcribe Ch’i thence it becomes clear both the meaning and the hitch in detecting it. Ch’i is not pure energy but energy times space, as a consequence its dimensions are J·m and being Planck costant so small we get a magnitude of ~ 1.99 ×10−25 (hardly detectable out of atomic laboratory)!
            Moreover we have to take the first order derivative, probably a partial one so I guess here comes acupuncturists experience in treating these boundary conditions (the deeper our analysis the wider our understanding).
            Besides we are dealing only with the imaginary part, here being i = √−1, (sorry if the analysis goes from left to right but I am using Western evidence)
            because as properly deduced from the Dao De Jing the Real Ch’i can’t be named.
            I wonder when ‘de(Ch’i) ’ is elicited do you get a total differential ?
            Moxibustion obviously being an “integral” part of the system, I bet we can dive deeper in QM thinking of meridians as (Feynman) path integrals.
            Q.E.D.

          • obviously!

  • Is it really surprising that controversy remains about the use of acupuncture in mainstream health? There are all kinds of misconceptions and negative associations tied up with acupuncture for many Western doctors, and the road towards true integration is a long and difficult one. Because certain current guidelines are tentative doesn’t for a minute mean that acupuncture is not effective (when properly delivered). However I totally agree it would be wonderful to have more comparative studies – let’s hope they eventuate.

    Safely and cost issues should also be considered.

    • There have been hundreds of studies to date (perhaps more) with no high-standard ones showing any benefit beyond placebo.

      sciencebasedmedicine.org for a thorough discussion. I’m sorry to upset your closely held belief, but that’s what it is–a belief.

      • I suppose that depends on your definition of ‘high-standard’. Here’s one recent meta-analysis of ‘high-quality RCTs’ which concludes acupuncture shows an effect statistically superior to placebo: http://archinte.jamanetwork.com/article.aspx?articleid=1357513

        • “However, these differences are relatively modest, suggesting that factors in addition to the specific effects of needling are important contributors to the therapeutic effects of acupuncture.”
          this was the most open admission yet by acupuncture-proponents that acupuncture relies almost exclusively on non-specific effects!!!

          • They say ‘relatively modest’, you say ‘almost exclusively…non-specific effects’. The fact is, acupuncture has been shown here by high quality trials to exhibit effects beyond placebo. I believe very good acupuncture to be a highly refined skill, and that these effect sizes can be much greater.

Leave a Reply

Your email address will not be published. Required fields are marked *

Please answer the following: *

Recent Comments

Note that comments can be edited for up to five minutes after they are first submitted.


Click here for a comprehensive list of recent comments.

Categories