Last year, JOSPT published a systematic review and meta-analysis of dry needling for upper quarter myofascial pain. We thought it was a really well conducted systematic review, but we were somewhat surprised by the abstract, the interpretation of results and the accompanying advice to patients. So we wrote a letter to the journal. They did not publish it because they felt it was too long after publication, which is fair enough. So, we are posting it, pretty much as is, here[9].
We think that systematic review and meta-analysis is currently the best methodology available to assess the state of evidence concerning clinical interventions. This contribution[9], on the effectiveness of dry needling for upper-quarter myofascial pain, appears to be no exception. We strongly believe that the scientific method is only complete once the results have been interpreted and communicated to the end user, and done so in light of the strengths and limitations of the identified evidence and in consideration of the differences in specialised knowledge between the scientists and their audience (that is, without jargon). On that matter, we have been warned of the perils of taking abstracts on face value[2], and, in line with such warnings, we have significant concerns regarding the interpretation and subsequent recommendations that were offered by the authors, and by the accompanying advice to patients[8].
In the abstract, arguably the major determinant of clinicians’ impressions of a study[6], and in the conclusions, the authors make a clear clinical recommendation in support of dry needling based upon what they report as “grade A evidence”. In a companion article ‘JOSPT perspectives for patients’, those recommendations are repeated: “..dry needling can be effective in providing pain relief….more research needs to be done to determine whether dry needling is better for this condition than other treatment options” (p 635). When we read the full article, we found these recommendations surprising. The article itself suggests the available data are flimsy (not least because of the so-called garbage in – garbage out phenomenon – if the original articles are crap, then the best the systematic review can do is to collate crap), but if anything, the systematic review actually showed that dry needling is probably no better than sham and very possibly, it is measurably worse than comparative treatments.
We contend therefore, that the recommendations made in the abstract and in the patient-targeted advice, do not fairly represent the data provided in the article. Given the severe limitations of the evidence-base, in terms of size, quality and methodological consistency, the proposed effect is not robust. When comparing dry needling to control/sham, there appears to be a short-term effect of dry needling, which is not significant at four-week follow up. The studies with the largest samples[4,5] show no statistically significant effect of dry needling at either time-point. Moreover, the study that reported the greatest effect[3] clearly stands apart from the other included studies, possibly because it failed to apply a convincing control/sham intervention. Removing that single study from the analysis renders the result non-significant. When compared to other interventions included in the review, dry needling was less effective than comparison interventions in the immediate term and no different at four weeks. One must interpret that result with caution too, for the very same reasons pointed out by the authors, but it clearly does not point towards recommendation of dry needling, and certainly not at ‘Grade A’ level.
There is another intriguing problem here – one that is a nuisance for all of us engaged with evidence-appraisal – that of giving due consideration to the null hypothesis. In any meta-analysis of trials of mixed quality, with subjective outcomes, where blinding can be hard to ensure, the purists agree that even a completely useless treatment will probably give a picture of unreliable and marginal results, trending towards favouring the treatment. The predominance of small trials in such an analysis introduces a further bias, substantially increasing the risk of false positive results and exaggerated effect sizes[1,7]. That is, we need to see data substantially better than trending to be satisfied the treatment is better than useless (but see note below).
We contend that a far more parsimonious interpretation of the meta-analysis is that dry needling is not convincingly superior to sham/control conditions and possibly worse than comparative interventions. At best, the effectiveness of dry needling remains uncertain but, perhaps more likely, dry needling may be ineffective. To base a clinical recommendation for dry needling on such fragile evidence seems rash and risks exposing patients to an unnecessary invasive procedure.
The note below: We should always remember that ‘completely useless’ is usually relative to something, not nothing. We think that is important because a treatment that is completely useless compared to another treatment, for example a ‘sham’, may actually be useful but for reasons not related to the component that is supposedly effective. That is, let’s say, for argument’s sake, that dry needling is indeed no better than sham dry needling. It is possible that both are better than nothing, which implies something else, wrapped up in the dry needling, might be very useful indeed.
Daniel S. Harvie, M Musc & Sports Physio
PhD candidate
Samson Institute for Health Research
University of South Australia
Neil O’Connell, PhD
Lecturer in physiotherapy
Brunel University
Uxbridge, UK
Lorimer Moseley, PhD, FACP
Professor of clinical neuroscience
Samson Institute for Health Research,
University of South Australia and Neuroscience Research Australia
References
1. Dechartres A, Trinquart L, Boutron I, Ravaud P. Influence of trial sample size on treatment effect estimates: metaepidemiological study. British Medical Journal 2013;346:f2304.
2. Editorial. “Read MEDLINE abstracts with a pinch of salt.” The Lancet 2006; 368: 1394.
3. Hsieh YL, Kao MJ, Kuan TS, Chen SM, Chen JT, Hong CZ. Dry needling to a key myofascial trigger point may reduce the irritability of satellite MTrPs. Am J Phys Med Rehabil. 2007; 86:397-403. http://dx.doi.org/10.1097/ PHM.0b013e31804a554d
4. Ilbuldu E, Cakmak A, Disci R, Aydin R. Comparison of laser, dry needling, and placebo laser treatments in myofascial pain syndrome. Photomed Laser Surg. 2004; 22:306-311. http:// dx.doi.org/10.1089/1549541041797878
5. Irnich D, Behrens N, Gleditsch JM, et al. Immediate effects of dry needling and acupuncture at distant points in chronic neck pain: results of a randomized, double-blind, sham-controlled crossover trial. Pain. 2002; 99:83-89.
6. Marcelo, A, et al. “A comparison of the accuracy of clinical decisions based on full-text articles and on journal abstracts alone: a study among residents in a tertiary care hospital.” Evidence Based Medicine. 2013; 18:48-53.
7. Moore RA, Eccleston C, Derry S, Wiffen P, Bell RF, Straube S et al. “Evidence” in chronic pain–establishing best practice in the reporting of systematic reviews. Pain 2010; 150:386-389.
8. Teyhen DS. “Painful and tender muscles: Dry needling can reduce myofascial pain related to trigger points. JOSPT 2014; 44:261-261.
9. Kietrys, D., Palombaro, K., Azzaretto, E., Hubler, R., Schaller, B., Schlussel, J., & Tucker, M. (2013). Effectiveness of Dry Needling for Upper-Quarter Myofascial Pain: A Systematic Review and Meta-analysis. JOSPT, 43 (9), 620-634 DOI: 10.2519/jospt.2013.4668