European Journal of Pain 13 (2009) 3–10 Contents lists available at ScienceDirect European Journal of Pain journal homepage: www.EuropeanJournalPain.com Review Acupuncture and dry needling in the management of myofascial trigger point pain: A systematic review and meta-analysis of randomised controlled trials Elizabeth A. Tough a,*, Adrian R. White a, T. Michael Cummings b, Suzanne H. Richards a, John L. Campbell a a Primary Care, Peninsula Medical School, Universities of Exeter and Plymouth, Room N32, ITTC Building, Tamar Science Park, Plymouth PL6 8BX, UK b British Medical Acupuncture Society, Royal London Homeopathic Hospital, UK article info abstract Article history: Pain from myofascial trigger points is often treated by needling, with or without injection, although evi- Received 24 October 2007 dence is inconclusive on whether this is effective. We aimed to review the current evidence on needling Received in revised form 17 January 2008 without injection, by conducting a systematic literature review. Accepted 19 February 2008 We searched electronic databases to identify relevant randomised controlled trials, and included stud- Available online 18 April 2008 ies where at least one group were treated by needling directly into the myofascial trigger points, and where the control was either no treatment, or usual care; indirect local dry needling or some form of pla- Keywords: cebo intervention. We extracted data on pain, using VAS scores as the standard. Acupuncture Seven studies were included. One study concluded that direct dry needling was superior to no interven- Myofascial trigger point pain Systematic review tion. Two studies, comparing direct dry needling to needling elsewhere in the muscle, produced contra- Randomised controlled trials dictory results. Four studies used a placebo control and were included in a meta-analysis. Combining Meta-analysis these studies (n = 134), needling was not found to be significantly superior to placebo (standardised mean difference, 14.9 [95%CI, À5.81 to 33.99]), however marked statistical heterogeneity was present (I2 = 88%). In conclusion, there is limited evidence deriving from one study that deep needling directly into myo- fascial trigger points has an overall treatment effect when compared with standardised care. Whilst the result of the meta-analysis of needling compared with placebo controls does not attain statistically sig- nificant, the overall direction could be compatible with a treatment effect of dry needling on myofascial trigger point pain. However, the limited sample size and poor quality of these studies highlights and sup- ports the need for large scale, good quality placebo controlled trials in this area. Ó 2008 European Federation of Chapters of the International Association for the Study of Pain. Published by Elsevier Ltd. All rights reserved. 1. Introduction cause local tenderness and referred pain (Simons et al., 1999). MTrPs are thought to develop in muscles in any part of the Myofascial trigger points (MTrPs) have been described as a body, in response to sudden injury or muscle overload. It has ‘common cause of pain in clinical practice’ Gerwin et al. (1997) been hypothesised that the injured muscle fibres shorten (form- and an ‘extremely common, yet commonly overlooked’ source of ing taut bands) either in response to excessive amounts of cal- musculoskeletal pain (Simons, 2002). Epidemiological studies from cium ions being released from within the damaged fibres, or in the US have claimed that MTrPs were the primary source of pain in response to the corresponding motor end plate releasing exces- 30–85% of patients attending on account of pain in primary care or sive amounts of acetylcholine (Simons et al., 1999). Local ten- specialist pain clinic settings (Fricton et al., 1985; Fishbain et al., derness and referred pain ensues as muscle nociceptors are 1986; Skootsky et al., 1989). MTrP pain may therefore constitute stimulated in response to reduced oxygen levels and increased a substantial burden for both individual patients and for society inflammatory chemicals present at the site of injury (Simons as a whole. et al., 1999). MTrPs have been defined as hyperirritable points located in Although the existence of MTrPs remains controversial, they taut bands of skeletal muscle or fascia which when compressed do provide a basis for treatments and are an active topic for clinical research (Borg-Stein and Stein, 1996; Tunks and Crook, 1999; Borg-Stein and Simons, 2002). A recent systematic review identified 73 clinical intervention trials, of which the most com- * Corresponding author. Tel.: +44 (0) 1752 764448; fax: +44 (0) 1752 764234. E-mail address: [email protected] (E.A. Tough). mon intervention examined was needling directly into the MTrP, 1090-3801/$34.00 Ó 2008 European Federation of Chapters of the International Association for the Study of Pain. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.ejpain.2008.02.006 4 E.A. Tough et al. / European Journal of Pain 13 (2009) 3–10 with or without injection of local anaesthetic, cortisone or bot- 2.2. Study selection ulinum toxin (Tough et al., 2007). Although the mechanism of effect is unknown, the practice of inserting needles into points Two authors (from ET, AW and MC) made a preliminary deci- of soft tissue tenderness to alleviate pain is long established, sion on which papers might be eligible. Studies were excluded if with clinicians commonly adopting either the orthodox ap- the ‘active treatment’ involved inserting needles: (i) superficially proach of injection or the Western medical acupuncture ap- over the site of a MTrP; (ii) into traditional acupuncture points; proach of dry needling. Indeed, one of the first studies (iii) into pre-specified MTrP locations, since our hypothesis con- investigating trigger point dry needling, concluded that dry nee- cerned treating clinical identified MTrPs. We included RCTs where dling was effective at alleviating chronic myofascial pain (Lewit, at least one intervention group were treated by direct insertion of a 1979). dry needle into the MTrPs after locating the patient’s area of ten- Nonetheless, a previous systematic review of dry needling and derness. After further scrutiny by all authors we excluded papers injection therapy for MTrP pain found no evidence that injection where the control intervention was considered to be an ‘active’ of any substance elicited a superior response compared with the treatment, classified as: (i) oral medication (ii) an injected sub- insertion of a needle alone (Cummings and White, 2001). Further- stance or (iii) traditional meridian acupuncture needling – in the more, the evidence at that time was inconclusive on whether or light of laboratory and radiological evidence which shows a direct not needling is effective. association between acupuncture and the stimulation of pain Our initial aim was to update the systematic review of Cum- inhibitory mechanisms (Han and Terenius, 1982; Pariente et al., mings and White (2001) and, in view of the large number of stud- 2005). ies now available, and the clinical heterogeneity of the interventions (e.g. needling superficially over the site of MTrPs or needling traditional acupuncture points) we decided to report 2.3. Data extraction the evidence on different interventions separately. In this paper, we focus on examining whether or not direct dry needling of Two authors (from ET, AW and MC) independently ex- MTrPs that have been identified clinically, in the absence of any tracted data from each paper before data were entered onto other potentially active treatment, is effective at reducing pain a piloted electronic spread sheet. Discrepancies remaining in for patients with a diagnosis of MTrP pain. data extraction were resolved with discussion between the two reviewers and where necessary the third reviewer adjudicated. 2. Methods We extracted data on pain outcomes which reported a visual analogue scale (VAS) or comparable pain score as the principle out- We undertook a systematic review of the literature to identify- come measure. Where available, we extracted changes in mean ing whether or not dry needling directly into MTrPs achieved supe- scores and the associated standard deviations. Data in graphs were rior pain reduction in patients with a diagnosis of MTrP pain when extracted by measuring with a ruler. Pressure pain threshold compared with either: no additional intervention; indirect local algometry readings were classified as a secondary outcome dry needling either superficially over the MTrP or elsewhere in measure. the muscle; or a placebo control such as a non-penetrating sham We classified an outcome measure as ‘long-term’ if it was ap- needle or sham laser. plied 1–6 months after the final reported treatment (taking the measure assessed closest to 3 months) and ‘short-term’ if applied 2.1. Search 24 h to 30 days after the final reported treatment (but a mini- mum of 72 h after the first treatment because of the potential In April 2007 we searched the following electronic databases for post treatment soreness) taking the measure assessed closest sequentially: Pubmed (from 1966); a combined search of EM- to 7 days. BASE (from 1974), AMED (from 1985) and MEDLINE (from For cross-over studies we analysed the results of the first 1950); Cochrane Central/Cochrane Reviews (from inception); arm only unless we agreed the data showed no carry-over ef- PEDro (from 1929) and SCI-EXPANDED (from 1970). Searches fect. In RCTs that tested multiple interventions, we included were limited (where database facilities allowed) to randomised only the treatment groups which met our criteria for controlled trials (RCTs) or clinical trials, reviews and human inclusion. studies. There was no language restriction. The search terms Additional data were extracted on study design and setting, used were: ‘myofascial pain’ OR ‘myofascial pain syndrome’ OR inclusion/exclusion criteria, the condition being treated, includ- ‘trigger point’ OR ‘trigger points’ and then in turn acup*, needl*, ing duration of symptoms and criteria used for diagnosis, and inject*, block*, *caine, tox* or percutaneous neuromodulation the results as reported.