Although many conservative management options are being promoted for shoulder conditions, there is little evidence of their effectiveness. This review investigated one manual therapy approach, thrust manipulation, as a treatment option.
A systematic search was conducted of the electronic databases from inception to March 2016: PubMed, PEDro, ICL, CINAHL, and AMED. Two independent reviewers conducted the screening process to determine article eligibility. Inclusion criteria were manuscripts published in peer-reviewed journals with human participants of any age. The intervention included was thrust, or high-velocity low-amplitude, manipulative therapy directed to the shoulder and/or the regions of the cervical or thoracic spine. Studies investigating secondary shoulder pain or lacking diagnostic confirmation procedures were excluded. Methodological quality was assessed using the PEDro scale and the Cochrane risk-of-bias tool.
The initial search rendered 5041 articles. After screening titles and abstracts, 36 articles remained for full-text review. Six articles studying subacromial impingement syndrome met inclusion criteria. Four studies were randomized controlled trials (RCTs) and two were uncontrolled clinical studies. Five studies included one application of a thoracic spine thrust manipulation and one applied 8 treatments incorporating a shoulder joint thrust manipulation. Statistically significant improvements in pain scores were reported in all studies. Three of 4 RCTs compared a thrust manipulation to a sham, and statistical significance in pain reduction was found within the groups but not between them. Clinically meaningful changes in pain were inconsistent; three studies reported that scores met minimum clinically important difference, one reported scores did not, and two were unclear. Four studies found statistically significant improvements in disability; however, two were RCTs and did not find statistical significance between the active and sham groups.
The authors concluded that there is limited evidence to support or refute thrust manipulation as a solitary treatment for shoulder pain or disability associated with subacromial impingement syndrome. Studies consistently reported a reduction in pain and improvement in disability following thrust manipulation. In RCTs, active treatments were comparable to shams suggesting that addressing impingement issues by manipulation alone may not be effective. Thrust manipulative therapy appears not to be harmful, but AE reporting was not robust. Higher-quality studies with safety data, longer treatment periods and follow-up outcomes are needed to develop a stronger evidence-based foundation for thrust manipulation as a treatment for shoulder conditions.
This is yet another very odd conclusion from an otherwise almost acceptable analysis (but why include non-randomised studies on a subject where randomised trials are available?) . If pain reductions are found within groups but not between real and sham manipulation, the evidence is as clear as it can be: manipulations have no specific effects. In other words, they are a pure placebo therapy.
And what about this nonsense: there is limited evidence to support or refute thrust manipulation as a solitary treatment for shoulder pain? For responsible healthcare, we don’t need such weasel words, all we need is to stress loud and clear that there is no good positive evidence. This means the therapy is not evidence-based and we therefore should not recommend or use manipulation for shoulder pain.
But, in my view, the worst part in the conclusion section is this: thrust manipulative therapy appears not to be harmful, but AE reporting was not robust. Even if there had been adequate reporting of side-effects and even if this had not disclosed any problems, the safety of manipulation cannot be judged on the basis of such a small sample. Any responsible researcher should make it abundantly clear that the nasty habit by chiropractic pseudo-researchers of not reporting adverse effects is unethical and totally unacceptable.
My conclusion from all this: yet another attempt to white-wash a dodgy alternative therapy.
I agree that solitary treatment of subacromiol impingement shoulder pain( SISP) using only manipulation appears as good as just placebo. However it is commonly observered that patients presenting with SISP have anterior translation of shoulders and forward head translation. Chiropactic practices and procedures that address this postural distortion causing shoulder retraction via a reduction of forward head translation improve shoulder biomechanics and possibly the cause of SISP.
No evidence because he is a crackpot chiro from Sydney, Australia, who believes chiro is magic.
Edzard, have you ever tangled with Dunning? He’s a DPT, teaches joint manipulation and dry needling.
I can’t remember that I did
You follow him on twitter Edzard!
Dunning is the physio who’s runs courses on manipulation and dry needling and they call themselves “Osteopractors”! His courses are the largest post grad courses for physio’s in North America!
The fact that he invented (or bastardized) the term “Osteopractor” was something I was sure you would enjoy checking into. His team has a very powerful foothold in physical therapy in the US it seems and seems to be full steam ahead with full on spinal manipulation as a part of physical therapy, as well as extremity manipulation.
Shoulder joint thrust manipulation for impingement syndrome? I don’t think such is indicated or supported very much by the literature. This being said, I think the title of Edzard’s post should have referenced shoulder manipulation, not chiropractic in general; of course, he simply couldn’t resist the shock value of the title he chose. There are reasons why his title is misleading and reductionist. First, chiropractic generally refers to SMT, not to extremity manipulation. Second, mainstream chiropractices use physiotherapeutic approaches when treating impingement syndrome, not manipulation, and certainly not manipulation alone; Edzard’s implication to the contrary is unrealistic. Third, an analogous title to a post might read “Allergic to antibiotics? Don’t see an MD if you’re sick.” Of course such a title suggests that the MD(a prescribing physician) would rx an antibiotic to which the patient was allergic instead of one which would be effective; it’s as doubtful such would occur with the MD as it would be a chiropractic doctor who would use strictly manipulation instead of PT to treat impingement syndrome.
It’s tough to read between the lines of some of Edzard’s comments, but I try.
This review was carried out by chiropractors, and was published in Chiropractic and Manipulation Therapies.
A number of relevant studies were included in their review; so the practice in question clearly does go on. Presumably by some, if not all chiropractors, and perhaps other, if not all manual therapists. The findings of the review does not reflect well on this this practice, though the reviewers seem reluctant to say so.
The bottom line is this: This review is evidence of a manipulative practice being carried out with no support from the evidence base in terms of efficacy. More alarming still there is a serious lack of safety data.
Would chiropractors et al. who employ this practice please take note.
In total agreement on safety data! In Australian Informed Consent has been mandatory for all professions since 2000! Adverse event reporting should also be mandatory for ALL professions and the physio’s have flown under the radar on this issue and they know it! The critics have also ignore the physio’s on this issue!
What I do like is the systematic review was of physiotherapy research yet it seems to be solely about chiro’s in Edzards blog! With systematic reviews I like to read the references as well and go beyond the abstract to see where it is coming from!
What would constitute evidence sufficient to support or refute the treatment? Why was this not pre-defined?
The evidence shows treatment is no more effective than placebo, but fails to show that the treatment is as safe as placebo.
Such as it is, the evidence lends no support for the treatment, but does lend some concern for its safety.
So a chiro reviews the PHYSIOTHERAPY literature and the conclusions are:
“Don’t see a chiro!”
“Even if there had been adequate reporting of side-effects and even if this had not disclosed any problems, the safety of manipulation cannot be judged on the basis of such a small sample. Any responsible researcher should make it abundantly clear that the nasty habit by chiropractic pseudo-researchers of not reporting adverse effects is unethical and totally unacceptable.”
Now lets see at the bottom of the article it is posted in:
“Posted in alternative medicine, bias, chiropractic, clinical trial, conflict of interest, medical ethics, pseudo-science, risk”
Edzard, if you are considering what to write about for your next blog, consider ‘giving CAM a bash’ a rest; what about doing a review of vaccine studies?
I’d like Prof Ernst to expound on renin angiotensin system inhibitors for patients with stable coronary artery disease.
Edzard’s expertise is in CAM. Vaccines don’t qualify as CAM. In case you didn’t know.
never mind, Greg probably also goes to his bakery and asks for a roll of loo-paper.
alternatively, he merely wanted to indicate that he is also an ardent anti-vaxxer.