Thursday, June 7, 2012

Is Spinal Manipulation Ineffective for Pain?

CAMRecent systematic reviews assessed the effectiveness of spinal manipulation therapy, such as chiropractic, for managing various types of pain. Overall, this modality was found to be ineffective for relieving most types of pain and of debatable clinical usefulness for others, such as back pain. However, there also have been many weaknesses in past research that need further attention.

Among the many treatments often suggested for pain is spinal manipulation, as commonly used by chiropractors, and also by osteopaths, physiotherapists, or other manual therapists. Writing in an early online edition of the journal Pain Medicine, Paul Posadzki, PhD — from the Department of Complementary Medicine, Peninsula Medical School, Exeter, UK — defines spinal manipulation as “the application of high-velocity, low-amplitude manual thrusts to the spinal joints slightly beyond the passive range of joint motion” [Posadzki 2012].

The technique aims to correct misalignments or so called “subluxations” of the joints (both spinal and peripheral). However, as Posadzki notes, it has been suggested that such subluxations lack biological plausibility, and the safety and cost-effectiveness of spinal manipulation have also been questioned. Yet, spinal manipulation is widely used for a range of pain-related and other conditions.

To further examine the issue of effectiveness, Posadzki conducted a review of evidence from systematic reviews of spinal manipulation in patients with any type of pain. Four electronic databases were searched, from inception to March 2011, to identify all relevant articles on the effectiveness of spinal manipulation for pain. For selection, the systematic reviews had to include at least some evidence from randomized controlled trials and the spinal manipulation techniques studied could not merely be one component of more complex or extensive therapeutic interventions.

Posadzki discovered and assessed 22 qualifying systematic reviews relating to the following pain conditions: low back pain (N = 6), headache (N = 5), neck pain (N = 4), any medical problem (N = 1), carpal tunnel syndrome (N = 1), dysmenorrhea (N = 1), fibromyalgia (N = 1), lateral epicondylitis (N = 1), musculoskeletal conditions (N = 1), and non-spinal pain (N = 1). Nine of the reviews included more than 10 primary studies and 5 included a meta-analysis of data. Overall, most of the reviews (17) were of high quality, and 5 were of poor quality.

For those pain conditions noted above that had a single systematic review, spinal manipulation was not found beneficial for any of them. Similarly, for those conditions with multiple reviews, unanimously positive conclusions favoring benefits of spinal manipulation were not evident regarding any; although, there were some contradictions.

For example, for low back pain 3 reviews arrived at positive conclusions while the remaining 3 reached negative conclusions. Two of 5 reviews examining headache reached positive conclusions and the rest were negative. For neck pain, 3 of 4 reviews arrived at negative conclusions and the remaining one reached an equivocal conclusion.

In sum, Posadzki observes that it has been relatively unclear whether spinal manipulation is an effective therapeutic option for pain management. He believes that his review demonstrates rather convincingly that spinal manipulation is an ineffective option in the management of many types of pain, particularly neck pain, and further research in other areas such as low back pain or headache are need to reach more definitive assessments.

COMMENTARY: Spinal manipulation therapy, such as chiropractic, is very popular among patients and particularly for back pain. As noted in a prior UPDATE [here], nearly three-quarters of back pain sufferers (74%) who use CAM (Complementary and Alternative Medicine) modalities choose chiropractic care, followed by massage therapy at 22%, and at less than 10% each were acupuncture, yoga/tai chi/qi gong (considered together), herbal remedies, and relaxation techniques.

A recent randomized controlled clinical trial discussed in an UPDATE [here] found that chiropractic or home exercises were more effective than medications for relieving nonspecific subacute neck pain. However, the effect sizes were modest and there were a number of limitations in this study that cast doubt on the validity of outcomes.

Individual studies or surveys can be misleading; whereas, systematic reviews are considered to be at the top of evidence-base hierarchy and may therefore provide the most conclusive answers regarding the effectiveness of spinal manipulation for pain management. Although Posadzki found hundreds of individual clinical trials of spinal manipulation in the literature, the data were less than uniform and evaluations of effectiveness often arrived at contradictory conclusions.

In this current review by Posadzki, merely 5 of 22 systematic reviews meeting eligibility criteria suggested that spinal manipulation is effective for pain relief of any sort. The highest degree of ambiguity was noticed for low back pain; whereas, results were more definitive and negative regarding spinal manipulation for neck pain and headache. For other conditions, such as musculoskeletal pain, non-spinal pain syndromes, fibromyalgia, lateral epicondylitis, carpal tunnel syndrome, or primary and secondary dysmenorrhea, the conclusions were all negative.

In the relatively few positive reviews, publication bias may have played an important role. Posadzki observed that systematic review published by chiropractors or osteopaths were of a much lower quality and more positive than those of independent authors. For example…

Four (80%) of the 5 systematic reviews published by either chiropractors or osteopaths arrived at overtly positive conclusions and only one arrived at a negative conclusion. Conversely, 16 (94%) of the 17 reviews by independent research groups reached negative or equivocal conclusions; only one (6%) arrived at a clearly positive conclusion.

In another very recent review, Posadzki and coauthor Edzard Ernst assessed the effectiveness of spinal manipulation (ie, chiropractic) interventions specifically for the treatment and/or prevention of sports injuries [Ernst and Posadzki 2012]. In all, 6 trials qualified for inclusion: 4 RCTs and 2 controlled clinical trials (CCTs). Methodological quality of the studies was generally poor, and only half of the trials — 1 RCT, 2 CCTs — suggested that chiropractic was an effective treatment for sports injuries. Thus, based on the scarcity of good trials and the equivocal outcomes, the authors conclude that the therapeutic value of spinal manipulation for athletes is uncertain.

Regarding safety of spinal manipulation therapies, Posadzki [2012] observes, “…several hundred severe complications after upper spinal manipulation have been reported… although the estimates as to the incidence of these complications vary hugely.” Of particular concern are vascular accidents caused by arterial dissection after upper-body spinal manipulation and “spinal manipulation should be discouraged as a treatment option for pain in the cervical area,” he concludes.

We would add that, like many other CAM therapies, the effectiveness and safety of spinal manipulation may be highly dependent on the training and skills of the individual practitioner. Even the personality of the practitioner might play an important role in exerting placebo effects. Hence, there could be many confounding and unknown factors that have thus far influenced research outcomes. Posadzki suggests…

“Future research in this area should control for placebo effects by employing sham spinal manipulation, using blinded design, and being of adequate sample size based on power calculations. Allocation to groups should be concealed, data analyzed based on intention to treat, and validated outcome measures ought to be used to increase internal validity in future research on spinal manipulation in pain management.”

The current reviews by Posadzki, and Ernst and Posadzki, suggest that spinal manipulation is of debatable clinical usefulness in pain management. Higher quality clinical trials are needed for more rigorous systematic reviews, focusing on clearly defined populations, precise therapeutic interventions, and specific types of pain. Until such reviews are available, it seems difficult if not impossible to draw firm conclusions from the research for endorsing the therapeutic value of spinal manipulation for pain.

> Ernst E, Posadzki P. Chiropractic for the prevention and/or treatment of sports injuries: a systematic review of controlled clinical trials. Focus on Alternative and Complementary Therapies. 2012(Mar);17:9-14 [
available here].
> Posadzki P. Is Spinal Manipulation Effective for Pain? An Overview of Systematic Reviews. Pain Med. 2012(May 23); online ahead of print [
abstract here].

ADDENDUM: Soon after we posted the above, a pro/con debate regarding cervical spine manipulation appeared online in the British Medical Journal (BMJ, June 7, 2012). Benedict Ward and colleagues argue that the technique carries a risk of stroke resulting from neck arteries and the technique is “unnecessary and inadvisable” and “may carry the potential for serious neurovascular complications” [BMJ. 2012:344:e3679, abstract here]. Furthermore, a Cochrane review of RCTs found that cervical spinal manipulation, as a stand-alone treatment, provides only moderate short-term pain relief. Therefore, risks outweigh benefits.

On the other side of the issue, Professor David Cassidy, from the University of Toronto, and colleagues argue that high quality evidence "clearly suggests that manipulation benefits patients with neck pain" and raises doubts about any direct relation between manipulation and stroke [BMJ. 2012;344:e3680, abstract here]. However, they want to see more research into the pros and cons of this and other techniques with the aim of identifying preferred safe and effective treatments for neck pain.

eNotifications Don’t Miss Out. Stay Up-to-Date on Pain-Topics UPDATES!
Register [here] to receive a once-weekly e-Notification of new postings.