Spinal manipulative therapy (SMT) is frequently used to manage cervicogenic headache (CGHA). No meta-analysis has investigated the effectiveness of SMT exclusively for CGHA.
The aim of this review was to evaluate the effectiveness of SMT for cervicogenic headache (CGHA). Seven RCTs were eligible. At short-term follow-up, there was a significant, small effect favouring SMT for pain intensity and small effects for pain frequency. There was no effect for pain duration. There was a significant, small effect favouring SMT for disability. At intermediate follow-up, there was no significant effects for pain intensity and a significant, small effect favouring SMT for pain frequency. At long-term follow-up, there was no significant effects for pain intensity and for pain frequency.
The authors concluded that for CGHA, SMT provides small, superior short-term benefits for pain intensity, frequency and disability but not pain duration, however, high-quality evidence in this field is lacking. The long-term impact is not significant.
This meta-analysis can be criticised for a long list of reasons, the most serious of which, in my view, is that it is bar of even the tiniest critical input. The authors state that there has been no previous meta-analysis on this topic. This might be true, but there has been a systematic review of it (published in the leading journal on the subject) which the authors fail to mention/cite (I wonder why!). It is from 2011 and happens to be one of mine. Here is its abstract:
The objective of this systematic review was to assess the effectiveness of spinal manipulations as a treatment option for cervicogenic headaches. Seven databases were searched from their inception to February 2011. All randomized trials which investigated spinal manipulations performed by any type of healthcare professional for treating cervicogenic headaches in human subjects were considered. The selection of studies, data extraction, and validation were performed independently by 2 reviewers. Nine randomized clinical trials (RCTs) met the inclusion criteria. Their methodological quality was mostly poor. Six RCTs suggested that spinal manipulation is more effective than physical therapy, gentle massage, drug therapy, or no intervention. Three RCTs showed no differences in pain, duration, and frequency of headaches compared to placebo, manipulation, physical therapy, massage, or wait list controls. Adequate control for placebo effect was achieved in 1 RCT only, and this trial showed no benefit of spinal manipulations beyond a placebo effect. The majority of RCTs failed to provide details of adverse effects. There are few rigorous RCTs testing the effectiveness of spinal manipulations for treating cervicogenic headaches. The results are mixed and the only trial accounting for placebo effects fails to be positive. Therefore, the therapeutic value of this approach remains uncertain.
The key points here are:
- methodological quality of the primary studies was mostly poor;
- adequate control for placebo effect was achieved in 1 RCT only;
- this trial showed no benefit of SMT beyond a placebo effect;
- the majority of RCTs failed to provide details of adverse effects;
- this means they violate research ethics and should be discarded as not trustworthy;
- the therapeutic value of SMT remains uncertain.
The new paper was published by chiropractors. Its positive result is not clinically relevant, almost certainly due to residual bias and confounding in the primary studies, and thus most likely false-positive. The conclusions seem to disclose more the bias of the review authors than the truth. Considering the risks of SMT of the upper spine (a subject not even mentioned by the authors), I cannot see that the risk/benefit balance of this treatment is positive. It follows, I think, that other, less risky and more effective treatments are to be preferred for CGHA.
5 Responses to Spinal manipulation for the management of cervicogenic headache: a new review of the evidence
Leave a Reply
This site uses Akismet to reduce spam. Learn how your comment data is processed.
Thank you for this review and your analysis. One needs proper ammunition when dealing with frauds as they can be very clever. One thing regarding ‘spinal manipulation’ I simply can never “get past”: millions if not billions of us humans can and do manipulate our own spines, especially our necks…getting whatever the momentary “relief” such an act might give (Like forcing a burp?). But recognizing it has ZERO long term value and likely has detriment on several levels…just like when done by a $150,000 “trained quack”. When I do it to myself I might argue It is a much more informed application…and I don’t need insurance coverage. I would also argue that biomechaniclly the joint that eventually ‘cracks’ must be already mobile, and thus likely not in need of such mobilization. I would further argue that IF 2 of the 7 bones in the neck were indeed “stuck, fixated or subluxed” I can think of no plausible reason a force-applied would get “them“ to move and not simply move the already mobile ones (specificity has been shown to be an illusion). Thus it may be the reason manipulation never generates long-term results past whatever is gained by firing some mechanoreceptors and/or placebo. It’s NEVER a real answer.
Most laypeople don’t understand it.
Ernst likes to make it sound so simple. Perhaps he just doesn’t understand.
“Current RCTs suggest that physiotherapy and SMT might be an effective treatment in the management of CEH. However, the RCTs mostly included participant with infrequent CEH. Future challenges regarding CEH are substantial both from a diagnostic and management point of view.” https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3381059/
“We conclude that current literature suggests that not all non-pharmacological interventions are effective for all headaches, and that multimodal, not isolated, approaches seem to be more effective for patients with headaches.” https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7312657/
“Physical therapy is considered the first line of treatment. Manipulative therapy and therapeutic exercise regimen are effective in treating a cervicogenic headache.” https://www.ncbi.nlm.nih.gov/books/NBK507862/#article-19263.s7
“Our systematic review suggests that most randomized controlled trials published to date have investigated headaches with a clinical presentation involving the neck that maybe better defined as “possible,” “probable,” or “definitive” CeH depending on how well the diagnostic criteria used align with the most recent edition (3rd) of the International Classification of Headache Disorders.” https://headachejournal.onlinelibrary.wiley.com/doi/full/10.1111/head.13719
etc, etc, etc
Not trustworthy? Interesting…
“Previous studies have examined the methods for AE collection and presentation only, and highlighted the inadequacies in AE reporting in journal articles.7–16” Phillips R, Hazell L, Sauzet O, et al. Analysis and reporting of adverse events
in randomised controlled trials: a review. BMJ Open 2019;9:e024537. doi:10.1136/ bmjopen-2018-024537
Here is the full discussion at the end of the paper (not abstract) so you can read it for yourselves free from Edzards spin:
“This systematic review and meta‐analysis demonstrated that SMT provides significant, small short‐term effects for pain intensity, frequency and disability but not pain duration among people suffering from CGHA. Effect sizes in our meta‐analysis generally reduced over the intermediate and long‐term follow‐up periods. The lack of high‐quality evidence utilizing the GRADE approach creates some uncertainly in these results.
Our findings are generally in agreement with previous contemporary systematic reviews regarding the efficacy of manual therapies for CGHA (Chaibi & Russell, 2012; Coelho et al., 2019; Garcia et al., 2016; Racicki et al., 2013). Our review adds to the existing literature by including a greater number of eligible trials in the pooled analysis. In contrast to other reviews (Chaibi & Russell, 2012; Coelho et al., 2019; Garcia et al., 2016; Racicki et al., 2013), our review isolated SMT as a single modality and found statistically significant differences favouring SMT over other manual therapies in the short‐term. Few other reviews on the topic have reported the effect sizes for key outcomes (Racicki et al., 2013). Another interesting finding from our results was that SMT confers small benefits in the intermediate term for CGHA pain frequency, with the magnitude of effect remaining both small and statistically significant throughout this follow‐up period (SMD − 0.32 [−0.63 to − 0.00] p = .05). Headache frequency is considered the most important primary measure in efficacy studies for headache, according to the IHS guidelines for controlled trials (Bendtsen, Bigal, & Diener, 2009).
Although our review identified statistically significant differences between SMT and other manual therapies in the short‐term, it is important to consider our findings in light of clinically meaningful effects. Minimal clinically important change (MCID) or the smallest worthwhile effect (Kamper, 2019a), refers to the smallest difference that patients perceive as beneficial with respect to the treatment they receive. By reporting the estimate of the size of effects in our review (as opposed to just the presence of effects), we can further evaluate whether these treatment effects are likely to be large enough to be considered important for patients.
Currently, there is little evidence on what defines the smallest worthwhile effect of interventions for CGHA. A recent analysis of psychometric properties for CGHA patient outcomes, determined a MCID threshold for self‐reported pain intensity of 2.5 points on the 0–10 Numeric Pain Rating Scale and 5.5 points on the 0–50 Neck Disability Index Scale at the short‐term follow‐up (Young, Dunning, Butts, Cleland, & Fernández‐de‐las‐Peñas, 2019). Although these new CGHA estimates (which utilized the CHISG diagnostic criteria) help determine a threshold for clinically meaningful difference, it remains unknown whether some patients may consider the small, superior short‐term effects in our review as clinically worthwhile. For instance a treatment effect which favoured SMT for pain intensity in the short‐term (MD − 10.88 [95% CI, −17.94, −3.82] p = .003), may be considered too small to be worthwhile. Yet the treatment effect of disability in the short‐term (MD − 13.31 [95% CI, −18.07, −8.56] p ≤ .00001) may justify the effort to seek SMT for CGHA.
Given that there is no universal agreement on the size of a worthwhile effect (Kamper, 2019b), clinicians in practice can discuss the estimated effect with patients, who can then set their own individual ‘cut off’ points as the smallest worthwhile effect. In doing so, the patient must consider and balance several factors related to undertaking an intervention, including perceived benefits and potential risks associated with treatment, personal expectations, duration of treatment and related costs—all which are incorporated into their treatment decision‐making process (Herbert, 2019; Kamper, 2019a). Further to these considerations is the paucity of research evidence for the relief of CGHA from drug treatments as well as other far more invasive available treatments (Anthony, 2000; Zhang, Shi, & Wang, 2011; Zhou, Hud‐Shakoor, Hennessey, & Ashkenazi, 2010).
A recently published evidence‐based guideline for the non‐pharmacological management of headaches associated with neck pain, recommends the use of SMT, spinal mobilization or cranio‐cervical‐scapula exercises for CGHA management (Côté et al., 2019). Importantly, this guideline did not recommend combining these therapeutic modalities (Côté et al., 2019). In line with this guideline recommendation, our study findings suggest SMT, in isolation, provides comparatively superior, small short‐term benefits to CGHA pain intensity, frequency and disability over other manual therapies. Communicating such findings may be valuable in the shared decision‐making process between practitioners and patients, when considering treatment options for CGHA. In doing so, clinicians need to discuss any potential treatment benefits, that is the positive impacts of SMT on the quality‐of‐life of those with CGHA (Suijlekom et al., 2003), as well as the potential harms, that is mild‐to‐moderate adverse events generally associated with SMT (Carnes, Mars, Mullinger, Froud, & Underwood, 2010). Accordingly, this process helps ensure clinicians support patients in choosing treatment options that best aligns with patient preferences and values (Hoffmann, Lewis, & Maher, 2019).
This systematic review and meta‐analysis provides an up‐to‐date summary of the impact of SMT among CGHA sufferers. Other strengths include a comprehensive search of numerous electronic databases, and most of the trials included were of high methodological quality (mean PEDro Score of 7). Furthermore, we utilized the GRADE system to appraise the overall quality of evidence. Our recommendations were based on the pooled treatment effects and provided clinically interpretable estimates (where possible), thus assisting clinicians and patients as to whether they would consider SMT for CGHAs as meaningful or important.
A limitation of the current literature was the limited number of trials (n = 7) with a relatively small sample size, ranging from 4 to 63 patients. Therefore, the overall quality of the evidence for all meta‐analyses were downgraded for reasons of imprecision as less than 400 patients were included in the analysis (Guyatt et al., 2011). Within our pooled studies, there was between‐trial heterogeneity (I 2 = 70%) for the short‐term pain intensity outcome. In general, the considerable heterogeneity in our study is likely due to the use of different outcome measures and methodologies. One example is the different diagnostic inclusion criteria utilized for CGHA diagnosis as assessed by the practitioners (Avijgan & Thomas, 2019). While published diagnostic criteria from the CHISG and IHS are similar, they also vary and can therefore result in patient presentation variability with respect to study eligibility criteria. For example IHS criteria includes headache with a temporal relationship (increasing and decreasing) with neck pain with no clear guidance for how this is determined (Fredriksen et al., 2015). Unlike the IHS, CHISG includes pain starting posteriorly moving anteriorly; unilateral head pain, without side shift and diffuse shoulder and/or arm pain as criteria (Fredriksen et al., 2015). More generally, neither IHS nor CHISG provide clear guidance on the magnitude or direction of the neck provocation needed to stimulate headache or the extent of the loss of cervical motion needed for a diagnosis (Fredriksen et al., 2015). Furthermore, all studies within our review did not fully adopt the diagnostic criteria, that is certain IHS criterion was not relevant to the clinical trial (Haas et al., 2018), or further subgrouping beyond the IHS was required (Borusiak et al., 2010). Some studies excluded the radiographic criterion from the IHS (Haas et al., 2010; Nilsson, 1995; Nilsson et al., 1997) or excluded nerve blockage from the CHISG (Chaibi et al., 2017; Dunning et al., 2016). In our review, the IHS criteria was utilized in five studies (Borusiak et al., 2010; Haas et al., 2010, 2018; Nilsson, 1995; Nilsson et al., 1997), whereas the CHISG was utilized in two studies (Chaibi et al., 2017; Dunning et al., 2016). A sensitivity analysis according the different CGHA diagnostic criteria utilized continued to show significant, small effects favouring SMT over other manual therapies, with respect to pain intensity in the short‐term for the IHS (5 studies, I 2 = 31%, MD − 7.32 [95% CI, −12.62, −2.02]) and the CHISG (2 studies, I 2 = 0%, MD − 19.52 [95% CI, −26.22, −12.82]), respectively.
It is also likely that some patients with the diagnosis of CGHA, may have co‐occurring migraine and/or tension‐type headache (Knackstedt et al., 2010). This is problematic for clinicians (and researchers) in determining the exact diagnosis and therefore the most appropriate treatment strategy to manage CGHA specifically (Avijgan & Thomas, 2019). Ultimately, the diagnosis of CGHA may only be demonstrable in a definitive manner, by way of anaesthetic blockade procedures (Avijgan & Thomas, 2019), which was not available in our included trials. Another limitation includes the different forms of SMT techniques that were incorporated within the included studies and the treatment frequency varied between trials and/or was not always clearly defined. In addition, other interventions were sometimes also used in conjunction with the other manual therapy study groups. Only English language studies published in peer‐reviewed journals were included and it was not possible to carry out all planned analyses due to insufficient data available within studies.
While this review provides clinically interpretable estimates, future trials that evaluate the intermediate‐ to long‐term impact of SMT for CGHAs are needed. These forthcoming trials should be adequately powered and utilize the current CGHA diagnostic criteria. Future trials should also reflect recommendations from the most recent guidelines with respect to SMT for CGHA (Côté et al., 2019). Additionally, trials should include a cost‐effectiveness analysis and patient subgrouping to better determine the profile of those with CGHA who are more likely to benefit from SMT. Although differential diagnosis can be challenging, there is a need for trial designs to effectively isolate CGHA participants from those with co‐occurring migraine and/or tension‐type headache.”