Conflicts of interest
When ‘strong’ might be wrong: Evaluating the evidence in CRPS management
Version of Record online: 9 JAN 2013
© 2013 European Federation of International Association for the Study of Pain Chapters
European Journal of Pain
Volume 17, Issue 2, pages 143–144, February 2013
How to Cite
O'Connell, N.E. and Wand, B.M. (2013), When ‘strong’ might be wrong: Evaluating the evidence in CRPS management. European Journal of Pain, 17: 143–144. doi: 10.1002/j.1532-2149.2012.00238.x
- Issue online: 9 JAN 2013
- Version of Record online: 9 JAN 2013
- Manuscript Accepted: 27 AUG 2012
In this issue, you will find a paper by Cossins et al., (2012) entitled ‘Treatment of complex regional pain syndrome [CRPS] in adults: A systematic review of randomised controlled trials published from June 2000 to February 2012’. The aim of this review was to provide an updated summary of the evidence of effectiveness for all treatments for CRPS. This is an important task as considerable uncertainty persists regarding the optimal management of this condition. It is also a difficult one. The review reveals a broad and somewhat disparate selection of approaches mostly represented by one or two small trials of variable quality. Heterogeneity in interventions, dose, delivery method, diagnostic criteria and stage of the disease are frequently compounded by incomplete reporting in the original trials. We sympathize with the challenges this presented to the authors as we are currently completing an overview of systematic reviews for the Cochrane Collaboration on the same topic (O'Connell et al., 2011).
Cossins et al., (2012) conclude that there is strong evidence that a number of treatments are effective. While some encouraging results do emerge from the evidence base, to label the supporting evidence as strong is problematic and maybe premature. The level of evidence statement is partly determined by trial quality and the authors have used a relatively low threshold for considering a trial to be of high quality (≥50/100 on the chosen quality assessment tool). Given the varied and subtle ways in which trials can be confounded by bias, it is likely that a trial achieving a score close to this threshold will suffer from multiple weaknesses that individually or combined could explain any observed effects. In addition, judging a trial as positive based solely on statistical significance brings further problems. Lenient analyses, which are not uncommon in small exploratory trials, increase the risk of spurious results. Further, where statistically significant results are robust, the true treatment effect may still be clinically trivial. In these instances a judgement of strong evidence may afford undeserved credibility to a treatment of limited clinical worth.
For example, the authors conclude that there is strong evidence that repetitive transcranial magnetic stimulation (rTMS) is effective. This judgement is based on two small trials with a combined n of 33 (Pleger et al., 2004; Picarelli et al., 2010), and quality scores of 63 and 50/100, respectively. Both studies carry potential threats to the integrity of participant and assessor blinding and in one (Picarelli et al., 2010), the processes of randomization and allocation concealment are not clearly described. Notwithstanding those concerns, these trials do not provide sufficient data to give us a precise estimate of the size of the treatment effect. Indeed, generally, in chronic pain, there is reason to doubt that rTMS delivers clinically meaningful reductions in pain (O'Connell et al., 2010).
In their discussion, Cossins et al., (2012) acknowledge some of these limitations. This is to be welcomed, but we would suggest that their importance requires that they be more clearly reflected in the overall conclusions, which may be all that some will read. In our view, while some treatments hold promise, there is currently no intervention for CRPS that can be considered to be supported by strong evidence of efficacy; although we agree that the evidence is strong that intravenous regional blockade with guanethidine is ineffective.
This review illustrates the wider challenges of reviewing the evidence for pain treatments. Reviews need to establish clear criteria for clinical importance, to avoid systems that might produce overly positive assessments and to clearly assess and communicate the potential for bias within the evidence. However, much ambiguity could be resolved if original trials were better conducted and reported in the first place. It is unacceptable that many contemporary published trials do not fully comply with the CONSORT statement (Schulz et al., 2010). Ensuring this and following the recent recommendations of the IMMPACT consensus group (Dworkin et al., 2008, 2009, 2010, 2012; Turk et al., 2008) in the design and reporting of future trials in CRPS would go a long way to creating an evidence base with genuine authority and credibility to guide clinical practice.
What this review makes very clear is that better-designed trials are needed. Sadly, the search for unpublished studies suggests that we are not on the cusp of a flood of high-quality data. Cossins et al., (2012) recommend a collaborative, multi-centre approach to addressing this issue, a sentiment that we share wholeheartedly. While we might not learn much from the existing data regarding the best treatment for CRPS, there are clear lessons for how to cultivate the evidence base. Furthermore, this review demonstrates that bisphosphonates, ketamine and graded motor imagery programmes all hold promise and are interventions worthy of further, rigorous investigation.
- 2012). Treatment of complex regional pain syndrome in adults: A systematic review of randomized controlled trials published from June 2000 to February 2012. Eur J Pain 17, 158–173. , , , , , (
- 2009). Interpreting the clinical importance of group differences in chronic pain clinical trials: IMMPACT recommendations. Pain 146, 238–244. , , , , , , , , , , , (
- 2010). Research design considerations for confirmatory chronic pain clinical trials: IMMPACT recommendations. Pain 149, 177–193. , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , (
- 2012). Considerations for improving assay sensitivity in chronic pain clinical trials: IMMPACT recommendations. Pain 153, 1148–1158. , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , (
- 2008). Interpreting the clinical importance of treatment outcomes in chronic pain clinical trials: IMMPACT recommendations. J Pain 9, 105–121. , , , , , , , , , , (
- 2010). Non-invasive brain stimulation for chronic pain in adults. Cochrane Database Syst Rev (9), CD008208. , , , , (
- 2011). Interventions for treating pain and disability in adults with complex regional pain syndrome. Overview (PROTOCOL). Cochrane Database Syst Rev (11), CD009416. , , , , (
- 2010). Repetitive transcranial magnetic stimulation is efficacious as an add-on to pharmacological therapy in complex regional pain syndrome (CRPS) type I. J Pain 11, 1203–1210. , , , , , , , (
- 2004). Repetitive transcranial magnetic stimulation of the motor cortex attenuates pain perception in complex regional pain syndrome type I. Neurosci Lett 356, 87–90. , , , , , (
- the CONSORT Group. (2010). CONSORT 2010 Statement: Updated guidelines for reporting parallel group randomised trials. BMJ 340, c332. , , , for
- 2008). Analyzing multiple endpoints in clinical trials of pain treatments: IMMPACT recommendations. Pain 139, 485–493. , , , , , , , , , , , , , , , , , , , , , , , , , , , , , (