Comparative efficacy trials with no placebo group cannot determine efficacyBMJ 2015; 350 doi: https://doi.org/10.1136/bmj.h3292 (Published 24 June 2015) Cite this as: BMJ 2015;350:h3292
- Chung-Wei Christine Lin, senior research fellow and associate professor1,
- Richard O Day, professor of clinical pharmacology2,
- Ian Harris, professor of orthopaedic surgery 3,
- Chris G Maher, professor 1,
- Andrew McLachlan, professor of pharmacy4
- 1George Institute for Global Health and Sydney Medical School, University of Sydney, Sydney NSW 2000, Australia
- 2St Vincent’s Hospital Clinical School and Pharmacology, School of Medical Sciences, Faculty of Medicine, University of New South Wales, NSW, Australia
- 3South Western Sydney Clinical School, Faculty of Medicine, University of New South Wales
- 4Faculty of Pharmacy and Centre for Education and Research on Ageing, University of Sydney
Cohen and colleagues’ trial of epidural steroids versus gabapentin in patients with lumbosacral radicular pain found no significant difference in the primary outcome.1 There are several problems with their conclusion that a trial with neuropathic drugs might be a first line treatment option.
Given the lack of a difference between treatment groups, there was no justification for endorsing one treatment over the other. In addition, the trial did not study first line treatment. All participants had experienced symptoms for at least six weeks; about 80% had chronic symptoms and 25% were taking opioids at baseline. More importantly, there was no convincing evidence that either treatment was efficacious against placebo.
In a meta-analysis we showed that epidural steroids have a significant short term effect on pain and disability compared with placebo.2 However, the small effect size indicates that the treatment may not be clinically worth while. Epidural steroid injections are weakly recommended for lumbosacral radicular pain by guidelines.3 Although the latest guideline recommends gabapentin as first line drug for neuropathic pain,4 evidence comes from trials in other neuropathic pain conditions. Our systematic review found only limited direct evidence5—one small unregistered trial showed a moderate short term benefit for gabapentin versus placebo in people with lumbosacral radicular pain.6
In view of the limited evidence supporting the efficacy of epidural steroid injections or gabapentin and the lack of a placebo comparison, Cohen and colleagues’ results do not suggest that these two treatments are equally effective—they could be equally ineffective or equally harmful (42.8% of participants withdrew from the study owing to negative outcomes after one month).1 This comparative efficacy trial does not help clinical decision making because neither treatment has evidence of efficacy against placebo.
Instead of comparing the efficacy of these treatments, the more urgent research question is whether either treatment, or another, is superior to placebo.
Cite this as: BMJ 2015;350:h3292
Competing interests: We have conducted a systematic review on the effects of epidural steroid injections (Pinto et al. Ann Intern Med 2012;157:865-77) and drug therapy (Pinto et al. BMJ 2012;344:e497) for sciatica, and are currently conducting an investigator initiated clinical trial evaluating the effect of pregabalin versus placebo in sciatica (Mathieson et al. Trials 2013;14:213).