Reference: Cochrane Database of Systematic Reviews 2017; 1 (CD010671)
Chronic low back pain (lasting more than 3 months) affects an estimated 19.6% of adults aged 20-59 (Rev Saude Publica. 2015;49) and is responsible for 60.1 million person-years lived with disability. It is also the most common cause of medically approved sick leave and early retirement in Europe. Chronic low back pain leads to more lost work days than any other occupational musculoskeletal condition in the US (Lancet. 2018 Jun 9;391(10137):2356). Low back pain has been described as “the leading worldwide cause of years lost to disability” (Lancet. 2018 Jun 9;391(10137):2384). There are few long-term safe and effective treatment options for chronic low back pain based on high-quality evidence. Several studies have investigated exercise as a treatment for chronic low back pain. Pilates (Clin Rehabil. 2018 Sep;32(9):1249), exercise in general, and tai chi have all shown some evidence for improvement of chronic low back pain (Ann Intern Med. 2017 Apr 4;166(7):493) though truly effective treatments are still lacking. This recent Cochrane review addresses the question: is yoga better than non-exercise interventions and other, non-yoga exercise interventions at improving function and pain in patients with chronic low back pain?
Twelve randomized trials with 1,080 adults evaluating yoga for chronic low back pain were reviewed. The forms of yoga varied and were compared to no or delayed intervention or patient education (7 trials), a back-focused form of exercise intervention (3 trials), or both back-focused exercise and non-exercise interventions (2 trials). Eleven trials included yoga sessions from 1 to 3 times per week lasting 45 to 90 minutes; 1 trial included an intensive residential yoga program. Primary outcomes were back-specific functional status as measured by the Roland-Morris Disability Questionnaire (RMDQ; scale 0-24 points, with lower score indicating less impairment, minimal clinically important difference was not pre-determined) and pain as measured by the Visual Analogue Scales (scores transformed to range 0-100 points, clinically significant change in score defined as ≥ 15 points). Most participants were assessed at 4-6 weeks, 3-4 months, and 6 months, though there was some variability between studies. All the included studies were at high-risk of both performance and detection bias, as all participants and investigators knew the intervention groups and outcomes were self-reported.
Compared to non-exercise, or no intervention, yoga was associated with improvements in back-related function at 3-4 months (mean difference [MD] -2.18 points on RMDQ, 95% CI -3.6 to -0.76) in analysis of 7 studies with 667 participants, and at 6 months (MD -2.15 points on RMDQ, 95% CI -3.23 to -1.08) in analysis of 6 studies with 630 participants. At 12 months these improvements attenuated to a MD of -1.36 points (95% CI -2.41 to -0.26) in analysis of 2 studies with 365 participants. Yoga improved pain scores at 3-4 months compared to non-exercise interventions (MD -4.55 points on 100-point scale, 95% CI -7.04 to -2.06) in analysis of 5 studies with 458 participants, but the pre-specified change for clinical significance (≥ 15 points) was not met. Similar results were found at 6-months (MD -7.81, 95% CI -13.37 to -2.25) in analysis of 4 studies with 414 participants and at 12-months (MD -5.4, 95% CI -14.5 to -3.7) in analysis of 2 studies with 355 participants. Yoga was associated with increased risk of worsening back pain compared to non-exercise (risk difference 5%, 95% CI 2%-8%) in analysis of 6 trials with 696 participants.
There was no significant difference between yoga and back-focused exercise in improving back function in analysis of 2 trials with 249 adults. There were no significant differences between yoga and back-focused exercise in adverse events in analysis of 3 trials with 314 adults. A small study suggested yoga may improve pain compared to back-focused exercise at 7 months, but the study may be at high-risk of bias due to lack of blinding as well as questionable allocation concealment, data reporting, compliance and baseline group similarity.
This review suggests that exercise-based interventions for back pain (including yoga) might be similar to non-exercise interventions. For patients with chronic low back pain who enjoy yoga or have easy access to it, yoga could be suggested as possibly beneficial with a low risk of harm.
For more information, see the Exercise therapy for chronic low back pain topic in Dynamed Plus. DynaMed users, see the Exercise therapy for chronic low back pain topic in Dynamed Classic.
Kelly Henkler, MD attended Albany Medical College and is currently a PGY-3 at University of Virginia. Interests include sports medicine, student health and public health.
Faculty contributions by Katharine C. DeGeorge, MD, MS.