Are stabilisation exercises different to other treatments in improving physical activity or reducing disability for people with persistent low back pain? A systematic literature review

Background/Aims Stabilisation exercises are commonly prescribed for people with persistent low back pain. However, for some patients, it has been hypothesised that stabilisation exercises could draw attention to protecting the core, promote hypervigilance and inhibit volitional movement. The aim of this study was to compare the effectiveness and reported adverse events, in particular fear avoidance, of stabilisation exercises compared with placebo or other treatments offered by physiotherapists on the outcome of disability and activity at 12- and 24-months' follow-up. Methods The following electronic databases were searched: Embase, Medline, AMED, CINAHL, from inception to June 2019. Only randomised controlled trails were included. Study selection, data extraction and appraisal of quality criteria using PEDro, were undertaken by two independent assessors. Results Seven studies (n=1820) were eligible. Of six studies that reported adverse effects in the group receiving stabilisation exercises, four reported none and two reported mild exacerbation of pain locally or elsewhere. Fear avoidance was not investigated in any of the studies. Across the studies, 12 analyses were reported and included seven different comparator groups and three outcome measures: Oswestry Disability Index (n=1), Roland Morris Disability Questionnaire (n=5), Patient Specific Functional Scale (n=4). Two studies included a 24-month follow up in addition to a 12-month follow up. Of the 12 studies, nine reported no significant differences between the effectiveness of stabilisation exercises and comparator groups. Stabilisation exercises were more effective than comparator groups for the following three analyses: compared to manual therapy or education at 12 but not 24 months for the Oswestry Disability Index (15.71, 95% confidence interval 19.3-10.01); compared to placebo for the Patient Specific Functional Scale (1.5, 95% confidence interval 0.7-2.2) but not the Roland Morris Disability Questionnaire; and compared to high load lifting for the Patient Specific Functional Scale (1.8 95% confidence interval 2.8-0.7). Conclusions Stabilisation exercises are safe and equally effective to other treatments, and possibly superior for some outcomes at some time points. No or only mild adverse effects were reported. However, none of the studies measured fear avoidance as an outcome and we recommend this be included in future randomised controlled trials measuring the effectiveness of stabilisation exercises.