Efficacy and safety of antidepressants for the treatment of low back pain and osteoarthritis: protocol for a systematic review and meta-analysis

Objectives: To systematically review the efficacy and safety of antidepressants for patients with spinal pain (including sciatica) and osteoarthritis of the hip and knee. Eligibility criteria: We will include randomised placebo-controlled trials published as full-text in peer-reviewed journals enrolling Participants with a diagnosis of non-specific spinal pain (low back and/or neck pain), spinal pain with radicular symptoms and/or radiculopathy; and hip and/or knee OA. Studies with mixed populations will be included. Studies that include participants with serious spinal pathologies (e.g. cauda equina syndrome, cervical myelopathy, spinal tumours, spinal infection) will be excluded. We will exclude studies including participants with inflammatory arthritis eg axial spondyloarthritis, rheumatoid arthritis unless data for knee and/or hip OA is reported separately. Studies where participants received previous spinal or OA surgery are eligible, but studies evaluating immediate post-operative pain management (ie within past month) will not be included. We will include trials testing any type of antidepressant drug prescribed at any dose, as treatment for spinal pain and/or hip and/or knee OA. Studies testing drug combinations (e.g. antidepressants in addition to other analgesics) will only be included if the treatment contrast is other analgesic in addition to placebo Types of antidepressants drugs to be included. Comparators will include both inert (e.g. an inert substance that does not contain an active drug treatment) and active placebos (i.e. drugs that have no known effect on pain but may mimic the adverse effects of antidepressants). Outcomes: Primary outcomes will be Pain intensity and disability. Secondary outcomes will include adverse events (serious and any adverse event), as well as the number and proportion of participants who stopped the study medicicine due to adverse events and who dropped out because of adverse events. Data synthesis: All outcomes will be converted to a 0-100 scale. We will group studies by condition and antidepressant class. If studies are considered to be sufficiently homogenous, results will be pooled. The I2 statistics will be used to analyse the between-trial heterogeneity, and a random effects model will be used among trials when I2 > 0%.