Randomised controlled trials (open-label or blind) of oral, sublingual or transdermal opioid analgesics or epidural steroid injections or steroid injections into the spinal canal, facet joint, sacroiliac joint or medial branch blocks were eligible for inclusion. Participants had to be aged over 18 with non-neoplastic chronic lower back pain or lumbar radiculopathy. Studies of patients with chronic pain after spinal surgery and studies that focused on partial or synthetic opioid analgesics were included. Primary outcomes were: pain relief measured with a visual analogue scale; functional benefit measured with the Roland-Morris questionnaire or Oswestry Disability Index; quality of life measured with Short-Form 36-item questionnaire; and mortality. Secondary outcome measures were other pain relief or functional benefit measures, psychological and quality of life measures and adverse events including aberrant opioid behaviours and procedural complications.
Interventions in the included randomised trials included tramadol, fixed-dose oxycodone, titrated dose oxycodone with sustained release morphine sulphate, methylprednisolone, prednisolone, bupivacaine, triamcinolone, betamethasone, procaine, dexamethasone, prilocaine and lidocaine. Comparators were placebo, physical therapy or another treatment. Patients presented with various conditions. Outcomes were measured at one week, between one week and one month, between one and three months, between three and six months and at more than six months.
One reviewer assessed titles and abstracts and a second reviewer checked 10% of these. It appeared that papers were selected by two reviewers independently and disagreements resolved by discussion or involvement of a third reviewer.