Study designs of evaluations included in the review
Randomised controlled trials (RCTs) with at least 20 patients in the smallest group were eligible. Diagnostic studies, descriptive and review studies, and case series or case reports, were excluded.
Specific interventions included in the review
Studies dealing with invasive procedures were excluded. No other inclusion criteria were specified. The actual placebo-controlled interventions included the following.
Non-steroidal anti-inflammatory drugs (NSAIDs), i.e. piroxicam (40 mg/day), indomethacin (dosage unclear) or phenylbutazone (600 to 1,200 mg/day), compared with placebo.
Traction, consisting of intermittent Tru-trac, Spina-Trac, greater than 45 kg traction daily, or autotraction and corset, compared with control therapy (fictitious traction, traction stimulation, infrared heat or corset alone).
Epidural steroids, i.e. methylprednisolone (2 to 10 mL of an 80 mg preparation), compared with control (epidural saline, intraspinal ligament saline, and sacral hiatus lignocaine).
Intramuscular steroids at two different dosing regimes, compared with neutral ampulla or saline injection.
Other therapies, comprising oral chymoral, manipulation or tizanidine, compared with placebo tablets or infrared heat.
The non-placebo-controlled interventions included: manual traction compared with isometric exercises; autotraction compared with manual traction; discectomy compared with physical therapy; comparison of various types of therapy (Maitland, manual traction, exercises, and corset); and tiaprofenic acid compared with ketoprofen.
The duration of therapy, where stated, ranged from one episode of therapy to 12 days.
Participants included in the review
Studies of patients with sciatica and/or nerve root compression as a well-demarcated main group or as a subgroup of the study were eligible. Studies that did not deal with lumbar disc herniation were excluded. The included studies were set in primary and secondary care. The patients included those with acute disease (less than 3 months' duration) and those with nonacute disease.
Outcomes assessed in the review
The inclusion criteria were not defined in terms of the outcomes. Improvement was assessed and patients were regarded as successfully treated if they reported overall improvement, improvement of pain, or return to work. The included studies assessed overall improvement using global measures of improvement, and evaluated pain using visual analogue scales.
How were decisions on the relevance of primary studies made?
The authors do not state how the papers were selected for the review, or how many of the reviewers performed the selection.