The role of MRI or CT in the management of patients with LBP, for whom there is no clear clinical indication for the use of sophisticated imaging, is uncertain. The aim of The Scottish Back Trial was to determine whether early use of MRI or CT influences clinical management and outcome of patients with LBP and whether it is cost-effective.

Elective patients were new referrals to orthopaedic or neurosurgeons with symptomatic lumbar spine disorders (without ‘red flags’). After obtaining informed consent, patients were randomised to ‘early imaging’ (imaging as soon as practicable) or ‘delayed, selective imaging’ (imaging only if an imperative clinical indication developed). Principal outcomes measures were the SF-36, questionnaire, the Aberdeen LBP Scale and the EQ-5D. Patients completed questionnaires at trial entry and after 8 and 24 months.

From 15 hospitals, 2657 patients were assessed and 783 were recruited and randomised. Eight months and twenty-four months after trial entry, comparison of data abstracted from case notes indicated that, apart from the use of imaging, both groups had received similar clinical management. At follow-up, an improvement in health status, as measured by the SF-36 and Aberdeen LBP Scale, was reported by both groups. At 24 months, there was a statistically significant but small difference in favour of the ‘early imaging’ group (p=0. 002) as measured by the Aberdeen LBP Scale but no difference in the SF-36 except a marginal improvement in the bodily pain subscale.

The use of MRI or CT imaging for this group of LBP patients did not significantly affect their management. The clinical significance of the marginal improvement in health status in the ‘early imaging’ group is uncertain. The results of the cost-effectiveness analysis may clarify whether a policy of ‘early imaging’ would be a cost effective use of resources.