A proportion of patients who are classified by MRI as having lumbar disc herniation or spinal stenosis could be receiving incorrect diagnoses, according to researchers in The Netherlands.
The authors noted that although MRI is used for diagnostic confirmation of serious underlying pathology in cases of low back pain, its role remains controversial. Thus, they sought to summarize the available evidence on MRI’s diagnostic accuracy in identifying pathology of the lumbar spine in adult sciatica or low back pain patients.
The authors searched three databases — MEDLINE, EMBASE and CINAHL — for studies comparing MRI to a reference test for lumbar spine pathology identification. The results of relevant studies that met the inclusion criteria were used to create pooled summary estimates of sensitivity and specificity for homogenous subsets of studies, the authors wrote.
In all, the authors used eight studies for the review, defining strata for separate pathologies. Five of the studies were used to compare MRI to surgical findings for the identification of lumbar disc herniation in a meta-analysis, with a summary estimate of sensitivity of 75% — and specificity of 77% — being demonstrated through a pooled analysis.
The authors noted they were not able to form such an analysis for spinal stenosis, and added, “the evidence for the diagnostic accuracy of MRI found by this review is not conclusive, since the results could be distorted due to the limited number of studies and large heterogeneity.”