Study design: A prospective cross-sectional analytic study.
Objectives: To assess in patients with chronic low back pain whether the presence or absence of pain originating from the lumbar zygapophysial joints correlates with changes seen on computed tomography.
Summary of background data: Results of studies have been divided as to whether or not radiologic imaging is able to predict those patients with pain originating from the zygapophysial joints.
Methods: Sixty-three patients with low back pain lasting for longer than 3 months underwent computed tomography and blocks of the zygapophysial joints at L5-S1, L4-L5, and L3-L4. The zygapophysial joints of all images were scored by three independent, masked radiologists.
Results: Interobserver agreement was poor with intraclass correlation coefficients of 0.34-0.66 using total joint scores for all three assessors. Using the results of a repeat assessment with two radiologists there was no statistically significant difference in joint scores between those with and those without pain originating from the zygapophysial joint.
Conclusions: Computed tomography has no place in the diagnosis of lumbar zygapophysial joint pain.