Abstract
Purpose and Background: Cerebrospinal fluid (CSF) is used as an intrabody reference for quantitative disc signal intensity measurements in the lumbar spine, but may be less ideal for investigating thoracic disc degeneration. The purpose of this study was to investigate the spinal cord as an alternative intrabody reference for thoracic disc signal intensity measurements.
Methods: T2-weighted MR images from the thoracic spine (T6-T12) were obtained using two 1.5-Tesla scanners from a population-based sample of 523 men (35–70 years). Quantitative data on signal intensities of the disc, CSF and spinal cord were acquired using custom-made software. The signal intensity of the spinal cord was correlated with the signal intensity of the adjacent CSF (criterion reference) in a sample of subjects with visually clean CSF-samples. The measured disc signal intensities were correlated with age in three ways; without adjustment and adjusting for CSF and spinal cord. Pearson’s r was used for the correlations and intraclass correlation coefficient (ICC) to examine the interobserver repeatabilities of spinal cord signal intensity measurements.
Results: Spinal cord signal intensity measurements were repeatable (ICC=0.99–1). Clean CSF-samples correlated highly with the signal intensities of the spinal cord (r=0.91–0.99). The correlations of spinal cord-adjusted disc signal intensity with age (r=−0.30 to −0.40) were somewhat higher than for age and CSF-adjusted disc signal intensity (r=−0.26 to −0.36). In comparison, correlations between unadjusted disc signal intensity and age were low (r=−0.11 to −0.19).
Conclusion: Spinal cord is a good alternative intrabody reference to the current gold standard (CSF) for disc signal intensity measurements in the thoracic spine.
Correspondence should be addressed to Mr J. O’Dowd, Honorary Secretary at SBPR c/o BOA, Royal College of Surgeons, 35–43 Lincoln’s Inn Fields, London, WC2A 3PE.
