Diffusion-Weighted Magnetic Resonance Imaging of Normal and Degenerative Lumbar Intervertebral Discs: A New Method to Potentially Quantify the Physiologic Effect of Physical Therapy Intervention

Abstract

Observational, repeated measures design. To determine the reliability of the apparent diffusion coefficient (ADC) calculated from diffusion-weighted magnetic resonance images (MRI) of the nuclear region of lumbar intervertebral discs (IVDs), to investigate the differences in the ADC based upon T2-signal intensity, and to examine the test-retest variation in these measures obtained from subjects undergoing serial, diffusion-weighted MRI scans. Impaired diffusion of water within the lumbar IVD is a central characteristic of degenerative disc disease. Diffusion-weighted MRI scans can provide quantitative estimates of water diffusion and may be useful to evaluate the physiologic effects of healing or the change in hydration related to interventions such as traction, manual therapy, or exercise on normal and degenerative lumbar IVDs. Thirty subjects underwent T2 -weighted and diffusion-weighted lumbar MRI scans. Twenty-one of these subjects underwent a second diffusion-weighted MRI scan 4 to 7 weeks after the initial scan. The ADC was calculated from midsagittal diffusion-weighted images for the IVDs of L1-2 to L5-S1. To assess reliability, repeated measures of the ADC were performed on the first 16 scans. The T2-signal of the nuclear region of each disc was classified as hyperintense, intermediate, or hypointense, and its relationship to the mean ADC of the nuclear region was determined. Test-retest variation in the ADC was described using the coefficient of variation (CV), plus or minus the width of the 95% confidence interval of the standard error of measurement (SEM). Intraclass correlation coefficients for estimates of intrarater and interrater reliability ranged from 0.95 to 0.99 and the SEM ranged from 0.006 to 0.026 X 10-3 mm2/s. The mean ADC was significantly greater for hyperintense IVDs compared to intermediate and hypointense IVDs. The CV plus or minus the 95% CI of the SEM between scans ranged from 9.0% to 13.6% for all discs, 6.1% to 10.1% for hyperintense discs, and 13.1% to 23.7% for intermediate discs. The prevalence of hypointense discs was too low to make meaningful judgments about their normal degree of variation over time. The ADC of the nuclear region of the lumbar IVDs may be reliably measured from diffusion-weighted images. Degenerative discs had lower mean ADC values than normal discs but demonstrated greater variation between scans. Diffusion-weighted imaging may be a useful procedure to assess change in diffusion of water in lumbar discs that occurs over time.