Purpose: Semiquantitative (SQ) grading of the intervertebral disk is based on signal intensity, intervertebral disk height and differentiation of the nucleus and annulus. Shortcomings of SQ grading are insensitivity to change in longitudinal assessment. The T2 values of intervertebral disks are a reflection of hydration and proteoglycan content, of collagen structure and fibre texture. Hydration and consequently degeneration of the nucleus is directly correlated with T2 values. T2 mapping of the nucleus might potentially be applied as an additive surrogate marker of cross-sectional and longitudinal assessment of disk integrity. Aim of study was to asses T2 relaxation times of the nucleus pulposus of intervertebral disks of the lumbar spine by T2 mapping and correlate mean T2 values with a descriptive SQ classification of disk morphology. Secondary aim was interand intrareader evaluation of SQ grading and T2 mapping. Methods: Lumbar MRI was performed on volunteers without lumbar pain on a 1.5 T system (Magnetom Avanto, Siemens, Erlangen, Germany). The intervertebral disks of the L1/2 to the L5/S1 levels were assessed. The protocol for MR measurements consisted of a multi-echo T2 spin echo sequence (TR 1,200 ms, 11 echoes of 12.7 ms increments, Field of view 280×280 mm, 8 mm slice thickness). Total acquisition time was 13 min 22 sec. The intervertebral discs were SQ assessed according to their morphology using the Pfirrmann-classification of disk degeneration. T2 relaxation times were obtained using a pixel wise, monoexponential, non-negative least squares (NNLS) fit analysis. Circular-shaped regions of interest (ROI) were drawn manually in the center of the nucleus pulposus by two readers separately using OsiriX version 3.6.1 as DICOM-viewer. Descriptive statistics using the mean T2 values and 95% confidence intervals were applied for comparison between T2 maps and morphologic grading. Interand Intraobserver reliabilitiy for T2 mapping and morphologic grading were calculated by generalized estimation equations (GEE). Statistical analyses were performed using PASW statistics version 17.0.2 for mac (SPSS Institute, Chicago, IL, USA). Results: 21 volunteers were examined. Mean age was 26.6 years (range 24-31), mean BMI 21.9 (range19.0-26.2). 105 disks were analyzed. The distribution of morphologic grading was as follows: 10 grade 1, 52 grade 2, 26 grade 3, 10 grade 4 and 7 grade 5. Mean T2 values for each grade were 152ms (± 22,5), 118ms (± 24,5), 86ms (± 19,3), 56ms (± 13,4) and 52ms (± 8,0) respectively. Values for grades 1-4 showed significant differences (p<0.001), while the difference of mean T2 values for grades 4 and 5 did not reach statistical significance (p=0.25). Interand intraobserver reliability for SQ grading and T2 mapping was good. Conclusions: T2 values in the nucleus pulposus significantly decrease with increasing grades of morphologic degeneration. T2 mapping might be a promising complimentary tool to assess cross-sectional and longitudinal changes of disk degeneration or disk repair.
Read full abstract