MRI evaluation of lumbar spinal stenosis: is a rapid visual assessment as good as area measurement?

Abstract

Lumbar spinal stenosis (LSS) is commonly assessed on MRI by measuring dural sac cross-sectional area (DSCA). A new method, morphological grading A-D, has recently been introduced as an alternative method. The aim of this study is to compare these two different methods for assessing LSS on MRI and study their reliability and intercorrelation. On pretreatment MRI of 84 patients, two experienced radiologists independently classified level L2/L3, L3/L4 and L4/L5 as no, relative or significant stenosis using both methods. Agreement was analyzed by weighted Kappa. The correlation between the two methods was analysed using Spearman correlation, and visualized in a box plot. The interobserver agreement (95 % CI) was 0.69 (0.61-0.77) and 0.65 (0.56-0.74), respectively. The intraobserver agreements for DSCA were 0.77 (0.60-0.74) and 0.80 (0.66-0.93). On morphological grading A-D it was 0.78 (0.65-0.92) and 0.81 (0.68-0.94). The correlation coefficient between the two methods was 0.85 (p < 0.001). Grades C and D were under the limit value for significant stenosis using the DSCA. The study shows that the inter- and intraobserver agreements of DSCA and morphological grading A-D were acceptable and their intercorrelation is strong. Both methods may be used in the MRI evaluation of LSS.