Lumbar Spine Anatomy in Supine versus Weight- Bearing Magnetic Resonance Imaging: Detecting Significant Positional Changes and Testing Reliability of Quantification.

Abstract

Testing between and within group differences and assessing reliability of measurements. To study and compare lumbar spine morphology in supine and weight-bearing (WB) magnetic resonance imaging (MRI). Upright lumbar MRI may uncover anatomical changes that may escape detection when using conventional supine imaging. This study quantified anatomical dimensions of the lumbar spine in the supine and WB MRI, compared specific morphometric differences between them, and tested the intra-rater reliability of the measurements. Repeated measures analysis was used to compare within- and between-session measurements performed on the supine and WB images. Reliability and agreement were assessed by calculating intraclass correlation (ICC) coefficient. Data from 12 adults without any history of back pain were used in this study. Sagittal T2-weighted images of the lumbar spine were acquired in the supine and WB positions twice (in two separate sessions scheduled within a week). Linear, angular dimensions, and cross-sectional areas (CSAs) were measured using proprietary software. Supine and WB data acquired from the two imaging sessions were tested for intra-rater reliability. Quantified data were normalized for each session to test the significance of differences. ICC was calculated to test the reliability of the measurements. Linear, angular, and CSA measurements demonstrated strong within-position (supine and WB) correlations (r -values, 0.75-0.97). Between-position (supine vs. WB) differences were significant for all measured dimensions (p<0.05). Between-session measurements demonstrated a strong correlation (r -values, 0.64-0.83). Calculated ICC showed strong agreement among the measurements. Anatomical dimensions of the lumbar spine may demonstrate consistent and significant differences between supine and WB MRI for specific structural parameters.