No abstract

Objective: To compare the interobserver reliability of measurements of psoas morphology and lumbar lordosis in different positions and to standardize the performance of magnetic resonance imaging in the prone and lateral positions. Methods: This is a cross-sectional study carried out with asymptomatic volunteers of both sexes, aged over 18 years, with no known pathological changes in the lumbar region. Magnetic resonance imaging of the lumbar spine was performed in the supine, right lateral decubitus and prone positions, obtaining images in T2-weighted sequences in the sagittal and axial planes. The distances were measured from the psoas to the vertebral plateau and from the psoas to the lumbar plexus. The exams were assessed by two independent, blinded orthopedists. Results: There was excellent agreement between the measurements of vertebral size (ICC=0.92), low agreement for plexus distance (ICC=0.63) and high agreement for the anterior margin (ICC=0.84). Conclusion: There was good reproducibility of 2 of the 3 measures proposed, suggesting that the technique in the lateral and prone positions is capable of generating quality images. Level of Evidence 3B; Prospective.