Lumbar Spinal Canal Quantification among a cohort of Saudi Population in National Oncology Tertiary Center: Magnetic Resonance Imaging–Based Study

Background Lumbar canal stenosis contributes to neuropathic pain and radiculopathy due to neural element compression. Qualitative grading of lumbar stenosis on magnetic resonance imaging (MRI) suffers from notable interobserver variability, limiting reproducible clinical decision-making. Objective To establish quantitative MRI reference benchmarks for lumbar spinal canal dimensions in a cohort of Saudi patients imaged at a national tertiary oncology center, assess interobserver measurement reliability, determine the impact of ligamentum flavum thickening, and evaluate cross-modal dimensional discrepancies compared with computed tomography (CT). Materials and Methods This retrospective observational study included 317 consecutive adult Saudi patients (aged 18–90 years) who underwent lumbar spine MRI at King Faisal Specialist Hospital and Research Center (KFSHRC) between March 2023 and April 2024. Patients with primary spinal malignancy, vertebral metastasis, fractures, scoliosis, or prior spine surgery were excluded. Quantitative metrics—cross-sectional area (CSA), midsagittal anteroposterior diameter (APD), and midsagittal transverse diameter (TD)—were measured at L1–S1 levels by two independent musculoskeletal radiologists. Interobserver reliability was a randomly selected subset (n = 30) using a two-way mixed-effects consistency intraclass correlation coefficient ICC(3,1). A pilot cross-modal validation sub-study was conducted in four patients (n = 4) with paired, same-patient MRI and CT scans. Results The study cohort comprised 317 subjects (mean age 45 ± 14 years; 63.1% female, 36.9% male; mean body mass index [BMI] 28 ± 6 kg/m²). Across lumbar levels, mean vertebral body CSA ranged from 239 to 283 mm², APD from 14.12 to 15.78 mm, and TD from 20.57 to 21.44 mm, with the greatest variability observed at L5 (standard deviation [SD] ± 5.00 mm) and L5–S1 (SD ± 7.00 mm). In the ICC analysis (n = 30), 11 of 15 continuous parameters demonstrated moderate-to-good reliability (ICC = 0.500–0.810), with highest agreement at L1 APD (ICC = 0.810) and L1 CSA (ICC = 0.748). In paired MRI–CT analysis (n = 4), MRI systematically underestimated dimensions relative to CT, with discrepancies expanding caudally: L5 CSA (−52.0%), L4 CSA (−31.4%), and L4 TD (−22.1%). Patients with ligamentum flavum thickening exhibited significant reductions in CSA and diameters across all levels ( p < 0.001 ). Conclusion This study provides preliminary institutional MRI reference benchmarks for lumbar spinal canal dimensions in a Saudi tertiary care cohort. Reliability is robust in upper segments but lower at L4–L5 due to trefoil morphology, soft-tissue crowding, and lordotic partial volume effects. Substantial MRI underestimation compared to CT at L4–L5 indicates that modality-specific measurements cannot be used interchangeably for surgical decompression planning.