Reliability in grading the severity of lumbar spinal stenosis

Stenosis of the lumbar spinal canal is a major cause of disability and lost productivity. Computed tomography (CT) is used commonly to assess the presence and severity of spinal stensosis, because it is relatively inexpensive, readily available, and has few adverse effects. The ability of four surgeons to agree about the presence and severity of lumbar spinal stenosis based on plain CT scans was evaluated from 30 scans of varying stenosis severity (normal to severe). Kappa, a measure of chance-corrected agreement, was calculated. Surgeons exhibited moderate agreement for the presence or absence of spinal stenosis (kappa = 0.58 +/- 0.06). Agreement regarding the severity of stenosis, when present, was poor (kappa = 0.26 +/- 0.04). The ability of surgeons to agree was not improved when individual features of the CT scans were assessed (facet joint arthrosis, ligamentum flavum hypertrophy, disk protrusion, and nerve root impingement). This study suggests that CT scans are not a reliable method by which to examine the severity of lumbar spinal stenosis.