Validation study of a clinical diagnosis support tool for lumbar spinal stenosis.

BACKGROUND: A clinical diagnosis support tool for lumbar spinal stenosis was developed by the Japanese Society for Spine Surgery and Related Research. However, the use of this tool has not yet been validated.
METHODS: Patients with symptoms in the lower extremities and who visited the Department of Orthopedics initially were recruited to the study. Orthopedic physicians who were not spine specialists completed the support tools. Spine specialists examined the patients, made a diagnosis, and completed the lumbar spine examination sheet made for the study. The support tool and lumbar spine examination sheet were sent to a central panel comprising four panelists who then decided on a final diagnosis.
RESULTS: In total, 118 patients were evaluated, including 62 males and 56 females. Lumbar spinal stenosis was diagnosed in 58 and nonlumbar spinal stenosis in 60 patients. The mean score in the lumbar spinal stenosis group was 12.2 points (median 13 points). In the nonlumbar spinal stenosis group, the mean score was 7.5 points (median 7 points). Sensitivity was 0.948, and specificity was 0.40.
CONCLUSIONS: Patients with lumbar spinal stenosis with a very low score were diagnosed with mild lumbar spinal stenosis, whereas nonlumbar spinal stenosis patients with a very high score were diagnosed as suffering from spine disease and needing special treatment by spine surgeons. Our results validate the use of the support tool for the diagnosis of lumbar spinal stenosis. Although the specificity observed in the present study was lower than that reported at development, we conclude that this support tool is useful for screening patients with lumbar spinal stenosis.