Abstract

BACKGROUND: Degenerative disease of the lumbar spine is characterized by symptoms related to the affected nerve root. A recently described method allows the classification of the roots in relation to the occurrence of compression on its course. This method can serve as a clinical selection tool and decision support for semi-invasive pain therapy in back pain patients.

MATERIAL AND METHODS: We examined 40 lumbar spine MRIs in 3 sessions of transcription each, according to the method being evaluated. Every MRI evaluation was performed by each of 3 different observers. Intra- and interobserver reproducibility was calculated using chance-corrected agreement using a weighted kappa value with quadratic weights to assess reliability for each nerve root separately.

RESULTS: We found high intraobserver agreement in indication of the root with most pronounced interference due to potential compression by degenerative changes, at the level mean kappa=0.81 (with 95% CI, range 0.04). Less agreement was observed in the interobserver evaluation test with the mean kappa=0.75 (95% CI within the range not exceeding 0.03), although it still reached the substantial agreement.

CONCLUSIONS: This zstudy provides evidence for substantial inter- and intraobserver agreement for the decision support method allowing selection of the most serious nerve structure compression in degenerative disease of the lumbar spine based on of the MRI description.

Keywords: Spinal Canal - pathology, Radiculopathy - pathology, Observer Variation, Reproducibility of Results, Neurodegenerative Diseases - pathology, Magnetic Resonance Imaging - methods, Lumbar Vertebrae - pathology, Spinal Nerve Roots - pathology