Comparison of inter- and intra-observer reliability among the three classification systems for cervical spinal canal stenosis.

PURPOSE: The aim is to analyze the agreement between different types of physicians in terms of the inter-observer and intra-observer reliability in addition to the agreement between the experienced and non-experienced physicians with respect to three different classification systems for diagnosis of cervical spinal canal stenosis.

METHODS: Total nine doctors including experienced group of three doctors and non-experienced group of six doctors classified the patients according to three different classification in an independent, blinded manner using magnetic resonance imaging (MRI) to diagnose cervical canal stenosis. MRI slice included sagittal plane (midline cut) and an image slice from each horizontal plane that penetrated the right center of each disk (C3-4, C4-5, C5-6, and C6-7) was made by PPT format.

RESULTS: For the inter-observer reliability, Vaccaro et al.'s classification system showed the excellent reproducibility, followed by Muhle et al. and Kang et al. All three classification systems showed excellent reproducibility and substantial agreement in terms of the intra-observer reliability.

CONCLUSIONS: All three classification systems showed excellent reproducibility and also displayed a substantial agreement. The classification system used by Vaccaro et al. was proven to be a method with substantial agreement both in the experienced group and the non-experienced group. It can be a useful classification system for simplifying communication among all physicians.