New grading system for lumbar foraminal stenosis: Evaluation of teachability and correlation of operability

Abstract

PURPOSE Our aim was to investigate the reliability and teachability of a new grading system for lumbar foraminal stenosis that was introduced by Lee et al. (i.e., the Lee system). The second aim was to evaluate the correlation between the new imaging grading system and operability.

MATERIALS AND METHODS The study was approved by the institutional review board of our hospital. A total of 100 patients (27 men, 73 women

mean age: 68.7 years

age range: 60-89 years) who underwent magnetic resonance imaging (MRI) of the lumbar spine were enrolled. Two musculoskeletal radiologists, two neurosurgeons, two orthopedic surgeons, and four radiology residents independently analyzed a total of 600 lumbar foramina (both L3-L4, L4-5 and L5-S1) and graded them using the Lee system. Interobserver and interobserver agreement was analyzed using the interclass correlation coefficient (ICC), the percentage agreement, and kappa statistics. Operability of each lumbar foramen was determined by four surgeons. The correlations between operability and stenosis grading were assessed using Spearmans correlation.

RESULTS Interobserver agreement among ten observers showed excellent reliability, as defined by the ICC value of 0.938. The average percentage agreement was 65.8%, and the overall kappa value was 0.55. The ICC for the intraobserver agreement of two observers was 0.897 and 0.787, indicating excellent agreement. The percentage agreements and kappa value were 83% (κ = 0.81) and 74% (κ = 0.65), respectively. Lumbar foraminal stenosis grading and operability were significantly correlated (rho Spearmans: 0.550, 0.717, 0.765, and 0.667

p < 0.0001)

CONCLUSION The Lee system for lumbar foraminal stenosis is a reliable classification system for surgeons, radiologists, and trainees.