MRI findings in candidates for lumbar disc prosthesis: reliability and relationship to disability and pain

Abstract

Background and objective: In candidates for lumbar disc prosthesis, limited reliability data exist for magnetic resonance imaging (MRI) findings relevant to planning of treatment and to evaluation of outcome after treatment. In this subgroup of patients with low back pain (LBP), we assessed the reliability of degenerative MRI findings and change in such findings over time. How the sum of MRI findings relates to the degree of disability and LBP was also evaluated since this is not clear. Materials and Methods: 170 of 173 patients aged 25-55 years, with LBP ≥ 1 year, Oswestry Disability Index (ODI) score ≥ 30 %, and localized degenerative MRI changes at L4/L5 and/or L5/S1 were included. On pre-treatment images three experienced radiologists independently rated Modic changes, disc findings, and facet arthropathy (FA) at L3-S1. Two of the radiologists rated progress and regress of the same findings on 2-year follow up images (n = 126). Agreement was analysed using the kappa statistic. How pre-treatment MRI total score related to the ODI (n = 170) and LBP intensity scores (n = 165) was analysed using multiple linear regression adjusting for age, gender, body mass index (BMI), smoking, and anxiety/depression. Results: Overall interobserver agreement was generally moderate or good (kappa 0.40-0.77) at L4-S1 for Modic changes, nucleus pulposus signal, disc height, posterior HIZ, and disc contour, and fair (kappa 0.24) at L4/L5 for FA. Intraobserver agreement was mostly good or very good (kappa 0.60-1.00). Image comparison indicated good interobserver agreement on progress and regress (prevalence and bias adjusted kappa (PABAK) 0.63-1.00) for Modic changes, posterior HIZ, disc height, and disc contour at L3-S1 and for nucleus pulposus signal and FA at L3/L4; and moderate interobserver agreement (PABAK 0.46-0.59) on decreasing nucleus signal and increasing FA at L4-S1. The MRI total score was not related to ODI (regression coefficient 0.12, P = 0.79) or LBP intensity (regression coefficient 0.64, P = 0.37). Conclusions and consequences: In candidates for lumbar disc prosthesis or fusion, Modic and disc findings, and change in these findings over time, have acceptable reliability for use in treatment planning and research, but the sum of these findings are unlikely to explain variation in current pre-treatment disability and pain.