The Total Disc Replacement Osteolysis Grading Scale - a simple, reliable and quantifiable tool for assessing, managing and reporting osteolysis after cervical total disc replacement.

Abstract

Cervical Total Disc Replacement (CTDR) related periprosthetic osteolysis is inconsistently reported. The purpose of this study is to assess the previously published TDR Osteolysis Grading Scale, a quantitative classification system of CTDR-related osteolysis, and to assess its reliability and utility using Computer Tomography (CT) and X-ray imaging. Participants were assigned to Groups A (CT, 20 participants, 27 implants) and B (X-ray, 20 participants, 26 implants). Four blinded raters independently measured osteolytic cysts and both vertebral endplates, calculated the percentage of osteolysis and assigned osteolysis grades: Grade 0, no osteolysis; Grade 1, < 50% of 1 endplate; Grade 2, < 50% of both endplates; Grade 3, ≥ 50% of 1 endplate; Grade 4, ≥ 50% of both endplates. Inter-rater reliability and sensitivity were assessed using intraclass correlation coefficients (ICC), Fleiss' and Cohen's Kappa statistics and sensitivity testing. Group A had good reliability for osteolytic cyst measurements (ICC = 0.78-0.79), poor reliability for endplate measurements (ICC = 0.41-0.45) and moderate reliability for osteolysis grades (κ-coefficient = 0.42). Group B had moderate reliability for osteolytic cyst (ICC = 0.55-0.57) and endplate measurements (ICC = 0.55-0.64) and fair reliability for osteolysis grades (κ-coefficient = 0.30). Reliability between reference grades and assigned grades was moderate for Group A (κ-coefficient = 0.46-0.60) and fair for Group B (κ-coefficient = 0.08-0.35). Sensitivity was greater for high grade osteolysis (grade 3 and 4) in Group A (0.80-1) than Group B (0.17-0.50). The TDR Osteolysis Grading Scale demonstrated good reliability and sensitivity amongst raters with CT. Implementing this tool may aid in standardising osteolysis reporting, monitoring osteolysis progression and clinical decision-making.