Surgical interventions for the treatment of chronic neck pain, which affects 330 million people globally, include fusion and cervical total disc replacement (CTDR). Most of the currently clinically available CTDRs designs include a metal-on-polymer (MoP) bearing. Numerous studies suggest that MoP CTDRs are associated with issues similar to those affecting other MoP joint replacement devices, including excessive wear and wear particle-related inflammation and osteolysis. A standard ISO testing protocol was employed to investigate a device with a metal-on-metal (MoM) bearing. Moreover, with findings in the literature suggesting that the testing protocol specified by ISO-18192-1 may result in overestimated wear rates, additional tests with reduced kinematics were conducted.

Six MoM CTDRs made from high carbon cobalt-chromium (CoCr) were tested in a six-axis spine simulator, under the ISO-18192-1 protocol for a duration of 4 million cycles (MC), followed by 2MC of modified testing conditions, which applied the same axial force as specified in ISO-18192-1 (50-150N), but reduced ranges of motion (ROM) i.e. ±3° flexion/extension (reduced from ±7.5°) and ±2° lateral bending (reduced from ±6°) and axial rotation (reduced from ±4°). Foetal bovine serum (25% v/v), used as a lubricant, was changed every 3.3×10⁵ cycles and stored at −20°C for particle analysis. Components were measured after each 1×10⁶ cycles; surface roughness, damage modes and gravimetric wear were assessed. The wear and roughness data was presented as mean ±95% confidence interval and was analysed by one-way analysis of variance (ANOVA) (p=0.05).

The mean wear rate of the MoM CTDRs tested under the ISO protocol was 0.246 ± 0.054mm³/MC, with the total volume of wear of 0.977 ± 0.102mm³ lost over the test duration (Fig. 1). The modified testing protocol resulted in a significantly lower mean volumetric wear rate of 0.039 ± 0.015mm³/MC (p=0.002), with a total wear volume of 0.078 ± 0.036mm³lost over the 2MC test duration. Under both test conditions, the volumetric wear was linear; with no significant bedding-in period observed (Fig. 1). The mean pre-test surface roughness decreased from 0.019 ± 0.03µm to 0.012 ± 0.002µm (p=0.001) after 4MC of testing, however surface roughness increased to 0.015 ± 0.002µm (p=0.009) after the additional 2MC of modified test conditions. Following 4MC of testing, polishing marks, observed prior to testing, had been removed. Consistently across all components, surface discolouration and multidirectional, criss-crossing, curvilinear and circular wear tracks, caused by abrasive wear, were observed. Reduced ROMs testing caused similar types of damage, however the circular wear tracks were smaller in size, compared to those produced during testing under the ISO protocol.

The wear rates exhibited by MoM CTDRs tested under ISO-18192-1 testing protocol (0.246mm³/MC) were lower, when compared to CTDR designs incorporating MoP bearings, as well as MoM lumbar CTDRs. Wear rates generated under a modified ISO testing protocol were reduced tenfold, similarly to findings that have previously been reported in the literature, and support the hypothesis that the testing protocol specified by ISO-18192-1 may overestimate wear rates. Characterisation of particles generated by MoM CTDRs and biological consequences of those remain to be determined.

For figures/tables, please contact authors directly.