Assessing the Clinical Value of Objective and Patient-Reported Audiovestibular Outcome Measures in the Risk Estimation of Systemic Cobalt Toxicity for Patients With a Metal-on-Metal Hip Implant

Objectives: Based on limited evidence from case reports and small cohort studies of metal-on-metal (MoM) hip implant patients with presumed systemic cobalt (Co) toxicity, and a few animal studies on Co-induced damage to the otovestibular system, it was hypothesised that Co exhibits an ototoxic potential alongside other systemic manifestations. Preliminary investigations from our research group in MoM patients confirmed this assumption for the auditory system, whereas no signs of Co-induced vestibular impairment were detected, and a clear dose-response relationship between the auditory function and the blood Co levels was lacking. Therefore, the current study aimed to extend and validate these findings in a larger sample of MoM patients and control subjects, to explore the potential clinical value of audiovestibular outcome measures in the risk estimation of systemic Co toxicity in this patient population.

Design: Fifty patients (32 to 68 years) with a primary unilateral/bilateral MoM hip implant were matched for age, gender, and noise exposure to 50 nonimplanted control subjects. Both groups underwent the same protocol, consisting of an objective auditory [i.e., conventional and high-frequency audiometry, transient-evoked and distortion (TEOAEs and DPOAEs), auditory brainstem responses] and vestibular (i.e., cervical and ocular vestibular evoked myogenic potentials, horizontal and vertical video head impulse tests) test battery, a questionnaire inquiring auditory, balance, and general neurological symptoms, and a blood sample collection to determine the plasma Co concentration.

Results: The auditory test battery presented consistently higher audiometric thresholds in the MoM patient group, with group differences ranging from 2.1 to 5.7 dB in the lower frequencies (0.25 to 6.0 kHz) and from 4.6 to 9.3 dB in the high frequencies (8.0 to 14.0 kHz). Group differences at high frequencies were statistically significant ( p ≤ 0.001). Additionally, significantly lower TEOAE ( p = 0.009) and DPOAE ( p < 0.001) amplitudes were observed in the MoM patients when the better ear was included in the analysis, and more absent TEOAE and DPOAE responses were found between 1.0 and 4.0 kHz (0.008 ≤ p ≤ 0.039). Within the vestibular test battery, the MoM patients showed longer N1 and P1 ocular vestibular evoked myogenic potentials latencies for the left ear, and lower video head impulse test gains for the left anterior and right posterior semicircular canals (0.005 ≤ p ≤0.035). The patient-reported (questionnaire) outcome delivered no significant group differences ( p > 0.01). Six patients had elevated Co levels according to our local institutional threshold (>4 or 5 µg/l for unilateral or bilateral MoM hip implants, resp.), but their audiovestibular outcome measures did not differ significantly from those of the other patients.

Conclusions: Corresponding to our preliminary investigations, the results indicate possibly Co-induced (predominantly high-frequency) auditory impairment, probably triggered by toxic damage to the cochlear structures. However, the low mean difference values, the lack of group differences for the patient-reported outcome measures, and the lack of any relationship with the blood Co levels strongly reduce the clinical relevance of these findings. Therefore, the risk of Co-induced ototoxic impairment is considered to be clinically negligible for the majority of MoM hip implant patients, and the use of auditory tests in the risk estimation of systemic Co toxicity should be decided on a case-by-case basis.