Short-term test-retest reliability of a few proxy measures of cochlear synaptopathy

Assessment of cochlear synaptopathy (CS) in humans depends on proxy measures of auditory nerve function; nonetheless, their clinical use fundamentally relies upon measurement reliability. This study assessed the short-term test-retest reliability of various proxy measures of CS across three test sessions.

The study employed a quasi-experimental, single-group, time-series design. Thirty adults with normal auditory abilities participated in three test sessions spaced at short inter-session intervals. Measures included absolute auditory brainstem response (ABR) Wave I amplitude, growth function of ABR Wave I amplitude across 60 to 90dBnHL, summating potential to action potential (SP/AP) amplitude ratio, acoustic middle ear muscle reflex (MEMR) threshold, and MEMR strength. Repeated-measures ANOVA (RMANOVA) was used to compare differences across sessions. Reliability was evaluated using intraclass correlation coefficient (ICC), within-subject coefficient of variation (wCV), and standard error of measurement (SEM).

RMANOVA revealed no significant differences across sessions for most measures, except MEMR threshold for broadband elicitor. Each measure exhibited a unique reliability profile. ABR wave I amplitude showed good to excellent ICC and low SEM, indicating high relative reliability and low measurement error. In contrast, ABR wave I growth slope and SP/AP ratio exhibited moderate to poor ICCs, respectively, relatively high wCV values, and low SEM, reflecting greater within-subject variability and relatively poor reliability. MEMR threshold measures demonstrated higher stability than MEMR strength measures.

The results indicated that whereas various proxy measures of CS showed good reliability, derived and ratio-based measures are marked by increased within-subject variability, thereby limiting their interpretability and consistency as reliable measures.