Analysis of Threshold Differences and Correlations Between Auditory Brainstem Response and Pure Tone Audiometry Across Degrees and Types of Hearing Loss

Purpose: Pure-tone audiometry (PTA) and auditory brainstem response (ABR) are widely used for hearing assessment. This study investigated PTA-ABR threshold differences, correlations, and agreement according to frequency, hearing loss severity, PTA averaging method, and audiometric configuration in adults with moderate-to-severe hearing loss.Methods: This retrospective study analyzed 1,132 ears from 712 adults who underwent PTA and ABR testing at a tertiary hospital. PTA thresholds (0.125-8 kHz) and five PTA averages (2 frequency average [FA] (L), 2FA(H), 3FA, 4FA, and 6FA) were compared with ABR thresholds.Results: PTA-ABR threshold differences were smallest at 1-2 kHz and generally increased above 3 kHz, particularly in steeply sloping audiograms. Among the PTA averaging methods, 6FA showed the smallest mean absolute threshold difference (3.89 dB), followed by 3FA (4.26 dB) and 4FA (4.75 dB). The 6FA also showed relatively high correlations with ABR (r = 0.890-0.927) and the highest intraclass correlation coefficient (0.904; 95% confidence interval, 0.896-0.911). Bland-Altman analysis revealed the smallest mean bias for 6FA (-1.33 dB), with 95% limits of agreement ranging from -9.62 to 6.97 dB. However, differences between 6FA and 4FA were relatively small, and individual-level variability remained.Conclusion: PTA-ABR threshold relationships varied according to frequency, averaging method, and audiometric configuration. Thresholds at 1-2 kHz and the 6FA showed relatively close correspondence with ABR. However, the observed differences among averaging methods were limited, and high correlations did not necessarily indicate clinical interchangeability. Further studies are needed to establish individual-level predictive accuracy and clinical applicability.