To analyse how the auditory brainstem response changes in patients with sudden sensorineural hearing loss.

Data were collected via retrospective medical chart review.

Forty-three patients were included in this study. The mean latency of auditory brainstem response wave 1 was significantly longer for the affected side than for the unaffected side (p = 0.003). The mean latency of auditory brainstem response wave 1 was significantly shorter, and the mean amplitude of auditory brainstem response wave 1 was significantly larger, in the good response group compared to the poor response group. In forward conditional logistic regression analysis, auditory brainstem response wave 1 latency was an independent predictor of a good response (odds ratio = 34.37, 95 per cent confidence interval = 1.56–757.15, p = 0.025).

In patients with sudden sensorineural hearing loss, the latency of wave 1 of the auditory brainstem response was significantly increased and was related to prognosis.

To investigate the hypothesis of cochlear and retrocochlear damage in scrub typhus, using evoked response audiometry.

Prospective, randomised, case–control study.

The study included 25 patients with scrub typhus and 25 controls with other febrile illnesses not known to cause hearing loss. Controls were age- and sex-matched. All subjects underwent pure tone audiometry and evoked response audiometry before commencing treatment.

Six patients presented with hearing loss, although a total of 23 patients had evidence of symmetrical high frequency loss on pure tone audiometry. Evoked response audiometry found significant prolongation of absolute latencies of wave I, III, V, and wave I–III interpeak latency. Two cases with normal hearing had increased interpeak latencies. These findings constitute level 3b evidence.

Findings were suggestive of retrocochlear pathology in two cases with normal hearing. In other patients, high frequency hearing loss may have led to altered evoked response results. Although scrub typhus appears to cause middle ear cochlear and retrocochlear damage, the presence of such damage could not be fully confirmed by evoked response audiometry.