Abstract
With increasing use of imaging in the investigation of cerebello-pontine angle (CPA) tumours, the role of audio-vestibular and electrophysiological testing has changed. Field performance data on the efficiencies of these tests to screen for CPA tumours are lacking, but must be known to choose an appropriate testing strategy. A prospective observational study of 237 patients attending a neuro-otology clinic for audio-vestibular investigation was carried out. The aim was to provide field performance data on which to base an effective protocol to screen for CPA tumours. All patients presenting at the ENT department and meeting any of the following criteria were referred to the neuro-otology clinic and included in the study: (1) asymmetrical sensorineural hearing loss, (2) unexplained asymmetrical tinnitus with normal hearing thresholds, (3) unilateral hearing difficulties with normal hearing thresholds and (4) other neurological indications. In addition to audio-vestibular and auditory brainstem response (ABR) investigation, every patient underwent computed tomography (CT), with magnetic resonance imaging (MRI) in cases having marginal results on CT, to exclude or confirm the presence of a tumour. Pass or fail on each test was based on a priori criteria from other studies. Eighteen patients were found to have CPA tumours. ABR testing was the only effective procedure for screening, but had some limitations. A contingent protocol using ABR in all cases except those with asymmetrical tinnitus and normal hearing thresholds, those with severe hearing loss, and those with neurological signs, was retrospectively defined: the exceptions would go straight to CT. This protocol would have missed two of the 18 tumour patients. CT scanning alone would have missed one small intra-canalicular tumour, which was picked up on MRI triggered by abnormal ABR. Based on the results from the present study we conclude there is no effective screening protocol for detecting CPA tumours, as MRI scanning with gadolinium enhancement will identify virtually all tumours. Where MRI is available but waiting lists are long, the described strategy using ABR to select priority referrals for MRI scanning is recommended.