ABSTRACT
OBJECTIVES: Evoked tympanic membrane displacement (TMD) measurements, quantified by V m, record small volume changes in the ear canal following stimulation of the acoustic reflex. V m shows a correlation with intracranial pressure (ICP) and has been proposed as an option to non-invasively measure ICP. The spontaneous pulsing of the tympanic membrane, driven by the cardiovascular pulse, may contaminate the recordings and contribute to high measurement variability in some subjects. This study hypothesised that the larger the spontaneous vascular pulse, the larger the variability in V m. MATERIALS AND METHODS: Spontaneous and evoked TMD data from each ear in the sitting and supine position were recorded from 100 healthy volunteers using the MMS-14 CCFP analyser. ECG was also recorded to identify each heartbeat. Using bespoke software written in Matlab, spontaneous data were analysed to produce average pulse amplitude (PA) waveforms and evoked data were analysed to calculate average V m and its standard deviation. Averaged spontaneous PA was plotted against V m variability and Pearson's correlation coefficient was calculated to test for a significant linear relationship. RESULTS: There was a strong positive correlation between PA and V m variability in all conditions: left sitting, r = 0.758; left supine, r = 0.665; right sitting, r = 0.755; right supine, r = 0.513. All were significant at p < 0.001. CONCLUSION: This study shows that large V m variability is associated with a large spontaneous vascular pulse. This suggests that efforts to reduce vascular pulsing from recordings, either by a subtraction technique during post-processing or ECG-gating of the evoking stimulus, may improve reliability of the V m measurement.
