Acoustic parameters of snoring sound to compare natural snores with snores during 'steady-state' propofol sedation

Abstract

Objectives: To investigate the acoustic similarity between natural and sedation-induced snores.

Design: Prospective observational study.

Setting: University Hospital Aintree, Liverpool, UK.

Participants: Twenty-one patients, who had already had overnight snore recordings, completed a pre-operative sleep nasendoscopic examination. Endoscopic examination of the upper aero-digestive tract was performed at sequentially increasing, steady-state sedation levels, using intravenous propofol administered according to a weight/time-based algorithm to predict blood and effect site (tissue) concentrations. At each sedation level at which snoring occurred, snoring sound was recorded. From these samples, snore files, comprising the inspiratory sound of each snore were created. Similarly, from natural snores recorded pre-operatively, snore files, comprising the inspiratory sounds of the first 100 snores with the patient sleeping in a supine position, were also created.

Main outcome measures: Snore duration (s), loudness (dBA), periodicity (%) and energy ratios for the frequency sub-bands 0-200, 0-250 and 0-400 Hz.

Results: Snore loudness increased significantly (P < 0.0001), whilst energy ratios for frequency bands 0-200, 0-250 and 0-400 Hz all decreased significantly as sedation level increased (P < 0.001). A significant difference between natural snoring and snoring induced at the lowest sedation level was shown (P < 0.0001). Endoscopic examination was not tolerated at this sedation level.

Conclusions: The acoustic characteristics of sedation-induced and natural snores are sufficiently different to recommend the need for further research to determine whether the technique of sleep nasendoscopy is, in fact, a valid predictor of outcome of snoring surgery.