Daily Tympanometry for High‐Resolution Measurement of the Time between Onset Of Cold‐Like Illness and Middle Ear Effusion

Abstract

Tympanometry is a simple method to assess middle ear pressure (MEP) and the presence of middle ear effusion (MEE), a marker of otitis media (OM). To determine whether daily parental tympanometry and illness sign recording in their children can be used to define the time between onsets of cold-like illness (CLI) and MEE at high resolution. Prospective, longitudinal, 7 month, daily follow-up on 169 children aged 1 through 8.6 years. Tympanograms and illness were recorded daily by a parent. Tympanograms were examined, rejected if artifactual, and MEP data were entered into a database, with "flat tympanograms" coded as -400 mm H2O = MEE. The incidence and burden of CLIs (>2 days) were calculated, and for each CLI, the presence/absence of concurrent MEE (>2 days) was determined. For each child, the average MEP for CLI and nonCLI days was calculated. Paired CLI and tympanometric results were aligned and the days between event onsets determined. Stepwise regression was used to assign risk predictors for the measured outcomes. A total of 566 CLIs were recorded, and the average CLI burden/child was 16%. Age was a significant predictor of CLI incidence/child, and age, history of colds, and daily environment were predictors of CLI-burden/child. Of the 433 evaluable CLI episodes, MEE was a complication in 37%, and MEE with a CLI was predicted by age, OM history, and environment. MEP was significantly more negative during CLI episodes, and the magnitude was predicted by age, race, and OM history. The average difference in MEE-CLI onsets was 1.2 +/- 4.0 days; approximately 32% of MEEs occurred prior to CLI onset and 17% on the same day as CLI onset. CLIs adversely affect the middle ear-ambient pressure balance and are frequently associated with MEE episodes. The distribution in onsets between those events suggests that chemoprophylaxis of a child with a newly identified CLI to prevent MEE would have a low expected efficiency.