Evaluation of the diagnostic value of pneumatic otoscopy in primary care using the results of tympanometry as a reference standard.

The aim of this study was to determine the value of pneumatic otoscopy in diagnosing otitis media with effusion in primary care. Pneumatic otoscopy was carried out for 111 children aged one to 16 years and the results obtained compared with those obtained from tympanometry. The children were those who had attended for a regular ear, nose and throat check up in the health centre of a school for the deaf during the period November 1989 to January 1990. Pneumatic otoscopy and tympanometry with the GSI 28 instrument (Grason-Stadler) were carried out by a trained ear, nose and throat nurse. All relevant features determined using otoscopy--the colour, position and mobility of the tympanum--and an overall assessment were recorded. The results of tympanometry were evaluated independently of the otoscopic findings. In the population examined the predictive values of positive and negative results of pneumatic otoscopy for diagnosing effusion were high; the sensitivity was low. Serious retraction of the eardrum and absence of mobility under positive pressure were the most predictive features but the colour of the tympanum did not show any relation to effusion. In the youngest age group (one to five years) 56% of the children had abnormal otoscopic findings in either one or both ears (odds ratio for this group versus remaining children 3.75; 95% confidence interval (CI) 1.75 to 15.2). The odds ratios of past and present history of upper respiratory tract infection with respect to abnormal results from otoscopy were 2.41 (95% CI 1.05 to 5.53) and 2.95 (95% CI 1.17 to 7.45), respectively. Pneumatic otoscopy carried out by an experienced health care worker is of high diagnostic value when compared with the results of tympanometry. Pneumatic otoscopy can improve the diagnostic capabilities of general practitioners and other primary care workers with regard to otitis media with effusion.