Abstract Introduction and objective In the first 6 months of life, 226 Hz tympanometry is considered an ineffective procedure for the diagnosis of otitis media with effusion. With the introduction of universal hearing screening, the use of high frequency 1000 Hz (1 kHz) tympanometry has been recommended. To optimise the diagnosis of neonatal hearing loss, we present this comparison, from the clinical point of view, of the results of 226 Hz and 1 kHz tympanometry in infants. Materials and methods We designed a prospective study of 100 children under 9 months of age proceeding from our hearing screening programme. We compare the result of tympanometry with binocular microscopy and transient evoked otoacoustic emissions. Results The application of transient otoacoustic emissions, otomicroscopy and 226 Hz and 1 kHz tympanometry has shown its usefulness in the management of otitis media with effusion of young infants, with a similar effectiveness between the 4 tests. Conclusion The joint use of otomicroscopy, transient otoacoustic emissions and 226 Hz and 1 kHz tympanometry, has allowed us to diagnose otitis media with effusion in young infants more accurately than each test separately. We recommend initial use of 1 kHz tympanometry, at least in children younger than 7 months, but in the presence of hearing loss or an unclear result, 226 Hz tympanometry is a good diagnostic complement.