PRACTICE RECOMMENDATIONS
In young children with persistent otitis media with middle-ear effusions, the insertion of tympanostomy tubes does not improve cognitive, language, or speech development.
Q&A
Tubes for otitis media do not improve developmental outcomes
- BACKGROUND: Ear infections are common occurrences in pediatric patients. Children who suffer from frequent bouts of these infections may have developmental delays due to hearing loss associated with the effusion. The insertion of tympanostomy tubesin these children after 3 months of persistenteffusion is currently recommended in hopes of preventing such complications. However, there are few data to show these procedures are necessary.
- POPULATION STUDIED: The investigators enrolled 6350 healthy infants aged between 2 and 61 days from urban and rural sites. The children were monitored at least monthly for the first 3 years of life for the presence of middle-ear effusions using otoscopy, supplemented with tympanometry.
- STUDY DESIGN AND VALIDITY: The investigators conducted a randomized, double-blinded, controlled trial with many strengths. It was likely large enough to find a clinically significant difference if one existed. Throughout the trial, study investigators were regularly assessed for interobserver variability. A series of standardized tests and questionnaires, validated instruments for evaluating pediatric development, were used after the hearing tests.
- OUTCOMES MEASURED: The primary endpoint of the trial was developmental progress (cognitive, language, speech, and psychosocial development) determined by a series of tests and parental questionnaires. The tests were the General Cognitive Index of McCarthy Scales of Children’s Abilities, the Peabody Picture Vocabulary Test–Revised, the Nonword Repetition Test, the Number of Different Words, the Percentage of Consonants Correct–Revised, the Parenting Stress Index–Short Form, the Mean Length of Utterance in Morphemes, and the Child Behavior Checklist. All 3 groups received the same testing.
- RESULTS: No significant differences were found in developmental tests between the early- and late-treatment groups. The one exception was a small but significant difference on the Nonword Repetition Test, which favored the late-treatment group (66.3 vs 69.7; 95% confidence interval [CI], –6.2 to –0.7; P=.01). Under no circumstances was early treatment favored in the subgroup analyses.