Abstract: Background：The clinical phenotypes of Williams syndrome (WS) include special facial features, cardiovascular diseases, connective tissue abnormalities, growth retardation, mental retardation, often accompanied with hearing loss. Previous reports on tympanogram in children with WS are rare and the sample size is small, and none was reported in China. Objective：To analyze the tympanogram characteristics of children with WS, and to provide clinical evidence for the follow-up of hearing in these children. Design：Case-control study. Methods：We included consecutive children followed by two WS follow-up physicians at the Department of Pediatric Health care of the Children's Hospital of Zhejiang University School of Medicine from June 2019 to June 2023 who underwent tympanometric testing and were diagnosed with WS by genetic testing or Lowery score. The control group consisted of healthy children undergoing physical examinations by the same two follow-up physicians at the same department during the same period. Tympanometry was performed using the Titan tympanometry module from Interacoustics in a standard soundproof room with background noise levels <30 dB SPL. The examiner was an audiologist with over 5 years of clinical experience after obtaining an undergraduate degree in audiology. The first tympanogram was analyzed, with type A and type As tympanograms considered negative results, and type B and type C tympanograms considered positive results. Main outcome measures：Tympanogram characteristics in children with WS. Results：The WS group included 83 cases (165 ears), with 50 males (60.2%) and 33 females, aged 1.0 to 7.8 years (4.3±1.8 years). The control group included 99 cases (198 ears), with 64 males (64.6%) and 35 females, aged 0.6 to 11.8 years (4.0±2.2 years). The proportion of abnormal ears was higher in the WS group (48/165, 29.1%) than that in the control group (25/198, 12.6%), with a statistically significant difference. The differences in the number of abnormal ears between the WS and control groups were significant in both the -5 years and ≥6 years subgroups. The positive rate of tympanograms was significantly higher in the -5 years (30/84, 35.7%) and ≥6 years (11/32, 34.4%) subgroups compared to the <3 years subgroup (7/49, 14.3%) within the WS group, while there were no significant differences in the positive rate among different age subgroups within the control group. There were no significant differences in the composition ratio of tympanogram types among different age subgroups in the WS group. The control group showed significant differences in the composition ratio of type A and type As tympanograms across age subgroups. The ear canal volume and tympanometric gradient were both significantly greater in the WS group compared to the control group, whereas compliance and peak pressure showed no significant differences. Peak pressure was significantly lower in the -5 years and ≥6 years subgroups compared to the <3 years subgroup within the WS group. In a rescreening of 20 cases (40 ears), 21 ears showed no change in test results, 7 ears changed from positive to negative, and 12 ears changed from negative to positive. Conclusion：Children with WS have a higher positive rate of tympanograms that may occur repeatedly. The abnormal rate increases with age and negative pressure intensifies with age. The composition ratio of tympanogram is relatively stable among different age groups, which is different from that of control group. Regular long-term tympanogram follow-up is recommended for children with WS.

Key words: Williams syndrome, Children, tympanogram