ASPO099 - Comparison of Genetic Diagnosis Rate Based on Age of Identification and Progression of Sensorineural Hearing Loss
Sunday, May 1, 2022
11:20 AM – 12:00 PM CT
Location: Landmark B
Elizabeth Liao, BA1, Michelle Florentine, BS2, Emily Taketa, BS1, Noura I. Mohamad, BS1, Dylan Chan, MD PhD1;
1Otolaryngology-Head & Neck Surgery, Univ. of California, San Francisco (UCSF), San Francisco, CA, 2Sackler Sch. of Med., Tel Aviv Univ., Tel Aviv, Israel.
Introduction: Sensorineural hearing loss (SNHL) affects 1:500 newborns and 1:300 children by age 9. Clinical guidelines recommend CMV testing, imaging, and genetic testing for diagnostic workup of a child with SNHL. We sought to understand how the rate of genetic diagnosis compares between SNHL that was early-identified (identified via newborn hearing screen) versus late-identified (identified after the neonatal period), as well as between stable and progressive SNHL.
Methods: We performed a retrospective cohort study on patients 0-18 years of age with SNHL with at least two audiograms, who completed hearing-loss gene panel testing (GeneDx) before April 3, 2021. Children were early-identified if they failed their newborn hearing screen (NHS), and late-identified if they passed or did not undergo NHS. Progression was defined as > 15 dB increase in pure tone average between the baseline and most recent audiogram or if progression was noted by audiology. Predictors of genetic diagnosis were assessed with univariate and multivariate logistic regression.
Results: 330 children were included. 247 children (75%) were underrepresented minorities (URM). 124 (37.5%) children were early-identified, and 206 (62.4%) late-identified. 30 (9%) children had progressive SNHL. On univariate analysis, genetic diagnosis was less likely among children who were late-identified (re: early-identified; OR: 0.53, p<0.014); had unilateral SNHL (re: bilateral; OR: 0.11, p<0.001); or were URM (re: not URM; OR=0.32, p<0.01). Progression (p=0.92) and severity of HL (p=0.70) were not associated with genetic diagnosis. On multivariate logistic regression, URM and late-identified children had lower odds of receiving a genetic diagnosis, while laterality, progression, and severity of HL did not. There was no association between URM and age at identification.
Conclusion: Children who are late-identified, URM, and/or have unilateral HL are less likely to receive a genetic diagnosis.