HHIssues

Frm JAMA, the "Journal of the American Medical Association"

RESEARCH LETTERS

Genetic Testing to Identify Deaf Newborns

To the Editor: Mutations in the GJB2 gene are the most common cause of prelingual deafness, accounting for approximately half of all nonsyndromic recessive deafness in many world populations.1 In select groups, such as Ashkenazi Jews, this proportion is even higher.2 Two deafness-causing GJB2 mutations are particularly common: the 35delG mutation, with a carrier frequency of 2.5% in the midwestern United States1; and the 167delT mutation, with a carrier frequency of 4.76% in the Ashkenazi population.2 The prevalence of these mutations has led to tremendous interest in the clinical use of mutation screening to identify newborns with GJB2-related deafness.

Report of Cases

Recently, we evaluated 2 deaf children for GJB2 mutations; neither child had been previously identified as a deaf neonate through the use of conventional newborn-screening audiometry. The first child had
a normal result on a newborn auditory brainstem response (ABR) screening test but was diagnosed as deaf at age 15 months when a second ABR, obtained at his parents' request, showed no responses through 90 dB. At age 5 months, the second child had normal findings on a free-field audiogram performed by an experienced audiologist but 4 months later had ABR-confirmed severe hearing loss. Mutation screening by direct sequencing demonstrated that both children are
homozygous for the 35delG GJB2 mutation.

Comment

While these results may suggest that genetic testing should be added to audiometric screening to identify children with prelingual sensorineural hearing loss, we believe that such an approach is premature. The prior normal audiometric results for these children imply that they developed the deaf phenotype later in infancy, and this underscores the unpredictable relationship between genotype and phenotype. Because a negative neonatal audiogram does not necessarily rule out a deafness-causing GJB2 mutation, a screening program for genetic deafness would need to include the entire population of newborns. This type of program would be very expensive and only marginally add to the predictive power of conventional screening. Until large population-based research studies clarify the relationship between common and rare GJB2 allele variants and their effect on hearing, the role of genetic testing in identifying deaf children remains to be established.

Glenn E. Green, MD
Richard J. H. Smith, MD
Molecular Otolaryngology Research Laboratories
University of Iowa
Iowa City

John P. Bent, MD
New York Otolaryngology Institute
New York, NY

Edward S. Cohn, MD
Center for Ear, Hearing and Balance Disorders
Boys Town National Research Hospital
Omaha, Neb

** HHIssues **