Cytomegalovirus (CMV)


1) Congenital CMV infection is associated with sensorineural hearing loss with high rate of progression. CMV-associated SNHL can be both congenital and with onset reported up to 6 years of age.

2) Symptoms in baby related to CMV infection (which include hearing loss, neurologic deficits, and systemic effects) occur ONLY with prenatal transmission of CMV. CMV infection of pregnant mom is asymptomatic

3) Postnatal transmission of CMV (usually through breastmilk) is very common - 12% of all babies at 3 months are positive for CMV, and 20% at one year (Peckham et al., 1987) and is NOT associated with any symptoms for baby.

4) Identification of congenital CMV infection as a cause of congenital and early-childhood-onset SNHL is important for the following reasons:

1) knowledge of the cause of hearing loss

2) potential for valganciclovir treatment to prevent progression

3) screening for other symptoms associated with congenital CMV, especially neurologic

5) Congenital CMV infection can be detected with high specificity either by urine culture/PCR before 3 weeks of age, or by CMV PCR of dried blood spots after 3 weeks of age. Direct CMV testing prior to 3 weeks of age is always preferable, as DBS testing misses many cases; thus, there is urgency for neonates with suspected CMV infection. Urine culture is performed at UCSF-BCH Oakland and SF; Urine PCR by Quest laboratories; DBS CMV PCR by labs at UW and Utah. CPT code for this test is 87496.

Important studies

1) Peckham CS, Johnson C, Ades A, Pearl K, Chin KS (1987). Early acquisition of cytomegalovirus infection. Arch Dis Child 62:780-785. Describes natural history of postnatal acquisition of CMV infection.

2) Kimberlin et al (2003). Effect of ganciclovir therapy on hearing in symptomatic congenital cytomegalovirus disease involving the central nervous system: a randomized, controlled trial. J Pediatr 143:16-25. This randomized, controlled, blinded study showed that the IV antiviral drug ganciclovir prevented worsening of hearing loss in babies with symptomatic CMV infection (0% vs 42% for placebo). This drug, however, had significant side effects, especially neutropenia (weakening of the immune system (63% vs 21%)

3) Kimberlin et al (2015). Valganciclovir for symptomatic congenital cytomegalovirus disease. N Engl J Med 372:933-43. This randomized, controlled, blinded study showed that the oral antiviral drug valganciclovir prevented worsening of hearing loss in babies with symptomatic CMV infection (13% with worsening of hearing at one year with 6 wks treatment; 8% with 6 months treatment), with much less side effects than ganciclovir (27% neutropenia). This paper is serving as the basis for considering antiviral treatment in babies with CMV-associated hearing loss.

4) Boppana et al (2010). Dried blood spot real-time PCR assays to screen for congenital cytomegalovirus infection. JAMA 303:1375-1382. Established the ability of CMV DBS testing to confirm diagnosis of congenital CMV infection. Low sensitivity (28%) but nearly 100% specificity. Sensitivity likely variable depending on how DBS are stored, blood extracted, and tested.

5) Ross et al (2014). Detection of congenital cytomegalovirus infection by real-time PCR analysis of saliva or urine specimens. Establishes equivalence of CMV PCR testing of urine or saliva to gold standard test (urine culture).

6) CHIMES study. Presented at AAP 2014 (Susan Ross, UAB). They screened 100,000 babies for CMV and performed newborn hearing screening at birth. 2,000 babies referred NHS; 460 babies were positive for CMV. 33 babies were positive for CMV and referred NHS. Of these 33 babies, 21 were found to have SNHL at initial audiologic testing before 3 months.

7) NCT01649869. Multi-institutional clinical trial (Valgan toddler trial) headed by David Kimberlin (UAB) to investigate efficacy of valganciclovir treatment on hearing outcomes in children up to the age of 4 with confirmed congenital CMV infection by urine PCR before 3 weeks of age, or by CMV DBS PCR testing up to 4 years of age.

8) Utah and Connecticut now have state legislation mandating CMV testing prior to 3 weeks of age in all babies with referred NHS.

Practice guidelines

CMV testing can be ordered by otolaryngology; if audiology sees a baby fitting one of these criteria, please contact otolaryngology ([email protected]) and we can order the necessary tests. This is especially time-sensitive for neonates who refer on NHS - otolaryngology should be notified and testing done prior to 3 weeks of age.

1) Babies under 3 weeks of age with referred NHS or confirmed hearing loss should undergo CMV testing (either urine culture or urine/saliva PCR) prior to 3 weeks of age. Ideally, this would happen prior to leaving the birth hospital.

2) Babies over 3 weeks of age with referred NHS should not be tested for CMV, but should undergo standard diagnostic audiologic testing to confirm hearing loss

3) Babies and children over 3 weeks and up to 4 years of age with confirmed sensorineural hearing loss should consider CMV DBS testing. Determination of age cutoff is based on the inclusion criteria for the Valgan toddler study. A negative result does not rule out congenital CMV, but there is no other reliable testing for this age group. A positive result is highly specific for congenital CMV infection. Discussion should then be had with family regarding further workup/treatment - this may include referral to ID, consideration of valganciclovir, referral to neurology, and/or MRI.

4) Children 4 years of age and up with confirmed sensorineural hearing loss may consider CMV DBS testing, purely as a diagnostic etiologic test for the cause of the hearing loss. Valganciclovir treatment is not currently being considered in this age group.