Because undetected hearing loss can have devastating consequences for child development, all 50 states have an early hearing detection and intervention program; 43 states, the District of Columbia, and Puerto Rico have mandatory newborn hearing screening. The effort seems to be paying off: Today, more than 95% of newborns born in the U.S. are screened for hearing loss shortly after birth, according to the National Institutes of Health (NIH).
Explore This Issue
July 2014Should we be doing more? Screening, while important, is only the first step in detecting and treating hearing loss, and NIH data suggest that as many as half of all infants who do not pass a newborn hearing screening are lost to follow-up. Furthermore, some infants who pass their hearing screening will go on to develop hearing loss in the early years of their lives.
“We know that at least two out of every thousand babies are born with a permanent hearing loss,” said Daniel Choo, MD, director of pediatric otolaryngology-head and neck surgery at Cincinnati Children’s Hospital Medical Center in Ohio. “By 6 years of age, six per thousand kids have hearing loss.”
According to the American Academy of Pediatrics, congenital cytomegalovirus (CMV) infection is the most common non-genetic cause of sensorineural hearing loss (SNHL) in children. Importantly, congenital CMV infection has also been proven treatable.
Utah is the only state that has passed a CMV public education and testing law, which went into effect in July 2013. The law, which states that infants who fail two newborn hearing screenings should be tested for congenital CMV before 21 days of age, also requires medical practitioners to talk to parents about the potential sequelae of CMV infection and inform them of available treatment options.
Should other states follow Utah’s lead? Is it time to add CMV testing to our arsenal of hearing screening and preservation procedures?
The CMV–SNHL Link
Congenital CMV infection can cause a host of health problems, or none. Symptoms may be apparent at birth, but most commonly are not. According to the NIH, only about one in 10 infants with a congenital CMV infection has symptoms at birth; symptoms may include retinal inflammation, jaundice, microcephaly, liver and spleen enlargement, and rash.
Approximately 90% of babies born with symptomatic congenital CMV will develop neurologic abnormalities, vision loss, and/or hearing loss. Ten percent of infants with asymptomatic congenital CMV will go on to develop neurologic abnormalities, vision loss, and/or hearing loss, which means that some newborns who appear perfectly healthy will develop CMV-associated hearing loss.
However, it’s difficult to determine exactly how many, primarily because CMV-associated hearing loss often is not apparent until a child is 2 years of age or older, and at that point, it’s next to impossible to determine whether CMV infection was present at birth. Because CMV is so endemic in the population, CMV testing must occur before three weeks of age to accurately determine if the infection was congenital.
A 2007 review of more than 11 studies determined that 15% to 20% of cases of profound bilateral hearing loss are associated with congenital CMV infection (J Clin Virol. 2008;41:57-62). A 10-year prospective study of more than 14,000 infants found a 0.53% prevalence of congenital CMV infection; of those, 22% developed SNHL (J Pediatr. 2008;153:84-88). Albert Park, MD, section chief for pediatric otolaryngology at the University of Utah in Salt Lake City, recently conducted a soon-to-be-published five-year retrospective study that concluded that “approximately 30% of the children we saw for pediatric hearing loss showed either confirmed or probable diagnosis for CMV.”
CMV-associated hearing loss is progressive and can fluctuate. In a 1997 study of children with congenital CMV, 50% of infants with demonstrated hearing loss at birth experienced further deterioration of hearing, with the median age of first progression 18 months (J Pediatr. 1997;130:624-630). Delayed-onset SNHL occurred in 18.2% of the children, and the median age of detection was 27 months.
CMV Testing Can Direct Monitoring and Treatment
Currently, practitioners in Utah, Ohio, parts of Massachusetts, and elsewhere are using CMV testing to evaluate and direct the treatment of newborns who fail to pass their newborn hearing screening. “In Utah, the testing requirement is based on the failure of two hearing screenings,” said Stephanie Browning McVicar, AuD, CCC-A, director of Utah’s Early Hearing Detection and Intervention program in Salt Lake City. An infant who fails his in-hospital newborn hearing screen will be re-screened at 10 to 14 days of age, often at the hospital of birth. Children who fail that second screening are scheduled for diagnostic auditory brainstem response testing and the child’s pediatrician is notified. By law, CMV testing must be done before the child reaches 21 days of age.
Infants who test positive for congenital CMV are referred to the University of Utah, where medical providers educate the family about CMV, including possible sequelae, monitoring, and treatment options, as well as ways to prevent the spread of the infection.
Education, an important part of the Utah law, may help prevent CMV-associated cases of SNHL in the future, because a child with CMV can shed the virus for years. Teaching families and childcare providers how to prevent CMV transmission (good hand washing, toy sanitation, not sharing utensils) may decrease future congenital CMV infections and CMV-related SNHL, said Dr. Park.
Treatment with antiviral medication can preserve, slow, or even reverse CMV-related SNHL (J Clin Virol. 2012;55:72-74; J Nippon Med Sch. 2012;79:471-477; J Pediatr. 2003;143:16-25.) “In some cases, we can actually bring back the hearing,” Dr. Choo said.
Drs. Choo and Park both treat patients with valganciclovir, an oral antiviral medication. (Valganciclovir is not FDA-approved for the treatment of congenital CMV; using it to treat CMV and hearing loss is an off-label use.) To be effective, though, the drug must be administered as early as possible. “If you catch the hearing loss within the first several weeks of onset, you have a much better chance of rescuing the hearing,” Dr. Choo said. “If you are several months into the hearing loss, it starts to get grey as to whether you’re too late, and if you’re a year or more out, most people would be fairly skeptical.”
Infants being treated with valganciclovir for congenital CMV-related hearing loss need frequent audiologic monitoring. “Typically, we confirm hearing loss, start them on valganciclovir, and then see them about every four to six weeks to repeat an auditory brainstem response test,” said Margaret Kenna, MD, MPH professor of otology and laryngology at Harvard Medical School in Boston. Children with CMV should also be referred to infectious disease, neurology, and ophthalmology specialists for further evaluation and monitoring.
Should CMV Testing Be Performed on All Infants?
Currently, Utah only performs CMV testing on children who have failed their hearing screenings. The problem with this approach is that about half of all babies with congenital CMV pass their screenings and show no signs of infection, said McVicar.
“It is estimated that up to 30% of the time when hearing loss develops within the first two years of life, the child passed their newborn hearing screen, lulling physicians into a false sense of security,” said Diego Preciado, MD, PhD, an otolaryngologist and director of the Cochlear Implant Program at Children’s National Medical Center in Washington, D.C. “If you’re going to screen for CMV, one should really consider screening every newborn, not just those with identified hearing loss. The issue is whether this approach would be considered cost-effective.”
“For any newborn screening program to be successful and embraced into practice, it has to be inexpensive,” said Mark Schleiss, MD, co-director of the Center of Infectious Diseases and Microbiology Translational Research at the University of Minnesota in Minneapolis. “The good news is that the technologies exist to use dried blood spot testing to quickly and inexpensively get an answer. The downside is that newborn blood spot screening may not the best way to screen for CMV.” Recent studies show that dried blood spot testing is cheap and convenient but not as accurate as testing urine or saliva (JAMA. 2010;303:1375-1382).
But CMV testing is cheaper than genetic testing and, given the prevalence of CMV-related SNHL, Dr. Park believes CMV testing is a cost-effective strategy. “We found that if you incorporate CMV testing first, instead of doing genetic testing or imaging first, it’s cost-effective in this population of children. CMV testing is actually the least expensive diagnostic test to detect causes of hearing loss than all the other tests that are currently being done,” he said.
Still, the lack of a standard treatment protocol causes some clinicians to wonder if CMV testing should be performed on all infants. “The treatment holds promise, but it’s nonstandard at this point,” said Dr. Preciado. “So, should CMV testing be legislated, especially when the legislation includes language that requires pediatricians to introduce a discussion about treatment if the test comes back positive? That’s certainly controversial.”
Testing all babies for congenital CMV would offer increased opportunity for early detection and potential treatment of hearing loss, but it would also certainly turn up babies who have congenital CMV but no adverse symptoms. And while many clinicians believe the risk-benefit ratio of valganciclovir treatment tilts in favor of treatment for infants with CMV-related hearing loss, treating asymptomatic, congenital CMV-positive infants is more problematic.
“Valganciclovir can result in adverse side effects. Short-term adverse effects include neutropenia and elevated transaminases; long-term effects may include increased risk for cancer and infertility, which have been reported in animals but not children. So if you have an asymptomatic patient and expose them to those side effects, the risk-benefit calculation doesn’t really work out,” Dr. Park said. A better approach, he said, would to be conduct frequent hearing tests and intervene if the child’s hearing begins to deteriorate.
The Future
The CMV and Hearing Multicenter Screening (CHIMES) study, currently in progress, is expected to provide some clarity regarding CMV testing and infant hearing. The seven-year study will screen 100,000 newborns for congenital CMV infection and correlate that information with the results of newborn screening programs. It will also follow children with congenital CMV for four years to determine the natural history of CMV-related hearing loss. Study results are expected to add clarity to future discussions on the utility of CMV testing in the effort to preserve hearing.
More effective, less toxic treatments for congenital CMV may also be on the horizon. Dr. Choo is working on developing antiviral eardrops to treat CMV-related SNHL. “We’ve shown intratympanic antiviral therapy to be effective in guinea pig models,” he said. “Proof of principle is there.”
The search for a safe, effective CMV vaccine also continues. “I think in the not-too-distant future, probably in the next decade, we’ll be seeing some vaccines approach licensure,” Dr. Schleiss said. “But, at the end of the day, even if we have a vaccine that’s licensed, it’s not the end of the story. CMV is a tricky virus.”
While the research continues, early consideration of congenital CMV as a possible etiology of SNHL may help decrease the incidence of SNHL and improve lives. “If you have a child with mild hearing loss and can identify and treat this child early, you may be able to prevent them from developing progressive loss,” Dr. Park said.
Jennifer L.W. Fink is a freelance medical writer based in Wisconsin.