Current Screening for Congenital CMV: Who Are We Missing?

My name is Dr Jeff Bergelson. I'm one of the infectious disease doctors at The Children's Hospital of Philadelphia and I'm a professor of pediatrics at the University of Pennsylvania Medical School.

I have been asked to talk about screening for congenital cytomegalovirus (CMV). Who should be screened, who do we screen now, and who might we screen in the future?

Given that there are 4 million babies born annually in the United States, with an incidence of just a little less than 1%, there are probably 30,000 babies infected with CMV born each year.

To give you an idea of the magnitude of the problem, about 1% of babies are born with CMV infection, which is diagnosed by a positive urine culture or polymerase chain reaction (PCR) in the first few weeks of life. Given that there are 4 million babies born annually in the United States, with an incidence of just a little less than 1%, there are probably 30,000 babies infected with CMV born each year. The potential consequences of CMV infection include:

long-term developmental problems, particularly long-term hearing loss;
Loss of vision; and
Loss of neurologic functions.

The risk of going on to have problems really depends largely on whether symptoms are recognized at birth or the infant is asymptomatic at birth.

About 10% of infants born with CMV infection will show clinical evidence at birth, and about 90% will not. The signs of infection at birth include microcephaly, hepatomegaly, hepatitis, jaundice, petechiae associated with thrombocytopenia, sensorineural hearing loss, or chorioretinitis.

Among babies who show those signs at birth, the majority will go on to have long-term hearing problems or cognitive or neurologic problems. That's probably about 3000 babies a year. Among the 90% of babies who are infected and asymptomatic (about 27,000 babies a year), about 10% will go on to have long-term problems, largely sensorineural hearing loss.

Current Screening Strategies

The screening issue concerns identifying babies who are likely to go on to have problems and are likely to benefit from some intervention, either drug treatment or stepped-up monitoring, hearing aids, and special language developmental support.

I've mentioned the symptoms and how we diagnose this infection. What we're doing right now is testing symptomatic babies for CMV infection. If they are found to have the infection, we evaluate them for all of the possible symptoms and signs. We do head ultrasound to look for intracranial calcifications and brain abnormalities. We do basic lab tests to look for hepatitis or thrombocytopenia. We have retinal exams to look for retinitis and hearing tests to look for sensorineural hearing loss. Then these babies are followed intensely for hearing loss and developmental problems so that they can have appropriate interventions.

Babies who are affected, especially those with neurologic problems evident at birth, can benefit from a 6-month treatment with valganciclovir, with improved outcomes in hearing and development at 2 years compared with babies who are not treated. There are side effects of valganciclovir treatment, particularly neutropenia. In animal studies it also shows the potential for long-term potentiation of tumors or infertility problems but we don’t have the evidence that this happens in people. Right now, we think that if an infant is severely affected by congenital CMV, the child is likely to benefit from valganciclovir treatment.

The Infants Who Are Missed

The way we are screening now, we are missing probably half of the babies who will go on to have developmental problems, especially hearing problems. Those babies are in the group of babies who are asymptomatic at birth. Remember, 10% of all infected babies will be symptomatic at birth while 90% will be asymptomatic. Among the asymptomatic babies, about 10% will go on to have hearing loss. Unfortunately, you cannot detect all of them by hearing tests at the time of birth. There has been the suggestion that we have universal screening of babies for CMV infection so that we can identify asymptomatic babies who would go on to develop problems and could benefit from close monitoring and intervention.

The way we are screening now, we are missing probably half of the babies who will go on to have developmental problems, especially hearing problems.

The upside of a universal screening strategy is that we could recognize these hearing problems earlier and potentially find and help babies for whom drug treatment would be appropriate. The real problem is that 80% of all infected babies, and 90% of asymptomatic infants, won't have any problems at all in the future, so we will wind up closely monitoring a large number of babies who will not benefit from this monitoring.

The downsides of this universal strategy are the cost, which may be a wash if you consider the benefits to babies and society at large of identifying infants for whom you can prevent deafness or language delay. There is also the cost of worry in the parents of a baby who is asymptomatic. You could be quite worried if you are told that your child is potentially vulnerable and may go on to have developmental problems and hearing loss but probably will not. The other downside is that people might be pushed to use drug treatment even when we don't know that asymptomatic babies benefit from drug treatment. You could get the side effects without any of the benefits.

To summarize, congenital CMV is a big problem whose major concerns are hearing loss and developmental delay. There are about 6000 babies [in total] born in this country each year who will go on to have symptoms at some time in their life. Right now, we are picking out the babies who are at highest risk, who are symptomatic at birth, but we are missing half of the babies who could potentially benefit from intervention. If we screen as a general approach, we will test millions of babies a year and identify thousands who have no potential for benefit. But we will also find some who do have the potential to benefit from treatments.

And that's where the debate stands right now. In the suggested reading list below, I have provided a couple of important references so that you can look at the evidence and review some of the things I have discussed.

Suggested Reading

Cannon MJ, Griffiths PD, Aston V, Rawlinson WD. Universal newborn screening for congenital CMV infection: what is the evidence of potential benefit? Rev Med Virol. 2014;24:291-307. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4494732/ Accessed November 30, 2016.

Demmler-Harrison GJ. Congenital cytomegalovirus infection: the elephant in our living room. JAMA Pediatr. 2016 Oct 10. doi: 10.1001/jamapediatrics.2016.2892. [Epub ahead of print]

James SH, Kimberlin DW. Advances in the prevention and treatment of congenital cytomegalovirus infection. Curr Opin Pediatr. 2016;28:81-85.

Kimberlin DW, Jester PM, Sánchez PJ, et al; National Institute of Allergy and Infectious Diseases Collaborative Antiviral Study Group. Valganciclovir for symptomatic congenital cytomegalovirus disease. N Engl J Med. 2015;372:933-943. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4401811/ Accessed November 30, 2016.