A Common Virus, a Routine Heel Prick, and 276 Babies Who Would Otherwise Have Been Missed

What New York State learned by testing every newborn for cytomegalovirus for a year — and what it still cannot tell families

New York State spent a year adding one extra test to the routine heel-prick blood card that every newborn already receives, looking for a common virus called cytomegalovirus. Among just over 208,000 babies, the test found 276 who had been infected before birth — and more than half of the babies who turned out to be visibly affected had not been recognised by their doctors until the screening result arrived. The study shows that screening every newborn is practical and that families accept it readily, but also that the blood-spot test misses cases and that the long-term value of finding healthy-looking infected babies is not yet known.

The study, led by Norma Tavakoli and colleagues at the New York State Department of Health and published in JAMA Network Open in January 2026, is the largest American attempt so far to find out whether testing every newborn for this virus can be made to work in practice ^[1].

The Problem That Made This Research Necessary

Cytomegalovirus, usually shortened to CMV, is a member of the herpes virus family. Most adults who catch it never know they had it; it produces at worst a mild flu-like illness and then settles quietly in the body for life. For almost everyone, this is a non-event.

The exception is pregnancy. If the virus is passed to a developing baby, it can damage the inner ear and the brain. Careful studies from countries that screen every newborn suggest that roughly one baby in 200 is born infected in wealthy countries, and closer to one in 70 in lower-income countries ^[2]. About one in ten of those infected babies has visible signs at birth — things like an unusually small head, a low platelet count, jaundice, or an enlarged liver and spleen — and among that group, perhaps four or five in ten will go on to have lasting problems. Just as important, one in seven or eight babies who look completely healthy at birth will later develop difficulties, most commonly hearing loss that appears in the first years of life ^[3]. CMV is, in fact, the leading non-genetic cause of hearing loss in children. How common the virus is among mothers varies enormously — somewhere between 45% and 100% of women of childbearing age have been infected at some point, depending on where in the world they live and their circumstances ^[4] — which is part of why the risk differs so much from one community to another.

For decades, families in this situation faced a frustrating sequence. A baby seemed fine. The baby went home. Months or years later, hearing loss emerged, and only then — sometimes never — did anyone work out that a virus had been responsible. By that point the opportunity to do anything about it had passed, because the treatments that exist only work if they are started in the first weeks of life. That fact was established by two carefully conducted trials: one showed that six weeks of an intravenous antiviral drug protected hearing in babies with signs of brain involvement ^[5], and a later one showed that six months of a liquid antiviral medicine given by mouth was modestly better than six weeks for hearing and development ^[6]. Effective treatment with a narrow window makes early detection the whole ballgame.

The obvious compromise was to test only babies who did not pass their newborn hearing screen. Several US states require this. It helps, but a large study found this approach missed 43% of the babies whose hearing loss at birth was caused by CMV, and it cannot possibly find the healthy-looking babies whose hearing will fade later ^[7]. To find everyone, you would have to test everyone — using a sample every baby already gives. That sample is the dried blood spot, the few drops collected from a newborn's heel and blotted onto a card. The catch has always been that the virus is present in blood at low levels, and a landmark study found the blood-spot test detected only about a third of infected babies ^[8]. A saliva swab worked far better in the same study ^[9], but state laboratories are set up to receive blood cards, not saliva. The question New York asked was whether newer versions of the blood-spot test had improved enough to build a programme on.

What the Researchers Did

From October 2023 to September 2024, the New York State Newborn Screening Program added a CMV test to every suitable blood card it received ^[1]. Parents could decline — a brochure explaining how was produced in 17 languages — and only 245 out of 208,322 families, about one in a thousand, chose to opt out. That is a striking level of acceptance.

Babies whose card tested positive were referred to one of 11 hospitals across the state where paediatric infectious disease specialists took over. Those specialists confirmed the result with a urine test, which is the most reliable way to detect the virus, and then examined the baby: a physical check, blood tests, an ultrasound scan of the brain, and a hearing assessment. Doctors sorted each confirmed case into one of three groups using an agreed international definition ^[10] — babies with visible signs of illness, babies with hearing loss and nothing else, and babies who appeared entirely healthy.

One detail matters more than it might seem. A baby can also catch CMV after birth, from breast milk or ordinary contact, and in a healthy full-term baby that is generally harmless. The only way to tell the two apart is timing: if the positive sample was collected in the first 21 days of life, the infection almost certainly began before birth.

What They Found

The test flagged 529 babies out of roughly 208,000 — about one in 393. After specialists evaluated them, 276 were confirmed to have been infected before birth, which works out to about one baby in every 738. Of those 276, sixty-eight had visible signs of illness, one hundred and ninety-seven appeared completely healthy, and eleven had hearing loss as their only finding. Forty-eight of the sixty-eight affected babies were started on antiviral treatment, as were four of the eleven with hearing loss alone. One baby, born slightly early, died at a week of age from severe disease caused by the virus.

The finding that stands out is about recognition. Of the sixty-eight babies who did have visible signs, thirty-nine — well over half, including the baby who died — had not been identified as CMV cases by their doctors before the screening result came back. Looking at what those signs actually were explains it. The most common finding by far was an abnormality on the brain ultrasound, present in forty-six of the sixty-eight babies, and twenty-six of those babies had no other sign at all. Small head size, low platelets, and low white cell counts each appeared in ten babies. Only two babies had the pinprick rash that textbooks describe. In other words, most affected babies did not look ill. They looked like ordinary newborns whose brain scans, had anyone thought to order one, would have told a different story.

The screening also had real limitations. The programme learned of twenty-five babies whose blood-spot test came back negative even though they were infected — and because these were only discovered when doctors happened to report back, the true number is certainly higher. In sixteen of those cases, simply running the test again on the very same card produced a positive result, which tells you how borderline the amounts of virus in blood can be. There were also seventeen false alarms, some traced to contamination of the cards themselves.

Finally, a quarter of all the flagged babies — 131 of 529 — turned out to have caught CMV after birth rather than before it. Most of them were babies who had spent time in intensive care, where the state's rules require extra blood cards to be collected later on. For those families, a positive screen meant testing and worry about an infection that was, in most cases, not the one the programme was looking for.

What This Means for Families

If your baby is born in a place that does not screen for CMV, which is still most places, the practical lesson from this study is one for clinicians rather than parents: doctors should think of CMV sooner, particularly when a newborn has an unexplained low platelet count, a small head, poor growth, or an unexpected finding on a brain scan. It is reasonable to ask about CMV testing if any of those apply to your baby.

If your baby is screened and the result is positive, the most likely outcome — by a wide margin — is that your baby is entirely healthy and will need hearing checks over the coming years rather than any treatment. That is not a small thing to be told, and it is worth knowing in advance that "positive" here usually means "we will keep an eye on this," not "your baby is ill." Treatment is currently recommended only for babies with meaningful signs of disease, and sometimes considered for babies with hearing loss alone; there is not yet good evidence that treating healthy-looking infected babies helps them.

If your baby's screen is negative but something still seems wrong, that negative result should not close the door. This study makes plain that the blood-spot test misses cases, and a urine test remains available and more reliable.

What Researchers Are Working On Next

Three things are unresolved. The first is the test itself: nobody yet knows exactly how many babies the blood-spot method misses, because answering that would mean doing urine tests on every baby who screened negative. The second is the after-birth infection problem — screening programmes need clearer rules so that samples taken weeks after birth do not generate referrals the programme never intended to make.

The third, and the one that matters most to families, is what finding a healthy infected baby actually achieves. Ontario, in Canada, ran a similar programme and found exactly the same infection rate as New York, yet classified a noticeably smaller share of babies as affected ^[11] — largely because the two programmes drew the line between "healthy" and "affected" in different places. A review of practice worldwide found that this line is drawn differently almost everywhere ^[12]. Until researchers follow these children for years, nobody can say with confidence whether early knowledge leads to better hearing, better development, and better lives — or mainly to more appointments and more worry. A large follow-up study of the New York families, funded by the US National Institutes of Health, has been set up to answer precisely that, and it will ask the children and their families about their own experience, not just their test results.

What the study has already settled is narrower but not trivial. Waiting for a baby to look unwell is not a reliable way to find congenital CMV, and in New York it missed more than half of the babies who needed finding.

References

1. Tavakoli NP, Sack V, Handel AS, et al. Universal Newborn Screening for Congenital Cytomegalovirus Using Dried Blood Spot Specimens. JAMA Network Open. 2026;9(1):e2554518. doi:10.1001/jamanetworkopen.2025.54518 ↩
2. Ssentongo P, Hehnly C, Birungi P, et al. Congenital Cytomegalovirus Infection Burden and Epidemiologic Risk Factors in Countries With Universal Screening: A Systematic Review and Meta-analysis. JAMA Network Open. 2021;4(8):e2120736. doi:10.1001/jamanetworkopen.2021.20736 ↩
3. Dollard SC, Grosse SD, Ross DS. New estimates of the prevalence of neurological and sensory sequelae and mortality associated with congenital cytomegalovirus infection. Reviews in Medical Virology. 2007;17(5):355–363. doi:10.1002/rmv.544 ↩
4. Cannon MJ, Schmid DS, Hyde TB. Review of cytomegalovirus seroprevalence and demographic characteristics associated with infection. Reviews in Medical Virology. 2010;20(4):202–213. doi:10.1002/rmv.655 ↩
5. Kimberlin DW, Lin CY, Sánchez PJ, et al. Effect of ganciclovir therapy on hearing in symptomatic congenital cytomegalovirus disease involving the central nervous system: a randomized, controlled trial. The Journal of Pediatrics. 2003;143(1):16–25. doi:10.1016/S0022-3476(03)00192-600192-6) ↩
6. Kimberlin DW, Jester PM, Sánchez PJ, et al. Valganciclovir for symptomatic congenital cytomegalovirus disease. New England Journal of Medicine. 2015;372(10):933–943. doi:10.1056/NEJMoa1404599 ↩
7. Fowler KB, McCollister FP, Sabo DL, et al. A targeted approach for congenital cytomegalovirus screening within newborn hearing screening. Pediatrics. 2017;139(2):e20162128. doi:10.1542/peds.2016-2128 ↩
8. Boppana SB, Ross SA, Novak Z, et al. Dried blood spot real-time polymerase chain reaction assays to screen newborns for congenital cytomegalovirus infection. JAMA. 2010;303(14):1375–1382. doi:10.1001/jama.2010.423 ↩
9. Boppana SB, Ross SA, Shimamura M, et al. Saliva polymerase-chain-reaction assay for cytomegalovirus screening in newborns. New England Journal of Medicine. 2011;364(22):2111–2118. doi:10.1056/NEJMoa1006561 ↩
10. Rawlinson WD, Boppana SB, Fowler KB, et al. Congenital cytomegalovirus infection in pregnancy and the neonate: consensus recommendations for prevention, diagnosis, and therapy. The Lancet Infectious Diseases. 2017;17(6):e177–e188. doi:10.1016/S1473-3099(17)30143-330143-3) ↩
11. Dunn JKE, et al. Outcomes of a Population-Based Congenital Cytomegalovirus Screening Program. JAMA Pediatrics. 2025;179(3):332–339. doi:10.1001/jamapediatrics.2024.5562 ↩
12. Boscarino G, Romano R, Tegoni F, et al. Congenital cytomegalovirus severity definitions and treatment decisions around the world: a systematic scoping review of the literature. Journal of Clinical Medicine. 2024;13(19):5997. doi:10.3390/jcm13195997 ↩