Vital Signs: Zika-Associated Birth Defects and Neurodevelopmental Abnormalities Possibly Associated with Congenital Zika Virus Infection — U.S. Territories and Freely Associated States, 2018

1Zika virus infection during pregnancy can cause serious birth defects, including structural abnormalities of the brain and eye. As infants with congenital Zika virus infection get older, problems such as epilepsy, vision loss, and developmental delays have been increasingly recognized. Early identification of and intervention for adverse neurodevelopmental outcomes have been determined to improve cognitive, social, and behavioral functioning and to be cost effective to society in general.

2The most critical time to intervene and promote optimal brain development is during the first 3 years of life. To facilitate early identification and intervention, CDC released clinical guidance for the evaluation and management of infants with possible congenital Zika virus infection in January 2016. The guidance was based largely on existing guidelines for pediatric health promotion and care; expert opinion was incorporated from clinicians and researchers with knowledge of congenital infections and of clinical care of infants with birth defects as described in early reports. Recommendations for the care and management of infants with possible congenital Zika virus exposure and infants with one or more clinical findings consistent with congenital Zika virus syndrome have remained largely unchanged through subsequent updates. Standard evaluation* at birth and during each well-child visit is recommended for all infants and young children with possible prenatal Zika virus exposure. Laboratory testing for Zika virus is recommended for infants born to mothers with laboratory evidence of confirmed or possible Zika virus infection during pregnancy and for infants with one or more clinical findings consistent with congenital Zika syndrome born to mothers with possible Zika virus exposure, regardless of maternal testing results. In addition to a standard evaluation, infants born to mothers with laboratory evidence of confirmed or possible Zika virus infection during pregnancy should have a cranial ultrasound or other brain imaging and a comprehensive ophthalmologic evaluation performed by age 1 month to detect subclinical brain and eye findings.

3To better understand the effects of Zika virus infection during pregnancy on mothers and children from a national surveillance perspective, CDC collaborated with state, territorial, and local health departments on the U.S. Zika Pregnancy and Infant Registry (USZPIR)† to monitor pregnancy and infant/child outcomes among pregnancies with laboratory evidence of confirmed or possible Zika virus infection (www.cdc.gov/pregnancy/zika/research/registry.html). The USZPIR currently monitors outcomes of approximately 7,300 pregnancies, over 4,800 of which are reported from the U.S. territories and freely associated states§ (https://www.cdc.gov/pregnancy/zika/data/pregwomen-uscases.html). This report is the first to provide data on Zika-associated birth defects and neurodevelopmental abnormalities possibly associated with congenital Zika virus infection identified during infancy and early childhood among children aged ≥1 year who were born in the U.S. territories and freely associated states.§

4Pregnancies with laboratory evidence of confirmed or possible Zika virus infection¶ and infants resulting from these pregnancies are included in the USZPIR and followed through active surveillance methods. Data on birth defects and neurodevelopmental outcomes were abstracted from prenatal, birth hospitalization, pediatric, and specialty care medical records using standardized methods and reported to the USZPIR. CDC provided technical assistance to all U.S. territories and freely associated states that reported cases to the USZPIR through the Zika Local Health Department Initiative (https://www.cdc.gov/pregnancy/zika/research/lhdi.html) and the Epidemiology and Laboratory Capacity for Infectious Diseases Cooperative Agreement (https://www.cdc.gov/ncezid/dpei/epidemiology-laboratory-capacity.html). This report includes children who, among pregnancies reported to the USZPIR, 1) were born in U.S. territories or freely associated states; 2) had a date of birth on or before February 1, 2017, and reached age 1 year on or before February 1, 2018; and 3) had follow-up care reported to the USZPIR by June 1, 2018. For the purpose of this analysis, follow-up care was defined as clinical care at age >14 days reported to the USZPIR. Children from multiple gestation pregnancies were counted separately; infants who died during the first year of life were excluded.

5Among children who met the definition for reported follow-up care, the percentages who were reported to have received each of the following types of clinical care or evaluations, recommended in CDC clinical guidance, were calculated: 1) neuroimaging (cranial ultrasound, computed tomography, or magnetic resonance imaging) any time after birth; 2) hearing screen by automated auditory brainstem response (ABR) or audiologic evaluation by diagnostic ABR (ABR-based hearing screen or evaluation) any time after birth; 3) ophthalmologic evaluation any time after birth; 4) developmental screening or evaluation at age >14 days; and 5) physical examination, as indicated by reported growth parameters (head circumference, length, or weight) at age >14 days.

6Data were reviewed by clinical subject matter experts to identify Zika-associated birth defects and neurodevelopmental abnormalities possibly associated with congenital Zika virus infection (Box). Data for each child were reviewed by at least two reviewers; discrepant review findings were discussed among clinical subject matter experts to reach agreement. Although the category of neural tube defects and other early brain malformations was initially included in the surveillance case definition for Zika-associated birth defects, it was excluded in this report because of growing evidence suggesting a lack of association of these defects with congenital Zika virus infection. Postnatal-onset microcephaly detected during follow-up care is distinct from microcephaly detected at birth and is included among neurodevelopmental abnormalities possibly associated with congenital Zika virus infection (Box). Neurodevelopmental findings such as hearing loss, seizures, or swallowing abnormalities consistently documented in reports of infants with possible congenital Zika virus infection were specifically selected for inclusion; however, the broad range of neurodevelopmental abnormalities possibly associated with congenital Zika virus infection necessitates inclusion of less specific but more prevalent categories, such as possible developmental delay. The percentages of these adverse outcomes were calculated among all children born to mothers with laboratory evidence of confirmed or possible Zika virus infection during pregnancy with reported follow-up care, as well as among the subset of children born to mothers with nucleic acid testing (NAT)–confirmed infection during pregnancy, with reported follow-up care.

7A sensitivity analysis to address the concern about possible misclassification of microcephaly at birth** was performed by excluding infants with a birth head circumference measurement indicating microcephaly and no other Zika-associated birth defects, who subsequently had normal neuroimaging and at least two postnatal measurements with a head circumference above the tenth percentile for the infant’s age and sex. Among infants tested after birth for Zika virus with either NAT or serologic tests (immunoglobulin M [IgM]) in serum, urine, or cerebrospinal fluid, the percent positivity is reported.

8The U.S. territories and freely associated states reported 4,816 pregnancies with laboratory evidence of confirmed or possible Zika virus infection by June 1, 2018, including 4,320 (90%) completed on or before February 1, 2018; 4,165 (96%) pregnancies resulted in 4,199 live-born infants, and 155 (4%) resulted in a pregnancy loss (Figure 1). Seven infants were excluded who would have reached age 1 year on or before February 1, 2018 and were reported to have died, including three who died during the first 14 days of life. By February 1, 2018, a total of 2,141 (51%) children were aged ≥1 year, 1,450 (68%) of whom had some follow-up care reported to the USZPIR after age 14 days.

9Among these 1,450 children 1,376 (95%) had at least one physical examination reported after 14 days of life, 1,106 (76%) had at least one developmental screening or evaluation, 864 (60%) had postnatal neuroimaging, and 695 (48%) had at least one ABR-based hearing screen or evaluation. An ophthalmologic evaluation was reported for 522 (36%) children (Figure 2).

10Among all 1,450 children with reported follow-up care, 203 (14%) had a Zika-associated birth defect, neurodevelopmental abnormality possibly associated with congenital Zika virus infection identified, or both: 87 (6%) had at least one Zika-associated birth defect, 136 (9%) had at least one neurodevelopmental abnormality possibly associated with congenital Zika virus infection, and 20 (1%) had both (Table). Among the 1,386 (96%) children who did not have microcephaly detected at birth, 822 (59%) received neuroimaging, including 14 (2%) who had at least one brain anomaly identified. In addition, among the 494 (36%) children who received an ophthalmologic evaluation, 12 (2%) had at least one eye anomaly identified. Thus, had these infants not received neuroimaging or ophthalmologic evaluation, 26 brain or eye anomalies in 23 children might have gone undetected.

11The sensitivity analysis to assess possible misclassification of microcephaly at birth identified 84 (6%) children with microcephaly among the 1,450 children who had follow-up care reported: five infants had microcephaly at birth with brain or eye anomalies identified at birth; 59 had microcephaly at birth with no brain or eye anomalies identified at birth; and 20 infants did not have microcephaly identified at birth but had postnatal identification of microcephaly. The 59 infants with only microcephaly at birth included 15 who had no other Zika-associated birth defects identified during follow-up care, had normal neuroimaging, and had at least two postnatal measurements with a head circumference above the tenth percentile for the infant’s sex and age. Excluding these 15 infants from the 87 with Zika-associated birth defects results in a decrease in the estimated percentage of affected children from 6% to 5%.

12Among the 1,450 children whose mothers had laboratory evidence of confirmed or possible Zika virus infection during pregnancy and who had follow-up care reported, 136 (9%) had neurodevelopmental abnormalities possibly associated with congenital Zika virus infection identified (Table). One hundred sixteen (8%) had one or more neurodevelopmental abnormalities possibly associated with congenital Zika virus infection identified, but no Zika-associated birth defects; of these 116 children, 58 (50%) had only possible developmental delay identified, and 25 (22%) had possible developmental delay with at least one other neurodevelopmental abnormality possibly associated with congenital Zika virus infection identified.

13Among 943 pregnancies with NAT-confirmed Zika virus infection, 144 (15%) had a Zika-associated birth defect, neurodevelopmental abnormality possibly associated with congenital Zika virus infection identified, or both: 62 (7%) had at least one Zika-associated birth defect identified. Ninety-nine (10%) had at least one neurodevelopmental abnormality possibly associated with congenital Zika virus infection identified, and 17 (2%) had both.

14Among the 1,450 children in this analysis, 607 (42%) did not receive testing for Zika virus infection in serum, urine, or cerebrospinal fluid. Among the 843 (58%) who did receive testing, 32 (4%) tested positive by either NAT or IgM (four of 69 tested by NAT only; zero of 18 tested by IgM only; and 28 of 756 tested by both NAT and IgM tested positive by either NAT or IgM). Zika-associated birth defects or neurodevelopmental abnormalities possibly associated with congenital Zika virus infection were identified in children with positive Zika virus IgM or NAT, negative IgM and NAT, and in those who did not receive testing.

15A total of 1,450 children aged ≥1 year were born to mothers with laboratory evidence of confirmed or possible Zika virus infection during pregnancy in the U.S. territories and freely associated states and were reported to the USZPIR. Among these children, approximately one in seven (14%) were identified during infancy or early childhood as having either a Zika-associated birth defect, a neurodevelopmental abnormality possibly associated with congenital Zika virus infection, or both.

16The 6% with Zika-associated birth defects in this report can be viewed in the context of the previously published baseline frequency of brain and eye abnormalities potentially related to Zika virus infection. Before the introduction of Zika in the Region of the Americas the baseline frequency of brain and eye abnormalities potentially related to Zika virus infection among live-born infants was approximately 0.16%, suggesting a more than 30-fold increase over baseline.

17Among all children aged ≥1 year by February 1, 2018, 68% had some follow-up care reported to the USZPIR. Of these children, 95% had at least a physical examination, 76% had developmental screening or evaluation, and 60% had neuroimaging. Approximately one half of the children (48%) had an ABR-based hearing screen or evaluation, and approximately one third of the children (36%) had an ophthalmologic evaluation reported to the USZPIR. Because the full spectrum of adverse outcomes related to congenital Zika virus infection is not yet known, careful monitoring and evaluation of children born to mothers with laboratory evidence of confirmed or possible Zika virus infection during pregnancy is essential for ensuring early detection of possible disabilities and early referral to intervention services that might improve outcomes. For example, with early identification of vision problems, a prescription for corrective eyeglasses might be beneficial to a child’s development. Among children without microcephaly detected at birth, brain or eye anomalies might have gone undetected without neuroimaging or ophthalmologic evaluation.

18Many infants did not have Zika virus testing results reported. This could be because of changing recommendations for laboratory testing of infants born to mothers with laboratory evidence of confirmed or possible Zika virus infection during pregnancy. Among infants with testing reported, only 4% tested positive for Zika virus infection by IgM or NAT. In addition, limitations of laboratory testing for Zika virus have been previously described; Zika virus RNA is only transiently present in body fluids; thus, a negative NAT result does not rule out infection. Zika virus-associated birth defects and neurodevelopmental abnormalities possibly associated with congenital Zika virus infection also were identified in children with negative Zika virus NAT or IgM test results. These finding are consistent with other reports of infants with clinical findings suggestive of possible congenital Zika syndrome but with negative laboratory results.

19Microcephaly is challenging to monitor accurately as an outcome because it is difficult to reliably measure head circumference in a newborn, it can be affected by inaccuracies in estimated gestational age, and it does not distinguish between a small head size related to underlying pathology and one that will subsequently exhibit typical brain development. The sensitivity analysis suggests that the number of infants with Zika-associated birth defects could be a modest overestimate.

20This is the first analysis assessing neurodevelopmental abnormalities possibly associated with congenital Zika virus infection in addition to Zika-associated birth defects among children born to mothers in the U.S. territories and freely associated states with laboratory evidence of confirmed or possible Zika virus infection during pregnancy. Although there are large cohort studies monitoring pregnancies with and without Zika virus infection in several countries, the data in this report come from the largest cohort of children born to mothers with laboratory evidence of confirmed or possible Zika virus infection during pregnancy in the world who are currently being monitored as part of an enhanced surveillance system.

21Whereas the cohort size is a strength of this analysis, the findings in this report are subject to at least five limitations. First, the data are limited to evaluations and clinical care received and reported to the USZPIR. The recommended services might not have been available to all children, and among those with reported follow-up care, information was limited for some children. In addition, data are limited to clinical records reported to the USZPIR; collecting these data are challenging because children might be seen in various outpatient settings and by multiple providers. To alleviate this barrier, territorial and state jurisdictions made extensive efforts to actively follow up, abstract, and report available data; CDC also provided substantial technical assistance. Second, it is possible that children with recognized health problems might have received follow-up care more frequently than did those without identified health problems, which might lead to an overestimate of the percentage of children with Zika-related health problems. Third, estimates of the baseline frequencies of neurodevelopmental abnormalities among very young children are available only for a few of the specific abnormalities; the lack of an appropriate comparison group limits assessment of whether the prevalence of reported neurodevelopmental abnormalities in the U.S. territories and freely associated states among children born to mothers with laboratory evidence of confirmed or possible Zika virus infection during pregnancy is an increase over baseline levels. Fourth, given the potential persistence, cross-reactivity, or nonspecific reactivity of IgM, some mothers included in the USZPIR might not have been infected with Zika virus during their pregnancy. For this reason, an analysis of child outcomes restricted to pregnancies with NAT-confirmed Zika virus infection was included, and similar percentages of children with a Zika-associated birth defect, a neurodevelopmental abnormality possibly associated with congenital Zika virus infection, or both were found. Finally, it might be difficult to distinguish between birth defects and neurodevelopmental abnormalities that might be causally linked to congenital Zika virus infection and those that might be attributable to unrelated causes; thus, this report describes occurrences without attributing causation.

22Despite the limitations, this report extends understanding about the impact of congenital Zika virus infection. Whereas approximately 6% of children with congenital Zika virus exposure have Zika-associated birth defects, more children have neurodevelopmental abnormalities possibly associated with congenital Zika virus infection, identified during follow-up care, albeit without an appropriate comparison group on the baseline prevalence of these neurodevelopmental abnormalities among very young children. Given that most children did not have evidence of all recommended evaluations according to data reported to the USZPIR, additional unidentified anomalies might exist in this population. Furthermore, it is recognized that there were substantial disruptions to the provision of clinical care in Puerto Rico and the U.S. Virgin Islands related to hurricanes in 2017; many families also were internally displaced or left the affected territories, potentially resulting in fewer follow-up care data reported to the USZPIR. Children who were most affected by Zika virus infection during pregnancy might have been either more or less likely to be displaced after hurricanes; there is no specific information on the impact of this displacement in these estimates. However, jurisdictional staff members attempted to find families and link them to the USZPIR in their new jurisdiction.

23It is essential that health care providers who care for children have access to information regarding maternal exposure to Zika virus infection during pregnancy. This will improve the identification of children born to mothers with laboratory evidence of confirmed or possible Zika virus infection during pregnancy so that they can receive recommended postnatal evaluations. Zika virus transmission is far less prevalent in the Americas in 2018 than during 2015–2017 (https://www.cdc.gov/zika/reporting/case-counts.html); however, information about this cohort of children can inform and guide future responses to outbreaks of Zika virus that will inevitably occur among susceptible populations and disproportionately affect pregnant women and their children.

24Zika virus infection during pregnancy can cause serious birth defects and might be associated with neurodevelopmental abnormalities.

25Among children aged ≥1 year born in U.S. territories and freely associated states to mothers with laboratory evidence of confirmed or possible Zika virus infection during pregnancy and who had follow-up care reported, 6% had a Zika-associated birth defect, 9% had ≥1 neurodevelopmental abnormality possibly associated with congenital Zika virus infection, and 1% had both.

26Given the potential benefits from interventions during early critical periods of infant development, health care providers should share information on maternal Zika virus exposure and closely monitor child health and development.