ABSTRACT
BACKGROUND: In collaboration with state, tribal, local, and territorial health departments, CDC established the U.S. Zika Pregnancy Registry (USZPR) in early 2016 to monitor pregnant women with laboratory evidence of possible recent Zika virus infection and their infants. METHODS: This report includes an analysis of completed pregnancies (which include live births and pregnancy losses, regardless of gestational age) in the 50 U.S. states and the District of Columbia (DC) with laboratory evidence of possible recent Zika virus infection reported to the USZPR from January 15 to December 27, 2016. Birth defects potentially associated with Zika virus infection during pregnancy include brain abnormalities and/or microcephaly, eye abnormalities, other consequences of central nervous system dysfunction, and neural tube defects and other early brain malformations. RESULTS: During the analysis period, 1,297 pregnant women in 44 states were reported to the USZPR. Zika virus-associated birth defects were reported for 51 (5%) of the 972 fetuses/infants from completed pregnancies with laboratory evidence of possible recent Zika virus infection (95% confidence interval [CI] = 4%-7%); the proportion was higher when restricted to pregnancies with laboratory-confirmed Zika virus infection (24/250 completed pregnancies [10%, 95% CI = 7%-14%]). Birth defects were reported in 15% (95% CI = 8%-26%) of fetuses/infants of completed pregnancies with confirmed Zika virus infection in the first trimester. Among 895 liveborn infants from pregnancies with possible recent Zika virus infection, postnatal neuroimaging was reported for 221 (25%), and Zika virus testing of at least one infant specimen was reported for 585 (65%). CONCLUSIONS AND IMPLICATIONS FOR PUBLIC HEALTH PRACTICE: These findings highlight why pregnant women should avoid Zika virus exposure. Because the full clinical spectrum of congenital Zika virus infection is not yet known, all infants born to women with laboratory evidence of possible recent Zika virus infection during pregnancy should receive postnatal neuroimaging and Zika virus testing in addition to a comprehensive newborn physical exam and hearing screen. Identification and follow-up care of infants born to women with laboratory evidence of possible recent Zika virus infection during pregnancy and infants with possible congenital Zika virus infection can ensure that appropriate clinical services are available.
Subject(s)
Congenital Abnormalities/virology , Fetus/virology , Pregnancy Complications, Infectious/virology , Zika Virus Infection , Brain/abnormalities , Brain/virology , Central Nervous System Diseases/epidemiology , Central Nervous System Diseases/virology , Congenital Abnormalities/epidemiology , Eye Abnormalities/epidemiology , Eye Abnormalities/virology , Female , Humans , Infant , Infant, Newborn , Microcephaly/epidemiology , Microcephaly/virology , Neural Tube Defects/epidemiology , Neural Tube Defects/virology , Pregnancy , Registries , United States/epidemiology , Zika Virus/isolation & purification , Zika Virus Infection/epidemiologyABSTRACT
In 2016, Centers for Disease Control and Prevention (CDC) established surveillance of pregnant women with Zika virus infection and their infants in the U.S. states, territories, and freely associated states. To identify cases of Zika-associated birth defects, subject matter experts review data reported from medical records of completed pregnancies to identify findings that meet surveillance case criteria (manual review). The volume of reported data increased over the course of the Zika virus outbreak in the Americas, challenging the resources of the surveillance system to conduct manual review. Machine learning was explored as a possible method for predicting case status. Ensemble models (using machine learning algorithms including support vector machines, logistic regression, random forests, k-nearest neighbors, gradient boosted trees, and decision trees) were developed and trained using data collected from January 2016-October 2017. Models were developed separately, on data from the U.S. states, non-Puerto Rico territories, and freely associated states (referred to as the U.S. Zika Pregnancy and Infant Registry [USZPIR]) and data from Puerto Rico (referred to as the Zika Active Pregnancy Surveillance System [ZAPSS]) due to differences in data collection and storage methods. The machine learning models demonstrated high sensitivity for identifying cases while potentially reducing volume of data for manual review (USZPIR: 96% sensitivity, 25% reduction in review volume; ZAPSS: 97% sensitivity, 50% reduction in review volume). Machine learning models show potential for identifying cases of Zika-associated birth defects and for reducing volume of data for manual review, a potential benefit in other public health emergency response settings.