Customize your JAMA Network experience by selecting one or more topics from the list below.
Human illness caused by Zika virus infection has been described for several decades, but this pathogen was in a sense better classified as an infectious diseases “trivia question” before reports of larger outbreaks appeared within the last 10 years.1 Nonspecific symptoms of viral infection, including fever, rash, arthralgia, and conjunctivitis, have been described for Zika infection, and asymptomatic infection is fairly common.1 However, now that strong and accumulating evidence has implicated Zika infection during pregnancy in severe central nervous system sequelae after infection of the fetus,2,3 there has been increased urgency in acquiring a greater understanding of the pathophysiology of Zika disease, and efforts to control the spread of this virus have escalated.
Among the many unanswered questions associated with Zika virus infection during pregnancy is whether the risk of congenital abnormalities is influenced by the gestational timing of maternal infection (early vs late). There is precedence with other congenital infections to anticipate that infection early in pregnancy will lead to greater risk of severe adverse neurodevelopmental outcomes. Congenital rubella syndrome, characterized by a variety of clinical features, including central nervous system, ocular, auditory, and cardiac effects, is well described as having a substantially increased risk of fetal infection when maternal infection is acquired early in pregnancy, with few clinical manifestations when maternal infection occurs after 20 weeks’ gestation.4 For cytomegalovirus, primary maternal infection is less likely to transmit the virus to the fetus early in pregnancy. However, if fetal infection does occur, the risk of severe, often neurologic birth defects is higher when transmission occurs earlier in pregnancy.5 Whether the timing of infection affects the risk of congenital abnormalities is critically important for families, physicians, and the development of public health approaches to screening.
The report from Honein et al6 in this issue of JAMA provides important preliminary data from the US Zika Pregnancy Registry (USZPR), a surveillance system organized by the Centers for Disease Control and Prevention (CDC) to track pregnancy and fetal/infant outcomes after maternal exposure to Zika virus. The authors report that among 442 completed pregnancies among women with laboratory evidence of possible recent Zika virus infection, birth defects potentially related to Zika virus were identified in 26 fetuses/infants, including 22 (85%) with brain abnormalities with or without microcephaly. Infants born to mothers in whom infection likely occurred in the periconceptional period or first trimester (9 cases) or in multiple trimesters including the first trimester (15 cases) had a higher risk of birth defects than in cases in which the exposure was later in pregnancy (11% vs 0%). None of the clinically examined anomalies were identified in infants born to mothers when evidence of Zika infection was only within the second or third trimesters of pregnancy. These data are important because they represent the outcomes associated with maternal Zika infection among US women, although the exposures to the virus occurred in countries with active Zika transmission rather than in the United States.
However, this analysis also has several limitations. The USZPR accumulates data on pregnant women for whom testing has been sent for Zika virus infection, likely resulting in an underestimate of cases in which infection was asymptomatic and correspondingly overestimating cases in which fetal or infant anomalies consistent with Zika virus were identified. An additional limitation of the registry data is that for many of the affected pregnancies, there was potential exposure to Zika virus in more than 1 trimester. Furthermore, the USZPR data likely only include the most immediate and severe neonatal sequelae; infants of many of the exposed women were not tested after delivery, potentially reflecting an underlying lack of comprehensive screening for subtler neurologic findings. In addition, longer-term neurologic outcomes are not yet available because of the preliminary nature of these data. Thus, for congenital Zika infection, based on these data it remains unclear whether the increased risk of congenital abnormalities observed with periconceptional or first-trimester exposure is a result of an increased propensity toward overall fetal infection or, rather, a reflection of an increased severity of adverse fetal effects when infection occurs earlier in gestation. Regardless, it does seem reasonable to suspect that early fetal infection can cause the reported severe central nervous system manifestations such as microcephaly, which at least in part may be related to direct effects of the virus on the developing brain, including destruction of neural progenitor cells7,8 and defects in neural cell migration.9
Honein et al used the USPZR data to address another important question: Are symptomatic and asymptomatic maternal Zika virus infections equally likely to lead to adverse outcomes in the fetus? Because the majority (approximately 80%) of Zika virus infections are thought to be asymptomatic,1 a better understanding of this relationship is important for prognostic and public health reasons and for improving understanding of the pathogenesis of this disease. Congenital malformations after asymptomatic Zika virus infection have been observed previously, most recently in preliminary data from Colombia, in which all 4 infants identified to have both microcephaly and laboratory evidence of Zika virus infection were born to mothers with prior asymptomatic infection.10 Although the findings are limited to some extent by the methods and preliminary nature of their analysis, Honein et al reported similar percentages of fetal/infant anomalies among pregnant women with symptomatic and asymptomatic infection (16/271 [6%] with asymptomatic infections and 10/167 [6%] with symptomatic infections).6 This observation supports current CDC recommendations to screen all pregnant women with exposure to Zika virus.
Although the data presented by Honein et al suggest that risk to a fetus may be greatest early in pregnancy, the outcomes measured in this study include the more severe birth defects that have been associated with Zika virus infection. It is possible that infection later in pregnancy or even early in infancy may still carry a risk for promoting more subtle central nervous system deficits, as brain development continues in the first few years of life and may be adversely affected by inflammation.11 Until more is known about the risk of Zika virus infection in very young infants, counseling caution among parents to minimize exposure of this population also seems prudent.
The CDC recommendations also currently include screening all pregnant women with exposure to the virus.12 The data presented by Honein et al would not seem to alter these recommendations, but a significant gap in practice was identified in this study. Specifically, the authors observed low rates of testing in newborns exposed in utero to Zika virus. Although laboratory evidence of Zika virus infection in the USZPR was predominantly based on maternal samples, 41% of infants born to women with completed pregnancies and laboratory evidence of Zika virus infection had no Zika virus testing completed. This suggests either a knowledge gap among clinicians who care for children or a communication gap between obstetric and pediatric practitioners who cared for these patients, as serologic and polymerase chain reaction testing of infant serum is recommended in this setting.12
Significant progress has been made in a relatively short time in the understanding of Zika virus and its ability to affect the developing brain after in utero infection. Despite this, many questions remain, including about the effectiveness of public health measures to minimize morbidity as the virus spreads into the United States. The population at risk in the United States, although large, has significantly less exposure to dengue virus than prior populations affected by Zika virus. This may have implications for risk of fetal disease13; notably, antibodies to dengue may enhance infection with Zika virus,14 supporting the hypothesis that the apparent increase in birth defects observed in Brazil compared with prior Zika outbreaks in other countries may reflect different epidemiological risks. However, without active and ongoing surveillance efforts such as the USZPR, it would not be possible to even begin to address such questions.
How should the findings of Honein et al influence current practice? These data highlight the critical importance of primary prevention as recommended by the current CDC guidelines. These recommendations include counseling pregnant patients to restrict travel to Zika-endemic areas, avoid mosquito exposure, and use barrier protection with potentially exposed sexual partners.12 Similarly, the risks of periconceptional exposure underscore the CDC recommendation to delay conception after potential Zika virus exposure for 8 weeks in the female partner and 6 months in the male partner.15 Implementation of these critical recommendations mandates access to comprehensive reproductive health care, including effective contraceptive options and early prenatal care. Based on these data, continued support of these services will be critical in curtailing potential adverse outcomes related to this evolving epidemic in the United States.
Corresponding Author: William J. Muller, MD, PhD, Department of Pediatrics, Northwestern University Feinberg School of Medicine, 320 E Superior St, Morton 4-685, Chicago, IL 60611 (email@example.com).
Published Online: December 13, 2016. doi:10.1001/jama.2016.18632
Conflict of Interest Disclosures: The authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest and none were reported.
Muller WJ, Miller ES. Preliminary Results From the US Zika Pregnancy Registry: Untangling Risks for Congenital Anomalies. JAMA. 2017;317(1):35–36. doi:10.1001/jama.2016.18632
Create a personal account or sign in to: