A decade later, what have we learned from the Zika epidemic in children with intrauterine exposure?

Abstract

Since the emergence of the Zika virus epidemic in 2014 and the associated novel sequalae that emerged, much has been learned about the effects of antenatal exposure to Zika virus. Zika virus in pregnancy carries severe teratogenic potential to the fetus, ranging from congenital Zika syndrome to milder neurodevelopmental sequelae. Congenital Zika syndrome is associated with a spectrum of alterations that can affect cognitive, language, and motor development. Among children with congenital Zika syndrome, dysphagia and seizures are common, as are hospitalisations for pneumonia and urinary tract infections; overall, morbidity and mortality are extremely high. Children without congenital Zika syndrome but exposed to Zika virus antenatally are also at risk of developmental disorders. In addition, in utero exposure to Zika virus does not lead to the production of neutralising antibodies. Although the epidemic has subsided, Zika virus remains endemic in many countries and continues to affect families. Maternal associations have been fundamental in advocating for health care for children with congenital Zika syndrome and economic support for families. Gaps in scientific knowledge include the absence of data on long-term outcomes among school-age children. Future research and investments are needed to improve diagnostics, restart the stalled development of Zika virus vaccines, and evaluate antiviral treatments.

Introduction

Zika virus, a mosquito-borne flavivirus that for decades had been endemic in Africa and was later identified in Asia, first emerged in the Americas in 2014. Following studies showing an association between Zika virus and microcephaly,^1–15 in 2016, WHO declared that the cluster of neurological conditions associated with Zika virus to be a Public Health Emergency of International Concern. Antenatal Zika virus exposure was shown to cause a spectrum of clinical conditions, ranging from congenital Zika syndrome to milder presentations. By 2016, Zika virus had rapidly spread across the Americas. The incidence subsequently declined in most countries by November, 2016, and the Public Health Emergency was declared over.

The epidemic’s long-term clinical outcomes can now be assessed. In adults, these outcomes include neurological complications such as Guillain–Barré syndrome.¹⁶ Focusing on children, we can characterise the effects of antenatal Zika virus exposure on development through school age and the socioeconomic burden caring for a child with congenital Zika syndrome poses to families and the health-care systems. This decade of perspective also yields insights about the trajectory of Zika virus vaccine development and the risk of future outbreaks, including the potential contribution of climate change.

Zika virus epidemiology

Zika virus illustrates the potential for re-emerging flaviviruses to cause epidemics. Zika virus, first identified in 1947 in the Zika forest in Uganda in a rhesus monkey,¹⁷ was a relatively unknown arboviral flavivirus that for seven decades caused sporadic disease in sub-Saharan Africa. In 2007, a major Zika virus epidemic was identified in the Yap islands of Micronesia, with 73% of the population affected.¹⁸ This outbreak was caused by the Asian lineage, which shares 90% homology with its ancestral African lineage.⁶ The virus reappeared in 2012–13, in a second major epidemic in French Polynesia, where 30 000 individuals sought medical attention.¹⁹ It was at that time that Zika virus was shown to trigger Guillain–Barré syndrome.²⁰ Cases of microcephaly were not identified during the outbreak, but were shown to occur upon retrospective analysis, after the epidemic in the Americas had begun.¹¹ The virus was subsequently identified in surrounding islands and migrated eastwards in the ensuing months.

By February, 2015, more than 7000 cases of a mild febrile illness with a skin rash were reported in northeastern states in Brazil.²¹ Evaluation for dengue, chikungunya, measles, and rubella viruses did not detect the pathogen, which was ultimately identified by molecular tests as Zika virus of the Asian lineage.^2,3 By October, 2015, 14 Brazilian states had reported local Zika virus transmission. In parallel, an increase in neonatal microcephaly cases occurred in the northeastern state of Pernambuco.^7,21 In November, 2015, Brazil declared a national public health emergency due to concerns about a potential association between Zika virus and microcephaly. The number of microcephaly cases reported in newborns then totalled 1248 cases—20 times more than expected.^9,21,22 Subsequently, the link between Zika virus, microcephaly, and other neonatal abnormalities was established. Zika virus was recovered from the amniotic fluid of two pregnant women whose fetuses had microcephaly on ultrasound.^4,14,22 Zika virus genome was also recovered from tissue and serum samples of a microcephalic infant.¹⁰ Reports of gestational Zika virus infections with fetuses diagnosed with severe central nervous system abnormalities started to surface globally.^23–27 In addition, the virus was isolated from pregnant patients who travelled or whose partners travelled to tropical areas.²⁸ A prospective preliminary report of pregnant patients with PCR-proven Zika virus infection with high frequency of abnormal gestational ultrasounds further supported the viruses’ teratogenic potential.¹ Causality was then shown between Zika virus infection and birth defects of the fetal central nervous system.¹²

In early 2016, WHO declared Zika virus-related microcephaly and other neurological disorders a Public Health Emergency of International Concern.²⁹ By that time, the virus had caused an estimated 750 000 infections in the Americas.³⁰ The name congenital Zika syndrome^31,32 was coined to identify children with severe, specific manifestations of congenital Zika virus infection, including: severe microcephaly with partly collapsed skull; thin cerebral cortices with subcortical calcifications; macular scarring and focal pigmentary retinal mottling; congenital contractures; and marked early hypertonia.^31–33

Between January, 2015 and March, 2017, of the 754 000 reported cases of Zika virus infections in the Americas, the majority were identified in Brazil, Colombia, and Venezuela.³⁰ Although it had been previously reported that Zika virus could be incidentally sexually transmitted,³⁴ studies during the epidemic showed that sexual transmission of the virus was not infrequent and could contribute to outbreaks in non-endemic areas. In the USA, cases of neonatal microcephaly were reported in children born to women who never left the country during pregnancy due to sexual transmission of the virus by a partner who travelled.³⁵ In addition, several cases were reported in Puerto Rico, where the virus was locally acquired through mosquito vector in women who never left the country.³⁰

By July, 2017, Zika virus cases had dropped considerably in Brazil, which was concurrent with a rise in the number of Chikungunya cases.³⁶ The shorter incubation period in the mosquito vector and declining temperatures during autumn favoured the rise of Chikungunya cases over Zika virus in endemic settings.³⁶ This observation suggests that other arboviruses can eventually outcompete Zika virus in endemic areas. Zika virus incidence subsequently declined in most countries and the public health emergency was considered over by mid-2017. In 2018, India experienced a small Zika virus outbreak.³⁷ In 2019, autochthonous Zika virus infections with the Asian strain of virus were reported in Europe (France³⁸) and Africa (Kenya and Angola³⁹). In the Western Pacific (Malaysia and Singapore), sporadic cases were reported and Zika virus-caused neonatal microcephaly was identified in Laos.⁸ In 2020, 22 885 cases of Zika virus disease were reported in the Americas.³⁷ Figure 1 shows the global epidemiology of Zika virus, with 92 countries reporting autochthonous, mosquito-borne Zika virus infections and 60 countries with established Aedes aegypti populations but no documented Zika virus cases.⁸

Figure 1: Geographic distribution of Zika virus cases and mosquito vectors.

Countries in dark grey have reported Zika virus transmission. Countries in light grey have competent vectors but no reported cases. White areas are not known to have Zika virus cases or mosquito vectors. Cases in France refers to rare, limited, autochthonous transmission in southeastern France in 2019.³⁸ The likely vector Aedes albopictus is found in France and nearby countries, including Spain and Portugal.

Pathogenesis and clinical features of postnatal Zika virus infection

The Zika virus public health emergency spurred a flurry of studies to better understand the pathogenesis and clinical features of Zika virus infection. Zika virus is a positive-sense, single-stranded RNA virus in the family Flaviviridae (appendix p 2), transmitted mainly by Aedes mosquitoes (A aegypti and A albopictus).⁴⁰ After a mosquito bite, the virus replicates in dendritic cells, spreading to the lymph nodes and distant parts of the body. The first blood cells infected by the virus are CD14⁺ cells, which allows the virus to avoid host immune responses and infiltrate immune-protected tissues, such as the testes and brain, leading to viral persistence.^41,42 Dendritic cells, dermal fibroblasts, and epidermal keratinocytes play an important role in pathogenesis as they can be infected by Zika virus.⁴³ Perinatal transmission has been established as another important mode of spread.^1,4,11

During the period of acute maternal viremia, Zika virus can cross the placenta and infect placental macrophages.⁴⁴ Type 1 interferon immune responses from cells infected with Zika virus can lead to suppression of placental development, damage to the maternal-fetal blood brain barrier, and placental hypoxia.⁴³ Zika virus is highly neurotropic; its preference for neural progenitor cells triggering apoptosis is a potential mechanism for microcephaly.⁴⁵ Studies in primates and human cohorts show that inflammatory responses induced by Zika virus can cause fetal loss.^13,46 Brain calcifications due to Zika virus are not dead tissue but occur after induction of human bone morphogenetic protein maturation next to brain perivascular cells by the viral NS3 protein, leading to osteogenic gene expression and calcification.⁴⁷

Zika virus infection is asymptomatic in 80% of cases.^18,40 Symptomatic infections have an incubation period of 3 to 14 days, and a duration of up to 7 days.⁴⁸ Clinical manifestations are mainly mild and include low-grade fever, exanthema, arthralgia, myalgia, conjunctivitis, and headache.^1,49–52 Thrombocytopenia has been described and might be attributable to immune-mediated mechanisms.^51,53 Clinical manifestations of acute Zika virus infection are similar across age groups, sexes, and in pregnant and non-pregnant people.^1,54,55 Neurological complications are prominent clinical features resulting from postinfectious immune responses (Guillain–Barre syndrome) or direct viral neurotropism (congenital Zika syndrome).^51,56 Pregnant patients do not need to have symptomatic disease to give birth to infants with features of congenital Zika syndrome.^57,58

Due to limitations of current Zika virus diagnostic methods (ie, high cost, complexity, and short detection window for molecular identification of the virus) laboratory confirmation often remains elusive, especially in non-pregnant, non-neonatal populations. A well defined Zika virus case definition is crucial to overcome the inherent limitations of using clinical symptoms to identify cases—a challenge during simultaneous circulation of multiple arboviruses in endemic areas. Clinical case definitions often vary,^9,29,59 which affects the accuracy of non-laboratory diagnoses.⁶⁰

Zika virus laboratory diagnosis

Before the Zika virus epidemic in the Americas, there were no commercially available tests for molecular diagnostics of the virus or commercially available serological assays. Diagnostics relied primarily on viral sequencing and plaque reduction neutralisation assays performed in research settings. When the epidemic ensued, Zika virus diagnostics rapidly expanded, first with the development of in house PCR and in house serological assays in areas affected by the epidemic, with eventual development of commercially available assays. The diagnosis of Zika virus infection should be considered in individuals presenting with typical clinical manifestations and a relevant epidemiological history, including residence or travel to an area where mosquito-borne transmission of Zika virus was reported or unprotected sexual contact with a person meeting these criteria.^61,62 WHO recommends prioritisation of specific patient groups for Zika virus testing as described in the appendix (p 7).

Nucleic acid amplification tests, such as reverse transcriptase PCR (RT-PCR), are rapid and sensitive methods for confirming Zika virus infection during the acute phase of illness.^59,61,63 RT-PCR assays might not be available in overwhelmed laboratories during outbreaks, hindering a timely diagnosis. Considering the short window period and relatively low levels of Zika virus RNA in serum, negative PCR results do not exclude recent infection.^{40,59,61,64,65} For optimal detection, testing of paired blood and urine samples collected within 2 weeks of symptom onset is recommended to enhance PCR detection.^5,61,62,65 In pregnant patients, Zika virus RNA can be detected in amniotic fluid; amniocentesis is only recommended for exclusion of other diagnoses in fetuses with prenatal findings consistent with Zika virus infection. ^54,59,66 Serological diagnoses for Zika virus, similar to other flaviviruses, primarily relies on ELISA techniques to detect IgM and IgG antibodies. The major limitation is the poor diagnostic specificity of ELISA tests stemming from high cross-reactivity with other flaviviruses, particularly Dengue viruses.^62,67 Assays such as the plaque reduction neutralisation test (PRNT) have greater sensitivity and specificity, measuring virus-specific neutralising antibody titres and can be done to confirm positive IgM antibody results.⁶² Due to its high specificity, PRNT is considered the gold standard for differential flavivirus serodiagnosis. Nevertheless, PRNT is cumbersome, time consuming, and expensive. The test is therefore primarily used for research purposes and is not available in commercial laboratories. The pronounced heterogeneity of laboratory methods is a limitation of seroprevalence studies leading to a gap in assessments of population-level susceptibility.⁶⁷

Testing algorithms are based on sampling relative to symptom onset and serological testing should be considered if samples are negative for Zika virus by RT-PCR.⁶⁴ Zika virus antibody assays can be used variably on serum, plasma, whole blood, or cerebrospinal fluid. IgM generally is detectable in the first week after onset of symptoms and persists for months. Results might be influenced by population immune responses to other circulating flaviviruses and false-positive results are more common with IgM than nucleic acid amplification tests.^61,62

Studies have not shown persistent anti-Zika virus IgG neutralising antibody immune responses in children with PCR-confirmed Zika virus antenatal exposure.⁶⁸ The postnatal diagnosis of antenatal Zika virus exposure is complicated by the fact that most children, even those with congenital Zika syndrome, do not develop neutralising antibodies of their own after maternal antibodies wane.⁶⁸ For this reason, it is not possible to retrospectively identify children with antenatal Zika virus exposure by PRNT assays. In addition, given the absence of neutralising antibodies, these children could potentially be susceptible to re-infection. Regular clinical follow-up is the only method to identify potential short and long-term consequences of Zika virus antenatal exposure. The absence of reliable biomarkers for congenital Zika virus infection remains a key gap requiring further research.⁶⁸

Few studies have reported vertical transmission rates for Zika virus, given the diagnostic challenges. Not infrequently, children born with congenital Zika syndrome whose mothers were infected in the first trimester of pregnancy might have negative RT-PCR results at birth, as the virus is no longer shed. This situation is similar to what happens in congenital varicella syndrome, in which the virus is no longer present after the neurological insult.⁶⁹ In some situations, Zika virus-specific IgM in sera or in the spinal fluid can be positive in the neonatal period, or the virus is identifiable by PCR in blood or urine after birth.⁷⁰ The greatest risk of fetal sequelae after maternal Zika virus infection is with first-trimester infections, but sequelae can occur after infection in any trimester.⁷¹ In a cohort of 130 infants from Rio de Janeiro born to mothers with PCR-confirmed Zika virus infection during pregnancy, the vertical transmission rate of Zika virus through PCR in infant blood or urine or a positive infant IgM result was 65%, independent of infant symptomatology. Among mothers infected in the first trimester of pregnancy, 78% of infants had a laboratory-confirmed infection.⁷² In an analysis of data from seven prospective studies, the estimated mean risk of vertical transmission was 47% (95% CI 26–76) after maternal infection in the first trimester, 28% (15–46) in the second, and 25% (13–47) in the third trimester of pregnancy.⁷³ However, a negative virological or serological test does not necessarily rule out antenatal Zika virus infection.⁷⁴ These findings suggest that vertical transmission of Zika virus is very frequent and highest early in pregnancy. Given the limitations in defining the actual infection, most studies identify infants at risk of perinatal infection as Zika virus-exposed.

Congenital Zika syndrome and antenatal Zika virus exposure

Antenatal Zika virus exposure can cause a spectrum of clinical conditions, ranging from congenital Zika syndrome, the most severe disease manifestation, to other neurodevelopmental, urinary, cardiac, oesophageal, and endocrine conditions to asymptomatic infection.^72,75,76 Congenital Zika Syndrome was first identified and characterised during the Zika epidemic in the Americas. Since that time, much has been established about its natural history and clinical features (panel 1). From 2015 to 2017, 3920 confirmed cases of congenital Zika syndrome associated with Zika virus infection were reported in the Americas (appendix pp 3, 5–6).⁷⁷ Congenital Zika syndrome is considered the tip of the iceberg at the tail end of a spectrum of disease severity which encompasses asymptomatic infection on the other end, following antenatal Zika virus exposure. Cohort studies of mothers with laboratory proven Zika virus infection have shown that microcephaly at birth occurs in about 3–7% of pregnancies affected by maternal Zika virus infection,^{13,57,58,78–82} with development of postnatal microcephaly occurring in another 1% of children in the first months of life. The inclusion of other abnormalities identified on neuroimaging, eye abnormalities, hearing deficits, and severe developmental delay further increase the frequency of neurosensory abnormalities to about 15% in longitudinal studies.⁸³ Microcephaly as a manifestation of congenital Zika syndrome can be disproportional (when only the head circumference is affected) or proportional (when the newborn is small for gestational age and the head is proportionally small). Both forms of microcephaly can be associated with adverse neurodevelopmental outcomes.⁸⁴

Panel 1: Principal findings of congenital Zika syndrome at a glance.

Structural lesions

• Microcephaly

• Cortical atrophy

• Lissencephaly

• Subcortical calcifications

• Ventriculomegaly

• Ocular abnormalities

• Arthrogryposis

Functional lesions

• Seizures

• Muscles hypertonia

• Dyskinesia or dystonia

• Breastfeeding dysphagia

• Visual impairment

• Hearing loss

• Neurodevelopment delay

Long-term consequences of congenital Zika syndrome

The realisation that Zika virus is vertically transmitted and is responsible for fetal complications prompted the development of protocols for the care of affected paediatric populations (panel 2).⁷⁰ A substantial number of infants with congenital Zika syndrome experience complications, such as respiratory infections, dysphagia, and seizures in the first year of life. Studies conducted since the Public Health Emergency of International Concern have shown that after the first year of life, central and peripheral nervous system diseases and congenital anomalies continue to be their main problems. In a Brazilian nationwide cohort study⁸⁵ of 11 481 215 live-born children, the overall mortality to 36 months of age in children with congenital Zika syndrome was 11·3 times higher (95% CI 10·2–12·4) than controls. Cardiovascular complications were responsible for 58% more deaths in the congenital Zika syndrome group compared with children who did not have this condition. Causes of death included cardiomyopathy, cardiac arrhythmias, and heart failure.⁸⁵ For surviving children, long-term developmental trajectories remain unclear. 10 years after the Zika epidemic in Brazil, children living with congenital Zika syndrome have varying degrees of cerebral palsy and require long-term care.

Panel 2: Management of children with antenatal Zika virus exposure*.

Children with microcephaly

• Complete physical and neurological examination every 6 months and when necessary

• Nutritional assessment: anthropometry and body composition

• Electroencephalogram to identify seizures and assist with management of dose adjustments for anti-seizure medications

• Gastrostomy in cases of dysphagia and malnutrition

• Referral to orthopaedic surgery in cases of contractures and pain

• Audiometry

• Echocardiogram

• Ophthalmological reassessment for lens correction, funduscopic eye examination, and visual function

• Inclusion in a school environment

• Stimulation, physical therapy, speech therapy, and support for families

Children without microcephaly

• Complete physical and neurological examination every year and when necessary

• Nutritional assessment: anthropometry and body composition

• Electroencephalogram

• Echocardiogram

• Audiometry and fundoscopic eye examination, visual function, and motility

• Neurodevelopmental follow-up using age-appropriate tools, for example:

  • Bayley-III or IV evaluation of cognitive, motor, and language functions
  • Ages and stages questionnaires 3
  • Wechsler intelligence scale for children
  • Social Responsiveness Scale or Modified Checklist for Autism in Toddlers (MCHAT) screening or Autism Diagnostic Observation Schedule (ADOS II) in cases of suspected autism spectrum disorder
  • Psychological Battery for Attention Assessment in suspected attention deficit hyperactivity disorder
  • School performance assessment

*Panel adapted from clinical practices from the Fernandes Figueira Institute.

Children with microcephaly have higher rates of craniofacial and dental anomalies compared with healthy children.⁸⁶ These individuals require liquid or soft diets, due to insufficient masticatory stimulation and frequent bruxism. ⁸⁷ Caregivers frequently report difficulty feeding their children, which is often linked to neurogenic oropharyngeal dysphagia, a common complication. Nearly 16% of children rely on gastrostomy tubes for feeding.⁸⁸ Another frequent finding is bilateral spastic quadriplegia, which severely affects daily care (diaper changes and dressing) and necessitates a multidisciplinary approach for spasticity management. These children often require orthopaedic surgeries to address contractures and alleviate pain.^88,89 Individuals with microcephaly often present with urinary tract anomalies and reduced bladder capacity.⁹⁰ These conditions, in combination with pneumonia, are common causes of hospitalisation and mortality.

Children with congenital Zika syndrome have a high incidence of severe epilepsy, which is often resistant to standard anti-seizure treatments.⁹¹ This finding is likely due to extensive Zika virus-induced damage to the brain’s cortical structures.^91,92 These children often experience speech, language, and hearing impairments including hearing loss, difficulty swallowing, tongue abnormalities, and delay in overall development.^93,94

Longitudinal assessments in hearing and vision have been performed in children with congenital Zika syndrome. These individuals often experience hearing loss that can affect their ability to speak and communicate. A Brazilian longitudinal study⁹⁴ following children with congenital Zika syndrome up to age 4 years found that most were non-verbal and incapable of producing vocalisations. The researchers note that a subset of children developed some ability to communicate basic needs using alternative methods. According to a case-control study of children aged 6 years in Brazil, individuals with microcephaly had deficits in auditory neural function, including delay in the ability to detect speech.⁹⁵ Associations between communication performance in daily life and cortical auditory evoked potentials highlight the value of auditory brainstem response testing in assessing clinical populations with pronounced neurodevelopmental disabilities.⁹⁵ On follow-up, worsening of both hearing and visual functions over time was not reported.^96,97

Since the Zika virus epidemic, a consensus emerged that children with antenatal Zika virus exposure without congenital Zika syndrome are still at risk for neurodevelopmental delay. In Viet Nam, some children with antenatal Zika virus exposure who were normocephalic at birth had below average developmental testing results at preschool age.⁹⁸ On the other hand, in Grenada, an upper-middle-income country, children of the same age scored no differently than unexposed controls on developmental tests.⁹⁹ In the USA and Colombia, follow-up of children with antenatal Zika virus exposure who were normocephalic at birth showed a higher risk for abnormal neurodevelopmental outcomes in the second and third years of life.^{78,82,83,100,101} Colombian children with and without Zika virus antenatal exposure had similar neurodevelopmental performance at preschool age; however, children with antenatal Zika virus exposure potentially had emerging differences in executive function, behaviour, mood, and adaptive mobility compared with unexposed children of the same age.¹⁰²

Although numerous studies have investigated neurodevelopment of infants antenatally exposed to Zika virus, most focused on the preschool age range or earlier,^{78,83,103–105} leaving a gap in knowledge about long-term outcomes. Neurodevelopmental endpoints in preschoolers have also been investigated in children exposed to antenatal Zika virus without congenital Zika syndrome. Although normocephalic children generally do not have the deficits in gross motor function and cognitive development as children who are microcephalic, some studies report neurodevelopmental alterations in this population. Compared with healthy controls, children aged 5 years born in the USA exposed to Zika virus antenatally due to parental travel have poorer emotional regulation and are less ready to attend school.^106,107 Colombian children exposed to antenatal Zika virus without congenital Zika syndrome showed overall typical development. However, these children showed potential emerging challenges in executive function, behaviour, mood, and mobility compared with their unexposed peers.^106,107 A cohort study of normocephalic children aged 3 years in Guadeloupe reported higher-than-expected rates of hyperactivity in this population.¹⁰⁸ Studies in Grenada and Nicaragua did not observe differences in neurological development at age 4 years between normocephalic children with antenatal Zika virus exposure and unexposed children.¹⁰⁹ These studies suggest that, when present, behavioural and cognitive alterations are more subtle in preschool-age normocephalic children. Nevertheless, a recent study¹¹⁰ in midwest Brazil identified statistically significant delays in neurodevelopmental outcomes in Zika virus antenatally exposed young school-aged children compared with age-matched controls. In this study, Zika virus exposure and non-exposure was defined based on maternal Zika virus PRNT90 assays during pregnancy, effectively ruling out exposure in the contemporary control group of children. Ruling out Zika virus exposure in contemporary control children has been a challenge in many settings given the difficulties in providing a retrospective diagnosis and could ultimately explain the absence of differences observed between cases and controls in some studies.

Children with mild disease should be evaluated for school performance but finding an instrument that can differentiate the spectrum of development delay across diverse cultural contexts has not been straightforward. A recent study¹¹¹ has shown that children exposed to Zika virus antenatally have persistent inflammation measured through serum proteomic profiling at age 2 years as compared with unexposed controls.¹¹¹ These children also have inferior antibody responses to childhood vaccines in comparison with children in the control group. The COVID-19 pandemic had a further negative effect on this population by reducing opportunities for healthy growth and learning and limiting medical access.^112–114

Understanding of long-term repercussions to children is enhanced by pooling data into consortia. An important effort in consolidating data from different Zika virus cohort studies is the WHO Zika Individual Patient Data Consortium that pools data from 65 cohorts in 24 countries.^115,116 However, few cohorts followed children through school age and most do not include a control group of unexposed children. In Brazil, older children are followed with the use of a common protocol for clinical evaluation (panel 2) through a collaboration referred to as the Zika Brazilian Cohorts Consortium.¹¹³ The protocol was created according to the needs reported by mothers and findings on clinical examination. It is subject to change, with inclusion and exclusion of actions or strategies depending on the spectrum of disease in question.

Although few studies of school-aged children are ongoing, available evidence suggests normocephalic children also have developmental disorders. Behavioural disorders, such as autism spectrum disorder and attention deficit hyperactivity disorder, associated with the Zika virus infection have been reported.^78,117–119 Language development has been one of the most affected domains in exposed children born without features of congenital Zika syndrome.^100,101 Because neurodevelopment unfolds over years and the ability to comprehensively evaluate cognitive and neuromotor functions improves with age, it is imperative to conduct long-term studies to fully grasp the consequences of antenatal viral exposure. Monitoring cognitive development, motor coordination, literacy, numeracy, executive functions, mood, and social skills to elucidate long-term consequences of antenatal Zika virus exposure are key. Monitoring future academic performance is particularly important in ongoing cohort studies. Longitudinal monitoring of growth and development might provide a more complete picture of the total congenital Zika syndrome burden^72,120 and long-term outcomes as some subtle learning and social development difficulties might only manifest later in life.

Economic costs of antenatal exposure to Zika virus

The care of children with disabilities due to congenital Zika syndrome can have extensive social and economic effects on families and health-care systems. However, there are scant data on the economic consequences; most studies investigating this issue have used cost parameters from developed countries which are not reflective of the economic realities of developing countries.^121,122 Nevertheless, studies are available for other types of childhood disabilities, such as cerebral palsy, which show a tight correlation between disease severity and economic burden.^115,123,124

At the global scale, it is estimated that Zika virus was responsible for an average of 44 000 disability-adjusted life years (DALYs) lost every year from 2015 to 2019.³⁷ In Brazil, the estimated incremental economic burden from the societal perspective over 10 years was US$178 million for severe congenital Zika syndrome and $433 million for moderate congenital Zika syndrome.¹²⁵ The incremental DALY burden over 10 years in Brazil was substantial, ranging between 24 000 and 47 000 DALYs for severe congenital Zika syndrome and between 50 000 to 100 000 DALYs for moderate congenital Zika syndrome.^37,125 These estimates are likely conservative because so many cases of antenatal exposure to Zika virus might go undetected. As it is not currently feasible to eliminate Zika virus by eradicating its mosquito vector,¹²⁶ society will need to provide resources to support the care of individuals with congenital Zika syndrome for the foreseeable future (figure 2).^92,127 According to a Brazilian study,¹²⁵ families of children with severe congenital Zika syndrome incur costs of more than $50 000 by the time the child reaches the age of 10 years, mainly due to health-care and education expenses. Governmental disability benefits were implemented to support these families, but the available evidence suggests that they were mainly effective at assisting families of children with severe congenital Zika syndrome (microcephaly at least—3 SD on the Fenton growth chart).¹²⁵ More support is needed for families of children with moderate congenital Zika syndrome (individuals with microcephaly at least—2 SD than the mean for sex and gestational age on the Fenton growth chart).¹²⁵ As these families currently receive less disability benefits but face high education and health-care costs, they often have a higher overall burden than households of children with severe congenital Zika syndrome.^125,127

Figure 2: Cumulative economic burden of congenital Zika syndrome for households and government until 2050.

The source of the cost data was the paper by Scherer and colleagues. The curves were created by projecting the model of¹²⁴ forward for the period from 2020–50.

Perspective of families affected by congenital Zika syndrome

The consideration of economic costs alone does not provide a complete picture of the full effect of congenital Zika syndrome on families as it does not account for psychological distress. During the Public Health Emergency period, there were little data available regarding the risk of microcephaly and other adverse neonatal outcomes after Zika virus infection in pregnancy.¹ The absence of robust risk estimates caused tremendous stress to pregnant women and their families. Other sources of distress include grief stemming from a child’s health condition or trauma after fetal or neonatal losses due to congenital Zika syndrome. In Brazil, many pregnancies were carried to term even after the diagnosis of major congenital anomalies due to unavailability of abortion services.¹²⁸ The generation of children with severe cerebral palsy due to congenital Zika syndrome have permanent movement and posture disorders, severe motor and communication disabilities, and high rates of epilepsy with complex needs, placing a substantial emotional burden on their families.¹²⁹ Children with congenital Zika syndrome often require multidisciplinary medical management, creating a medical visit burden for the family.¹²⁵ Another problem for families is the transportation of children from their homes to medical clinics, rehabilitation centres, or schools, especially as they become older and heavier or require a wheelchair or prostheses for mobility. This situation can be particularly overwhelming when children receive treatment in a public clinic, with diminished capacity to provide timely access to specialty care.

These responsibilities can be extremely demanding, affecting the lives of caregivers and the entire family.⁸⁸ The role of caregivers is crucial for the assessment of a child’s health status, comfort, wellbeing, functional skills, and ease of care. Enhancing the health-related quality of life for families with children who have disabilities requires advocating for sustainable public policies that support families and developing coordinated, well planned rehabilitation goals.⁸⁸

The motto of nothing about us without us reflects the attitude of mothers of children vertically exposed to Zika virus. In Brazil, several parent associations were formed with the aim to provide mutual support. These associations, at first, sought to promote and include affected children and families in government programmes, such as the Instalment Benefit Continuada, Minha Casa Minha Vida, and Bolsa Família, in addition to providing information on health and legal assistance. The WHO toolkit was developed for the care and support of people affected by complications associated with Zika virus. The toolkit intends to provide a systems approach involving public health planners and managers so that the necessary infrastructure and resources can be identified and incorporated as needed as well as technical and practical guidance for health-care professionals and community workers. In this way, families claim the right to participate in all processes of building responses to the disease, as they are most knowledgeable about the effects of the virus on their children.¹³⁰

Future perspectives

Recent post-epidemic autochthonous transmission of Zika virus in Europe from 2019 onwards might provide insight into countries at risk for future outbreaks. The mosquito vector A albopictus has expanded its range into southern Europe in the last decade where the population is largely unexposed to Zika virus. We can conjecture that other countries with similar characteristics might be at an increased risk of new outbreaks. Another determinant of future outbreaks is the probability of reinfection in individuals previously exposed to Zika virus. For many infectious disease, immunity wanes over time, leading to outbreaks occurring in a repeating cycle at regular intervals punctuated by interepidemic periods. Whether Zika virus epidemics will reoccur in a cyclic pattern remains unknown. Cases of probable human Zika virus reinfection have been reported in Brazil since the end of the Public Health Emergency period; however, the mean duration of protective Zika virus humoral immune responses is unknown.¹³¹ If waning immunity to Zika virus is widespread, future outbreaks might be possible in countries that have already experienced epidemics. Conversely, experimental infection of non-human primates suggests that Zika virus immunity could be lifelong.¹³² If true, countries that experienced epidemics in 2015–17 might have new outbreaks when individuals born after the Public Health Emergency period comprise substantial proportions of the population. Simulation studies suggest this could occur in the Americas in the mid-2030s.¹³³

An important driver of Zika virus outbreaks is the population dynamics of mosquito vectors. The range limit of Aedes mosquitos and the season suitable for their reproduction are predicted to increase due to global climate change.¹³⁵ By the mid-21st century, due to climate change, the land area that is environmentally suitable for Aedes mosquitoes could expand so much that the human population at risk for Zika transmission could increase by 1·3 billion people.¹³⁴ Another aspect of arbovirus transmission that could be influenced by climate change is the extrinsic incubation period, defined as how long after consuming a blood meal the mosquito becomes capable of transmitting Zika virus. According to a feeding study,¹³⁶ increased temperatures shorten this extrinsic incubation period (EIP) for A aegypti. Simulation studies¹³⁷ with shorter EIPs predict that future incidence of Zika virus could be twice as high as during the last epidemic. Combined with forecasts of warmer temperatures with enhanced migration of mosquito vectors, there is an increased risk of Zika virus outbreaks due to climate change.

Regarding prevention, non-pharmaceutical approaches include use of repellents, netting, staying indoors in air conditioned spaces, wearing long clothing to avoid mosquito bites, and using condoms to prevent sexual transmission.¹²¹ Unfortunately, for some populations, many of these protective measures are unavailable or too costly. Poverty contributes substantially to an increased risk of exposure to the mosquito vector; individuals working outdoors without access to air conditioning and without access to mosquito repellents are more likely to contract Zika virus,¹³⁷ including pregnant women. In many countries that experienced Zika epidemics, the population’s knowledge of these prevention practices remained poor.¹³⁸ Nevertheless, awareness campaigns might be capable of increasing prevention behaviours if these are affordable.¹³⁹ Public awareness is also heightened by the use of awareness campaigns to travelers.¹⁴⁰

Prevention also includes pharmaceutical interventions. Within a year after the Zika epidemic, there were 45 vaccine candidates undergoing in-vitro or animal testing, with six in phase 1 human clinical trials and one in phase 2 trial development.¹⁴¹ A total of 13 vaccine candidates based on various manufacturing platforms have undergone phase 1 trials.¹⁴² The safety and immunogenicity of inactivated Zika virus vaccines in humans were tested in three placebo-controlled clinical trials of two doses of 5 μg 28 days apart (NCT02963909, NCT02952833, and NCT02937233), and a fourth dose reduction study examining safety and immunogenicity was completed in 2021 (NCT03008122). Results showed that vaccines were safe and well tolerated in a two-dose regimen and generated potent neutralising antibody titers. Takeda Pharmaceuticals developed a similar inactivated Zika virus vaccine candidate called TAK-426. A phase 1 clinical trial (NCT03343626) was completed in 2020, and showed that the vaccine was well tolerated, had an acceptable safety profile, and was immunogenic in both flavivirus-naive and flavivirus-primed adults.¹⁴² A live virus attenuated vaccine was developed, attenuated chimeric vaccine rZIKV/D4Δ30–713. This vaccine underwent a phase 1 clinical trial (NCT03611946) which evaluated its safety and immunogenicity in 56 adults with no previous flavivirus infection. This trial was completed in 2022. Although this vaccine was safe and well tolerated, seroconversion rates were poor because chimerisation generated vaccine viruses that were too attenuated to elicit adequate humoral responses.¹⁴² DNA vaccines have also shown encouraging results, being safe and well tolerated in phase 1 clinical trials. VRC5283 advanced to a phase 2 trial and preliminary results indicated that it was highly immunogenic. Three vaccines subsequently advanced to phase 2 trials: two inactivated virus vaccines and a DNA vaccine. Other vaccine modalities have also been investigated. To be licensed by regulatory agencies, phase 3 vaccine trials would be necessary; however, vaccine development was negatively affected by the COVID-19 pandemic. Although there was initially great enthusiasm for development of effective Zika virus vaccines, after the COVID-19 pandemic, Zika virus vaccine studies stalled due to absence of momentum, financing, public interest, and the inability to perform stage 3 trials when the virus is not circulating.

Other prevention technologies, such as prophylactic monoclonal antibodies and repurposed antivirals, were also investigated but are still at a preclinical stage of development.¹⁴³ Human monoclonal antibodies that target key epitopes in the viral envelope region and elicit broad neutralising responses were identified.¹⁴⁴ An obstacle to progress in these interventions is that there is little incentive for private sector investment during interepidemic periods, particularly because there is no way to predict when, where, and whether a new outbreak will occur. Therefore, considerable resources would need to come from public sources.