Viral Infections of the Fetus and Newborn

Epidemiology

The first description of congenital CMV disease was in 1904, when Ribbert observed the large inclusion-bearing cells that represent the typical histopathologic finding of CMV end-organ disease in a stillborn infant. In 1920 a viral cause was proposed for the “cytomegaly” seen in tissue sections of these inclusion-bearing cells, but it would be several more decades before the ubiquitous nature of this virus and the depth and breadth of the disease it produces would be elucidated. In the developed world, CMV transmission occurs in 0.5% to 1% of all successful pregnancies, making CMV infection the most common congenital viral infection. Seroprevalence rates for CMV differ significantly globally, but in general increase with age and are inversely correlated with socioeconomic status. Although the lifetime risk of acquiring CMV infection is high, approaching 90% by the eighth decade of life, seroprevalence is substantially lower among women of childbearing age. Seronegative women are therefore at risk of acquiring primary infections during pregnancy. Among pregnant women, the overall estimated annual seroconversion rate in the United States is 2.3%. Acquisition of CMV occurs via contact with infectious secretions, including blood, saliva, urine, semen, vaginal fluid, and breast milk. Primary sources for maternal CMV acquisition are through sexual activity and exposure to young children.

Young children in group day care settings have particularly high rates of salivary CMV shedding, thus day care providers and parents of young children in day care are at elevated risk of acquisition of primary CMV infections. Health care providers in contrast are not at increased risk of acquisition of a primary CMV infection.

The risk of intrauterine transmission after primary infection is estimated to be 30% to 50%, with increasing rates of fetal infection at later stages in pregnancy. However, the risk of symptomatic congenital CMV is highest for women with primary CMV infection during periconception and the first trimester, with acquisition during the second or third trimester rarely leading to adverse fetal outcomes. In women with preexisting immunity, the risk of congenital infection is significantly lower (1% to 3%). However, reinfection with a different CMV strain, or reactivation, can lead to symptomatic infant disease and sequelae ( Fig. 34.2 ). Paradoxically, therefore the highest rates of congenital CMV are seen in populations with higher CMV seroprevalence.

Postnatal Acquisition of Cytomegalovirus: Implications for the Premature Infant

In addition to congenital infection, CMV can be acquired in the postnatal period. Postnatal CMV infection in healthy term infants is typically asymptomatic, with no convincing evidence of any adverse neurodevelopmental sequelae. In contrast, primary CMV infection in preterm infants can lead to significant morbidity. Potential routes of neonatal infection include exposure to infectious cervicovaginal secretions during delivery, ingestion of breast milk, contact with saliva, and via blood transfusion. Of these, the most common is via breast milk. Detection of CMV by PCR or culture in breastmilk samples of CMV positive women is reported between 63% and 97% and approximately 20% (range 6% to 59%) of premature infants fed breastmilk from CMV-infected mothers will become infected. Risk for severe illness is greatest in very low birth weight (VLBW) infants, in whom rates of symptomatic illness or death are as high as 17% to 18%. Symptoms in these infants include hepatopathy, neutropenia, thrombocytopenia, pneumonitis, and sepsis-like deterioration. The Centers for Disease Control and Prevention (CDC) has estimated that annually up to 4.5% of VLBW and premature infants, or an estimated 2000 infants, in the United States may develop CMV sepsis syndrome because of breast milk-acquired CMV infections. Acquisition of CMV infection in the premature infant may contribute to the development of chronic lung disease. The neurodevelopmental consequences of postnatal CMV infection in preterm infants are unclear, with some studies finding a detrimental effect and others not.

Proposed efforts to reduce the infectivity of breast milk from seropositive mothers have included freezing breast milk at −20°C, Holder pasteurization, and short-term pasteurization. Of these methods, freezing is the most studied and most likely to maintain the salutary immunologic properties of breast milk. Although freezing of breast milk may lower the incidence of postnatally acquired CMV infection, it does not eliminate the risk. Further evidence is necessary to make recommendations regarding what, if any, interventions are appropriate in low birth weight, preterm infants receiving breast milk from CMV-seropositive mothers.

Transfusion-associated CMV infections were at one time a major problem in the neonatal intensive care setting. Two approaches are currently used to decrease the risks of transfusion-associated CMV infection: leukocyte reduction and directed transfusion of CMV-negative blood products. A prospective, multicenter birth-cohort study conducted at three neonatal intensive care units (NICUs) in the United States found no evidence of transfusion-associated CMV infection when leukocyte-reduced blood was used for transfusion of VLBW infants. These data support the practice of transfusing leukocyte-reduced blood products to premature infants, including blood from CMV-seropositive donors, to prevent transfusion-associated CMV infection in the NICU.

Pathophysiology

The mechanisms by which CMV injures the fetus involve a complex interplay of viral gene products, maternal immune response, and placental biology. The pathogenesis of disease associated with acute CMV infection has been attributed to lytic virus replication, with end-organ damage occurring either secondary to virus-mediated cell death or from pathologic host immune responses targeting virus-infected cells. The factors that contribute to fetal injury include the timing of infection relative to the gestational age of the fetus, the maternal immune status, the extent of associated placental injury, the magnitude of the viral load in the amniotic fluid, the induction of host genes occurring in response to infection, and possibly the genotype of the particular strain of CMV infecting the fetus. Much of the injury that CMV produces in the newborn may be caused by placental insufficiency. CMV infection of the placenta might also contribute to IUGR and fetal injury via induction of proinflammatory cytokines and modulation of normal trophoblast gene expression. The developing fetal brain is also highly susceptible to CMV-induced injury. The pathogenesis of CMV infection in the central nervous system (CNS) seems to be strongly related to perturbations in neural migration, neural death, cellular compositions, and the immune system of the brain. In infants with symptomatic congenital CMV infection, histopathologic evidence of viral dissemination can be found in the brain, ear structures, retina, liver, lung, kidney, and endocrine glands. The pathogenesis of sensorineural hearing loss (SNHL) is related to an inflammatory labyrinthitis. CMV-infected cells can be seen in the semicircular canals, vestibular membrane, cochlea, and other structures of the ear.

Clinical Presentation

Signs and symptoms are apparent at birth in 10% to 15% of all children with congenital CMV infection. Infection in the symptomatic infant can involve any organ and manifests itself along a spectrum from mild illness to severe disseminated multi-organ system disease (see Table 34.1 ). A wide spectrum of disease can be observed, including hemolysis, bone marrow suppression, hepatitis, pneumonitis, enteritis, and nephritis. Infants with symptomatic disease are often premature and small for their gestational age. Clinical features include jaundice, hepatosplenomegaly, lethargy, respiratory distress, seizures, and petechial rash. Common laboratory abnormalities include thrombocytopenia, anemia, and elevated transaminase and conjugated bilirubin levels.

CNS disease is common in symptomatic infants: signs and symptoms may include hypotonia, lethargy, seizures, and microcephaly. Imaging by cranial ultrasound may reveal intracranial calcifications, periventricular cysts, ventriculomegaly, cerebellar hypoplasia, or lenticulostriate vasculopathy. Additional findings such as white matter abnormalities and neuronal migration disorders may be identified by MRI.

Ophthalmologic abnormalities are seen in 20% to 40% of symptomatic infants, but rarely in asymptomatic infants. Common findings include retinal scars, optic atrophy, and chorioretinitis. SNHL is either present at birth or develops in 40% to 70% of symptomatic infants and 10% to 15% of infants who are otherwise asymptomatic. SNHL can be progressive and fluctuating in both asymptomatic and symptomatically congenitally infected infants. It ranges in severity from a unilateral, mild hearing deficit to severe, bilateral, profound deafness. CMV-induced SNHL may be present at birth or can appear later in childhood. Delayed-onset hearing loss usually presents in the first few years of life but has been reported to evolve and progress through 5 years of age and beyond.

Mortality due to symptomatic congenital CMV disease in the neonatal period is uncommon, likely less than 5%. However, long-term neurodevelopmental disabilities are observed in 50% to 75% of children who are symptomatic at birth and can include motor deficits (paresis or paralysis), cerebral palsy, intellectual disability, seizures, vision impairment, hearing loss, and learning disabilities. In contrast, several studies have suggested that the intellectual development of asymptomatic congenitally infected infants appears to be normal.

Diagnosis

Management

The benefits of ganciclovir in neonates with symptomatically congenital CMV have been demonstrated in controlled trials. Neonates with virologically confirmed symptomatic congenital CMV disease involving the CNS who received a 6-week course of IV ganciclovir therapy had a statistically higher likelihood of normal or improved hearing at 6 months of age compared with untreated neonates. Infants in this trial who received IV ganciclovir therapy also had fewer developmental delays at 6 and 12 months, as assessed by the Denver Developmental Screening Test, compared with untreated infants. A subsequent study compared 6 weeks versus 6 months of oral valganciclovir in neonates with symptomatic congenital CMV disease. In this study, all participants received valganciclovir for 6 weeks. Participants were then randomized to receive either continued valganciclovir therapy or placebo for 4.5 months. Hearing was more likely to be improved or to remain normal at 12 and 24 months in the 6-month treatment group than in the 6-week treatment group, and this group also had better neurodevelopmental scores on the Bayley Scales of Infant and Toddler Development at 24 months. Six months of oral valganciclovir therapy ( Table 34.2 ) should thus be considered for all neonates with symptomatic congenital CMV infection with any evidence of CNS involvement (microcephaly, abnormal CNS imaging, CSF positive for CMV DNA, chorioretinitis, or evidence of SNHL). Careful monitoring for neutropenia, the major side effect of valganciclovir therapy, is essential.

Ganciclovir therapy should also be used in any infant with severe or life-threatening end-organ CMV disease (see Table 34.2 ), whether acquired via congenital infection or by a postnatal route, such as via breastfeeding in a low birth weight premature infant. CMV chorioretinitis, if present, should be managed in consultation with an ophthalmologist and infectious diseases expert and may require prolonged ganciclovir treatment.

Although ganciclovir/valganciclovir improves hearing and neurodevelopmental outcomes for symptomatic infants when initiated within the first month of life, it is not yet clear whether valganciclovir holds the promise of improving outcomes in infants with asymptomatic congenital CMV infection, or for infants presenting outside the neonatal period with isolated SNHL. A clinical trial is ongoing to determine whether valganciclovir therapy could prevent progression of hearing loss in older infants with isolated SNHL.

Ganciclovir has been demonstrated to cross the placenta, and therefore could theoretically be used to treat CMV infection in utero, however there are concerns for potential teratogenicity. There have been case reports of treatment of congenital CMV infection in utero with valganciclovir, but no systematic trials. High-dosage oral valacyclovir therapy (8 g/day) in recent studies has been found to be safe and to decrease the risk of fetal CMV infection and symptomatic congenital disease.

Passive immunization with CMV human immunoglobulin (HIG), a pooled, high-titer immunoglobulin preparation, has been studied for in utero treatment and prevention of congenital CMV infection. Reduction in congenital infection and symptomatic neonatal disease has been seen in CMV HIG-treated pregnant women in several uncontrolled unblinded studies. However, other observational studies and a randomized, blinded placebo-controlled trial of CMV HIG for the treatment and prevention of congenital CMV infection failed to demonstrate a benefit. A National Institutes of Health-funded multicenter randomized, placebo-controlled trial of CMV HIG during pregnancy in the setting of primary maternal infection was recently stopped early due to futility with no significant difference in transmission between the treatment and placebo arms of the study. Therefore, it remains unclear if CMV HIG is useful in prevention of viral transmission or sequelae.

Prevention

One important strategy for addressing the problem of congenital CMV infection is the education of women of childbearing age about the risks of CMV transmission and strategies for prevention. Because most maternal CMV infections are asymptomatic, a major goal is education of all women of childbearing age on hygienic practices. Hygienic strategies are important because the saliva and urine of infected children are significant sources of CMV infection among pregnant women. Strategies include washing hands whenever there is contact with a child's saliva or urine, not sharing food, utensils, or cups, and not kissing a child on the mouth or cheek. In several surveys, only 14% to 30% of women had previously heard of CMV. Therefore, educating women on methods to prevent CMV transmission may be effective to decrease seroconversion during pregnancy.

Prenatal maternal screening for CMV antibodies is controversial. Women who are CMV immune can be infected with new strains with subsequent fetal infection and developmental sequelae. A positive preconception titer result for CMV IgG antibody may provide a false sense of reassurance and decrease a pregnant woman's motivation to engage in careful hygienic practices.

Ultimately the control of congenital CMV infection could be realized by the development of an effective vaccine. No CMV vaccines are currently licensed; however, multiple candidate vaccines are in various stages of development including in active clinical trials.

Epidemiology

The annual incidence of rubella cases has decreased dramatically from 58 per 100,000 population in 1969 to only 10 cases reported in the US in 2012. CRS cases in the United States have demonstrated a similar dramatic decline, and in 2004 the CDC concluded that endemic rubella had been eliminated from the United States. Rubella is still an important and potentially preventable cause of birth defects globally, with more than 100,000 cases of CRS occurring annually in developing countries in the last decade ( Table 34.3 ). This number is likely lower now as significant progress toward elimination of rubella has been made worldwide, although there are still countries with suboptimal national vaccination programs.

The frequency of congenital infection after maternal rubella is 70% to 85% if infection occurs during the first 12 weeks of gestation, 30% to 54% during the first 13 to 16 weeks of gestation, and 10% to 25% at the end of the second trimester. The classic findings of congenital rubella are most typically associated with the onset of maternal infection during the first 11 weeks of gestation, and the risk of any teratogenic effect is extremely low after 17 weeks' gestation.

Pathogenesis

Rubella virus is transmitted via respiratory droplets. Viral replication begins in the upper respiratory tract and nasopharyngeal lymphoid tissue, followed by contiguous spread to regional lymph nodes and hematogenous dissemination to distant sites. Fetal infection is believed to occur as a consequence of maternal viremia. The mechanism by which fetal rubella infection leads to teratogenesis has not been fully determined, but the cytopathology in infected fetal tissues suggests necrosis, apoptosis, or both, as well as inhibition of cell division of precursor cells involved in organogenesis. Macrovascular fetal endothelial cells have been found to be highly permissive for rubella virus replication, suggesting that the vascular diseases in CRS are triggered by persistent rubella virus infection of endothelial cells.

Clinical Presentation

The typical illness in adults and children with acquired rubella consists of an acute generalized maculopapular rash, fever, and arthralgias, arthritis, or lymphadenopathy. Conjunctivitis is also common. Viremia can be detected as early as 9 days prior to the onset of rash. Up to 50% of primary infections are asymptomatic.

Infants with congenital rubella are usually born at term but are often small for their gestational age. Transient manifestations may include skin lesions described as resembling a blueberry muffin which represent extramedullary dermal hematopoiesis, thrombocytopenia, hyperbilirubinemia, and leukopenia, hepatosplenomegaly, jaundice, pneumonia, and meningoencephalitis. Radiographic findings include a large anterior fontanel, linear areas of radiolucency in the long bones (i.e., celery stalking), increased densities in the metaphysis, and irregular provisional zones of calcification. Permanent findings include SNHL, heart defects, eye abnormalities, neurologic defects, and developmental delay ( Table 34.4 ). The most common isolated finding of congenital rubella is SNHL. Cardiac lesions are noted in 45% to 70% of infants with the most common abnormalities being patent ductus arteriosus, peripheral pulmonic stenosis, and valve abnormalities. Ocular findings include cataracts, pigmentary retinopathy, microphthalmia, glaucoma, and strabismus. The triad of deafness, cataracts, and congenital heart disease constitutes the classic CRS.

The clinical manifestations of CRS differ to some extent depending on the timing of fetal infection. In a prospective study following up pregnant women with confirmed rubella by trimester, a full range of rubella-associated defects (including congenital heart disease and deafness) were observed in nine infants infected before the 11th week. Thirty-five percent of infants (9 of 26) infected between 13 and 16 weeks’ gestation had deafness alone.

Evaluation

Diagnosis of congenital rubella syndrome is based on both clinical and laboratory findings. In the first year of life, laboratory evidence of congenital rubella infection can be demonstrated by a positive rubella IgM titer; infant rubella IgG level that persists at a higher level and for a longer period than expected from passive transfer of maternal antibody; isolation of rubella virus from nasal, blood, throat, urine, or CSF samples; or detection of viral nucleic acid by PCR from the throat, CSF, or lens (obtained during cataract surgery). An infected infant can shed the virus for many months after birth despite the presence of neutralizing antibody and thus may pose a hazard to susceptible individuals.

Women who are exposed to rubella during pregnancy should be screened for evidence of previous immunity. If rubella IgG is detectable at the time of exposure, the fetus is probably protected. If no antibody is detectable, additional samples for rubella IgG and IgM should be obtained at 2 to 3 weeks after exposure and again at 4 to 6 weeks after exposure. These samples can be run concurrently with the first serum sample to ascertain whether infection has occurred (i.e., seroconversion).

Management

There is no specific antiviral therapy for congenital rubella. Initially the infant may need general supportive care, such as administration of blood transfusion for anemia or active bleeding, seizure control, and phototherapy for hyperbilirubinemia.

Outcomes

Long-term care requires a multidisciplinary approach consisting of occupational and physical therapy, close neurologic and audiologic monitoring, and surgical interventions as needed for cardiac malformations and cataracts. Over half of surviving infants have neurologic and/or developmental abnormalities, which may not be diagnosed until the infants are older. Other delayed manifestations include late-onset glaucoma and other ophthalmologic abnormalities, progressive SNHL, autism spectrum disorders, progressive rubella panencephalitis, abnormal dental development, and immunologic and autoimmune endocrine disorders (see Table 34.3 ). A higher-than-expected incidence of autoimmune diseases, such as thyroid disorders and diabetes mellitus, has also been reported years after the diagnosis of congenital rubella. Delayed onset immunologic abnormalities include abnormalities in immunoglobulin levels and cellular immune responses.

Prevention

The critical intervention in prevention of CRS is to ensure that women who are considering pregnancy get appropriately vaccinated. The Advisory Committee on Immunization Practices recommends screening of all pregnant women for rubella immunity and postpartum vaccination of those who are susceptible. Immunity to rubella appears to confer almost complete protection against CRS. Rare cases of documented subclinical maternal reinfection with rubella virus leading to CRS have been reported.

Live attenuated rubella virus vaccine is safe and effective, although the duration of immunity is uncertain. Rubella vaccine is currently administered in the United States in a trivalent formulation in combination as measles–mumps–rubella (MMR) vaccine. The vaccine is recommended for children at 12 to 15 months of age and at 4 to 5 years of age. It is also recommended for women of childbearing age in whom the results of both a hemagglutination inhibition antibody test and a pregnancy test are negative. Although no cases of symptomatic congenital rubella have been reported after vaccination during pregnancy in the more than 500 cases monitored, vaccination is not recommended during pregnancy because of the theoretical hazard to the fetus. If a pregnant woman is found to be nonimmune, vaccine should be administered during the immediate postpartum period before discharge. Breastfeeding is not a contraindication to postpartum immunization. Immunization in the postpartum period has rarely produced polyarticular arthritis, neurologic symptoms, or chronic rubella viremia. Immune globulin does not prevent rubella infection after exposure and is not recommended for routine post-exposure prophylaxis in pregnancy.

Epidemiology

Parvovirus B19 infection is common in childhood and continues at a low rate throughout adult life. The virus is spread primarily by respiratory droplets but can also be transmitted by blood products and transplacentally. Seroprevalence in the population varies by age and geographic location, but averages about 40% to 60% in women of childbearing age. Seroconversion during pregnancy is reported to be about 1% with increased rates noted during periodic epidemics.

It is estimated that one-fourth to half of maternal parvovirus B19 infections result in transmission of infection to the fetus. The spectrum of diseases associated with parvovirus B19 infection remains incompletely defined, as there is no routine surveillance during pregnancy; however, most pregnancies are unaffected. The overall risk of adverse fetal outcome is between 3% and 12%. The risk of adverse fetal outcome is increased if maternal infection occurs before 20 weeks' gestation.

Pathophysiology

Fetal infection most likely occurs transplacentally during maternal viremia. Parvovirus B19 binds to a glycolipid receptor (P-antigen or globoside) present on erythroid progenitor cells and placental trophoblasts. P-antigen expression by villous trophoblasts is gestation dependent and is highest in the first trimester. It has been postulated that parvovirus B19 infection leads to cytotoxicity and subsequent anemia by inducing apoptosis of infected red blood cells. The virus targets the erythroid lineage cells in the fetal liver, the primary site of erythrocyte production in the fetus. The fetus is especially susceptible to adverse consequences of red blood cell infection secondary to the intrinsic short fetal erythrocyte life span and rapidly expanding blood volume, especially during the second trimester. The P-antigen is also expressed on fetal cardiac myocytes, enabling parvovirus B19 to infect myocardial cells, leading to myocarditis. Myocarditis induced by parvovirus B19 can contribute to high-output cardiac failure, and the myocardial inflammation and subendocardial fibroelastosis may also contribute to fetal hydrops.

Clinical Presentation

Parvovirus B19 infection causes erythema infectiosum in normal hosts, aplastic crisis in patients with hemolytic disorders, and chronic anemia in immunocompromised hosts. A substantial proportion of infected adult women may also have arthropathy in association with parvovirus B19 infection. Once the virus establishes infection, viremia occurs, accompanied by mild systemic symptoms such as fever and malaise. Viremia is short-lived, lasting only 1 to 3 days, and the characteristic immune-mediated rash develops 1 to 2 weeks later. Once the rash appears, an individual is no longer infectious. Infection with parvovirus B19 infection during pregnancy may be asymptomatic more than half of the time.

The major clinical presentation of parvovirus B19 infection in the fetus is nonimmune hydrops fetalis (NIHF). Various estimates suggest that human parvovirus B19 infection contributes from 10% to 27% of cases of NIHF. The risk of NIHF secondary to maternal parvovirus B19 infection is between 4% and 12% in most reports and is highest for women who acquire parvovirus B19 infection between 13 and 20 weeks' gestation. NIHF usually develops 2 to 8 weeks after maternal parvovirus B19 infection. The role that parvovirus B19 plays in intrauterine fetal demise in the absence of hydrops fetalis is incompletely defined. Some studies show increased risk of fetal loss, spontaneous abortion, and stillbirth, while others do not. The estimated risk of fetal loss in infants infected with parvovirus B19 prior to 20 weeks gestation is 13% and decreases to 0.5% after 20 weeks’ gestation.

Parvovirus B19 has been implicated in some cases of congenital anemia. Investigation for parvovirus B19 infection should be included in the evaluation for congenital anemias. Thrombocytopenia is common in parvovirus B19-infected fetuses with NIHF. In one series, 46% of fetuses with parvovirus B19-associated NIHF had severe thrombocytopenia.

Evaluation

In children and adults, the diagnosis of human parvovirus B19 infection is most commonly made clinically through recognition of the characteristic rash. Women exposed to parvovirus B19 during pregnancy should be tested for virus-specific IgG and IgM antibodies. Those who are IgG positive and IgM negative are immune and therefore not at risk for infection. Women who are IgM and IgG negative should be monitored for seroconversion by repeat serologic testing in 4 weeks. Women who develop anti-parvovirus B19 antibodies or who demonstrate parvovirus B19 IgM at initial testing should be monitored for potential fetal infection ( Fig. 34.3 ). Parvovirus B19 DNA can be detected in plasma of infected women early in the course of infection and is more sensitive than serology for diagnosis of acute infection. A combination of serology and PCR can be used to improve the accuracy of diagnosis.

Pregnant mothers with evidence of acute infection should be counseled regarding risks of fetal transmission, fetal loss, and hydrops, and serial ultrasound examinations should be performed every 1 to 2 weeks, up to 12 weeks after acute infection, to detect development of anemia and hydrops. Monitoring should include assessment for ascites, placentomegaly, cardiomegaly, hydrops fetalis, and impaired fetal growth. Middle cerebral artery Doppler imaging for peak systolic velocity should be performed as this is an accurate predictor of fetal anemia which can be detected with this technique before fetal hydrops is evident. Maternal parvovirus B19 infection should also be considered when prenatal ultrasounds show evidence of fetal anemia, cardiomegaly, or hydrops fetalis. As fetal manifestations of parvovirus B19 infection occur weeks after maternal infection, absence of maternal parvovirus B19 IgM or plasma B19 DNA does not exclude infection. In one study, women tested due to suspicion of fetal infection were IgM positive 62.5% and DNA PCR positive 87.5%. Detectable parvovirus B19 DNA by PCR in amniotic fluid or fetal blood establishes the diagnosis in the fetus.

Management

Spontaneous resolution of fetal anemia without hydrops has been reported in about 50% of fetal infections, however, spontaneous resolution of hydrops is less common (5% to 30%). Because most fetuses do not recover without intervention, fetal transfusion is usually recommended. The earlier intrauterine transfusion (IUT) is attempted, the more likely it is to be successful. Cordocentesis allows precise assessment of the magnitude of fetal anemia, which can then be corrected by blood transfusion, typically using packed red blood cells. With this approach, outcomes have been favorable in most reported series, even among severely anemic fetuses. Because of the frequency of concurrent thrombocytopenia, platelet transfusion is also frequently required at the time of IUT. Small studies suggest a survival rate of 76-85% for fetuses with severe hydrops fetalis who receive IUT, compared to 100% mortality for non-transfused fetuses. Although high-dose IVIG has been used in the setting of acute infection to attempt to prevent hydrops fetalis during pregnancy, treatment with this modality in the pregnant woman has not been shown to improve fetal outcomes.

Outcomes

Results from several long-term prospective studies of infants born to mothers with documented primary parvovirus B19 infection during pregnancy found no evidence of increased risk of long-term morbidity, development delay, or death in childhood. Increasingly, however, there have been reports of neurologic morbidity in infants surviving severe anemia or NIHF due to parvovirus B19. Structural brain abnormalities have been seen in infants with severe anemia or NIHF, including small cerebellum, polymicrogyria, and heterotopia. Neurologic abnormalities have been noted on follow-up in approximately 10% of infants surviving IUT for B19-associated NIHF.

Prevention

There has been limited progress in the development of a candidate parvovirus B19 vaccine. At the current time, routine screening of pregnant women for parvovirus B19 is not recommended. If a pregnant woman has a significant exposure to an infectious case of parvovirus B19, counseling should be provided regarding the potential risk of infection and testing and follow-up as described above should be recommended.

Pregnant healthcare providers should be counseled about the potential risks to their fetus from parvovirus B19 infections. They should consider not caring for immunocompromised patients with chronic parvovirus B19 infection or patients with parvovirus B19-induced aplastic crises, or should, at a minimum, follow strict infection control procedures including standard droplet precautions.

Epidemiology

Like many flaviviruses, Zika is spread through mosquito vectors, in this case primarily Aedes aegypti . The virus is therefore limited to the equatorial range of this mosquito, although climate change is predicted to enlarge this range. The primary reservoirs are thought to be humans and non-human primates. In addition to blood, Zika virus can be isolated from seminal fluid (as long as 69 days after initial infection) and, to a lesser extent, from saliva, vaginal fluid, and breast milk. Sexual transmission has been well documented. Although it likely only contributes to a small percentage of overall cases, sexual transmission can contribute to new cases in non-endemic areas. Transmission via breast milk has been proposed but documented evidence is quite rare. Zika transmission is seasonal. Although cases can occur any time during the year, the majority occur during summer months. Serological studies have demonstrated a relatively high prevalence in endemic areas: 10% in Southeast Asia and 13% in West Africa. At the peak of the outbreak in the Americas in the first half of 2016 seroprevalence reached levels as high as 63%. While new cases in the Americas have decreased sharply since 2016, local spread continues in endemic areas including West Africa, the Caribbean, and Southeast Asia. Fetal anomalies have only been associated with the Asian lineage of Zika and have been documented in the Americas, Southeast Asia, and Africa. Up-to-date epidemiologic data is available through the World Health Organization (WHO).

Pathophysiology

Initial inoculation with Zika virus is followed by an incubation period estimated to last 6 days (range: 3 to 14), followed by a symptomatic phase that corresponds to viremia and typically lasts for 5 to 6 days. In a pregnant mother, the virus can gain access to the fetus via transplacental spread. Zika has a high degree of tropism for placental trophoblasts and the fetal central nervous system, establishing infection in these tissues that may persist for months. For Zika infection, as for most congenital infections that cause fetal anomalies, the consequences for fetal development are worst when infection occurs late in the first or early in the second trimester. Congenital Zika syndrome (CZS) includes a heterogeneous set of anomalies that reflect the virus’ tropism for neural tissue and harmful consequences on brain development: microcephaly, thin cerebral cortex with subcortical calcifications, retinal abnormalities, congenital contractures, and early hypotonia. The mechanism underlying these findings remains the subject of active investigation, but the pattern of injury mimics fetal brain disruption sequence, a syndrome in which loss of cells in the brain parenchyma results in collapse of the fetal skull during development. Viral infection of neural progenitor cells is thought to be central to CZS, leading to disruption of the fetal brain either by directly causing cell death, altering developmental programs, or inducing a type I interferon response. Epidemiological studies have only linked CZS to Asian lineage Zika virus. There is speculation that this is because congenital infection with African lineage viruses results in severe pathology and ultimately loss of pregnancy, whereas pathology resulting from Asian lineage virus infection is less severe resulting in a viable pregnancy, albeit with abnormal development.