33: Infections in pregnancy

Infections in pregnancy

Jennifer Amorosa1, Jane Goldman2, and Rhoda Sperling1,3

1Department of Obstetrics, Gynecology and Reproductive Sciences, Icahn School of Medicine, Mt Sinai Hospital, New York, NY, USA

2Division of Maternal‐Fetal Medicine, Department of Obstetrics and Gynecology, The Valley Hospital, Ridgewood, NJ, USA

3Medicine, Infectious Diseases, Icahn School of Medicine, Mt Sinai Hospital, New York, NY, USA


A primary CMV diagnosis in pregnancy is complicated for both the patient and the physician. The outcomes are variable, so counseling the patient about short and long‐term risks to the developing fetus is difficult. The timing of the infection impacts the risk of vertical transmission to the fetus as well as the long‐term outcomes of the affected neonate. Current methods of diagnosis cannot accurately predict if a fetus will be affected and what the long‐term consequences may be. Additionally, there is no proven in utero treatment for infected fetuses.

CMV is a double stranded DNA herpes virus. Infection can be primary or nonprimary. Primary CMV infection is the initial acquisition of the virus. Nonprimary infection results from reactivation of latent virus or reinfection with a different CMV strain. Transmission occurs by person to person contact with infected bodily secretions such as blood, urine, and saliva. Health care workers and those in contact with young children are at greatest risk. Seroprevalence increases with age and varies by geographic area and socioeconomic background.

The incubation period ranges from 28 to 60 days. Viremia can be detected two to three weeks after primary infection [1]. In pregnancy, rates of primary infection range from 1% to 7% [2]. Sometimes, primary CMV causes a mild febrile illness; however, 90% of women are asymptomatic, making the diagnosis difficult. A history of pre‐existing maternal CMV seropositivity decreases, but does not eliminate, the risk of fetal infection because maternal antibodies to CMV cannot prevent reactivation or reinfection. Secondary infection in pregnancy may occur in up to 13% of patients [1].

While viral culture or polymerase chain reaction (PCR) of infected body fluids can be used to diagnose CMV, infection in adults is usually established by serologic testing. Maternal primary infection can be diagnosed by comparing anti‐CMV IgG levels drawn three to four weeks apart. Seroconversion from negative to positive or greater than a fourfold increase in the anti‐CMV IgG titers indicates infection. CMV avidity testing can also be used to aid in determining recent from prior CMV infections. High index values indicate that the infection occurred more than three months prior to testing, making avidity testing in the first trimester particularly helpful in excluding a diagnosis of an acute CMV infection post‐conception.

Table 33.1 Risk of fetal transmission based on timing of maternal CMV infection

Preconception Periconception First trimester Second trimester Third trimester
1–10 weeks before conception One week before last menstrual period to 4–six weeks pregnant

17% 35% 34–42% 43–44% 64–73%

Source: Adapted from Enders et al. [6].

CMV is the most common congenital infection, occurring in up to 2% of all neonates [3, 4]. The risk for severe fetal infection is higher after primary infection than after recurrent infection. Primary CMV is associated with 30–50% risk of mother‐to‐child transmission with a spectrum of disease that encompasses severe multi‐organ disease, neurological impairments, and sensorineural hearing loss (SNHL). Chronic (latent) CMV has been associated with a lower overall risk of vertical transmission (1–3%) and infrequent severe multi‐organ disease. SNHL is the most frequent sequelae [1].

At birth, the majority of infected newborns born to women with primary CMV will be asymptomatic; however, 12–18% will have clinical signs and symptoms of CMV and up to 25% will have neurological sequelae [1, 3]. Thirty percent of severely infected infants die, and 65–80% will have neurologic sequelae [1, 5]. Asymptomatic infants have a lower risk of developing long‐term neurologic problems which include progressive hearing and/or visual loss as well as cognitive impairment.

Vertical transmission occurs transplacentally. The timing of infection affects transmission rates in maternal primary infection (Table 33.1). Enders et al. evaluated 248 pregnancies with primary CMV infection. The mean rate of intrauterine transmission was approximately 38%. Transmission was significantly higher in the third trimester when compared to the first trimester. However, more serious sequelae occur after first trimester infection compared to other trimesters [6]. The neonate can also be infected by exposure to infected breast milk and cervical secretions. In contrast to infections acquired in‐utero, infections acquired post‐natally are often asymptomatic and are generally not associated with long‐term adverse sequelae.

Ultrasound image depicting intracerebral calcification.

Figure 33.1 Intracerebral calcification.

Ultrasound image depicting echogenic bowel.

Figure 33.2 Echogenic bowel.

Ultrasound image depicting abdominal calcification.

Figure 33.3 Abdominal calcification.

Ultrasound image depicting ventriculomegaly.

Figure 33.4 Ventriculomegaly.

Prenatal diagnosis is an option for patients with a known primary infection or with ultrasound findings suggestive of CMV infection such as echogenic bowel, cerebral ventriculomegaly and calcifications and IUGR (Figures 33.133.4). Amniocentesis has a greater sensitivity after 21 weeks gestation and should be performed at least six to eight weeks after onset of maternal infection to perform PCR for CMV DNA. An earlier negative amniocentesis may provide false reassurance so a follow‐up amniocentesis should be considered after 21 weeks gestation. Serial ultrasound assessment can also detect stigmata suggestive of fetal sequelae (Tables 33.2 and 33.3).

Table 33.2 Possible ultrasound findings in fetuses affected with CMV

Abdominal and liver calcifications
Echogenic bowel
Echogenic kidneys
Intracranial calcifications
Cerebral ventriculomegaly
Intrauterine growth restriction
Enlarged placenta
Intrauterine fetal demise

Table 33.3 Potential results of serum testing after expose and management guidelines

IgM Result IgG Result Interpretation and management
Negative Negative Susceptible, repeat testing in four weeks
Negative Positive Immune, not at risk of transplacental transmission
Positive Negative Acute maternal infection, monitor for signs of fetal infection
Positive Positive Subacute maternal infection, monitor for signs of fetal infection

Although fetal infection can be detected by PCR (sensitivity 78–98%), fetal prognosis is difficult to predict. Quantitative determination of viral load in amniotic fluid may help to predict fetal outcome. An abnormal ultrasound examination suggests a poor prognosis, while a normal ultrasound examination does not exclude the possibility of a symptomatic neonate or long‐term neurologic morbidity. A negative result from prenatal diagnosis after primary infection in early pregnancy is associated with favorable outcomes [6].

During pregnancy, there is no proven treatment to prevent fetal disease or reduce the risk of sequelae [7]. Antiviral drugs have not been well‐studied during pregnancy for the prevention of mother‐to‐child transmission and CMV specific hyperimmunoglobulin therapy to reduce symptomatic infection in fetuses and neonates is still investigational. In three prospective studies, CMV‐specific hyperimmune globulin administered to pregnant women with primary CMV infection was associated with decreased transmission and decreased severity of infection [810]. However, a recent randomized placebo‐controlled double blind study of 124 pregnant women with primary CMV at 5–26 weeks did not demonstrate improved outcomes [11]. Rates of congenital infection and symptomatic neonates were similar in both the treatment and placebo group (30% versus 44%, 30% versus 24%). Additionally, there were more adverse obstetrical events in the hyperimmune globulin group than in the placebo group (13% versus 2%), including preterm delivery, preeclampsia, and intrauterine growth restriction.

Universal screening for CMV during pregnancy has not been recommended by ACOG as there are no efficacious treatments for CMV in pregnancy nor is there a CMV vaccine. However, this is an active area of research investigation. Good personal hygiene prevention strategies such as hand washing are recommended to prevent primary infection [1].


Parvovirus B19 is a single stranded DNA virus that is transmitted by respiratory secretions or hand to mouth contact. Most often it is a childhood illness. About 35–65% of pregnant women are immune [1214]. The incidence of acute B19 infection in pregnancy is approximately 3.5% [12] and the risk of vertical transmission is 25% [15]. The risk of maternal parvovirus infection varies by level of exposure to the infected individual. Exposure to infected household members confers the highest risk [14, 16].

Virema from B19 begins roughly six days after exposure and continues for approximately one week [17]. By the time a patient is symptomatic, she is rarely contagious. Children develop a “slapped cheek” appearance and a “lace‐like” rash on the extremities and trunk. Adults often develop a rash on the trunk, preceded by arthropathy of the hands, wrists, knees and ankles. Some patients are asymptomatic [18, 19]. Patients with an underlying chronic anemia, such as sickle cell anemia, may experience a transient aplastic crisis.

Intrauterine B19 infection can spontaneously resolve with no adverse sequelae or it can lead to severe fetal anemia resulting in fetal loss or hydrops fetalis. The overall rate of fetal loss reported in the literature ranges from 5% to 9%. The risk is highest if infection occurs during the first half of pregnancy [13, 1921]. A prospective observational study of 1018 women with acute B19 infection found a fetal death rate of 6.3%. In all cases, maternal infection occurred before 20 weeks gestation. A total of 80% of the fetal deaths occurred within four weeks of maternal infection. Stillbirth, defined as a fetal death at or greater than 22 weeks gestational age, occurred in 0.6% of the pregnancies [19]. A Swedish study examined the etiology of intrauterine fetal deaths occurring at 28 weeks gestation or greater. During the seven‐year study period, data was collected on 33 759 women, 93 of whom had an intrauterine fetal demise (IUFD) (0.3%). Parvovirus B19 was associated with 7 (7.5%) of the fetal deaths in this group making the overall rate 0.02% [22]. None of the seven fetuses were hydropic.

In addition to fetal loss, B19 can lead to hydrops fetalis. This risk is highest if the mother is infected in the first half of pregnancy [19, 23]. Enders et al. demonstrated that pregnancies affected by B19 had a 3.9% risk of developing hydrops fetalis; however, that rate increased to 7.1% if parvovirus B19 was acquired between weeks 13–20 of gestation. The risk decreased to less than 1% if the infection occurred after 32 weeks gestational age [19]. In a later study, the same group found a similar overall rate of hydrops (4.2%). Approximately 11% of fetuses infected between 9 and 20 weeks developed hydrops [23]. Mild hydrops may spontaneously resolved and be associated with favorable perinatal outcomes. On the other hand, severe hydrops can rapidly lead to fetal death if no intervention is undertaken. In fetuses that develop severe hydrops, fetal transfusion helps prevent fetal death. In a prospective study by Enders et al., 85% of fetuses who developed severe hydrops and were transfused survived. All of the fetuses with severe hydrops who were not transfused, died [19]. Fetuses with hydrops may be severely thrombocytopenic so exsanguination at time of transfusion is a concern [24, 25]. Platelet counts should be determined and platelets available at the time of any fetal procedure.

While several case reports have described congenital anomalies in infants affected by parvovirus infection, most intrauterine parvovirus infections are not associated with anomalies [12, 20, 21, 2630].

IgM antibodies to parvovirus are detectable approximately 10 days after exposure and persist for at least three months. IgG antibodies are detectable several days after IgM and typically persist for years. Enzyme‐linked immunosorbent assays (ELISA) and enzyme immunoassays (EIA) are 80–100% sensitive in diagnosing maternal B19 infection. IgM negative patients with significant exposure history, PCR testing may be used to detect disease when the patient’s IgM levels are below the detection limit of ELISA [31, 32]. PCR is the best method for fetal diagnosis because it can detect small amounts of B19 DNA from amniotic fluid [15].

ACOG recommends that pregnant women who are exposed to Parvovirus B19 undergo serum screening as soon as possible after exposure to determine if further monitoring is needed. Women who are IgM negative and IgG positive are immune and are not at risk of transplacental transmission. Women who are IgM positive need to be monitored for potential fetal infection no matter what their IgG status. Women who are both IgM and IgG negative are susceptible to B19 infection and should have repeat testing performed in four weeks. If either IgM or IgG becomes positive, these women should be followed for potential fetal infection [1].

Pregnant women who become infected with parvovirus should be monitored with ultrasound every 1 to 2 weeks for 8 to 12 weeks following exposure to assess for developing anemia. The ultrasound exam should look for signs of hydrops such as ascites and placentomegaly and intrauterine growth restriction. Additionally, fetal middle cerebral artery Doppler assessment should be performed as this can reliably predict fetal anemia [33, 34]. Severely hydropic fetuses have a poor prognosis. If hydrops fetalis or severe fetal anemia is suspected, fetal blood sampling should be performed and intrauterine transfusion considered if the fetus is severely anemic [1, 3537].

ACOG does not recommend routine screening for parvovirus in pregnancy given the low rate of seroconversion during pregnancy as well the variable rate of fetal transmission and range of potential sequelae. Instead testing should be performed on symptomatic patients and patients with exposure to suspected or confirmed cases [1].

Varicella zoster

Varicella zoster virus (VZV) is a highly contagious DNA herpes virus that is transmitted by respiratory droplets, direct contact, or rarely airborne spread. Primary infection causes a diffuse vesicular rash commonly known as chickenpox. Children are most often affected; less than 5% of reported cases occur in adults older than 20 [38]. Since most US adults have immunity to VZV, varicella during pregnancy is rare with an incidence of 1–5 cases per 10 000 pregnancies [39]. Maternal infection during pregnancy can have serious maternal, fetal and neonatal consequences.

Besides the characteristic rash, pregnant women infected with varicella are at an increased risk of varicella pneumonia which can progress to hypoxia and respiratory failure. Before the introduction of antiretroviral treatment, reports estimated the incidence of VZV pneumonia in pregnancy as 10–20% [40] with mortality rates of up to 40% [41]. Recently, Zhang et al. studied a cohort of nearly 1000 pregnancies from the Healthcare Cost and Utilization Project – Nationwide Inpatient Sample (HCUP‐NIS) admitted with VZV infection. The incidence of VZV pneumonia was 2.5% (95% CI, 1.6–3.7) [39]. No maternal deaths were attributable to VZV pneumonia.

Varicella infection between weeks 8–20 of gestation puts the fetus at risk of developing congenital varicella syndrome [42], a rare syndrome is associated with multiple abnormalities (chorioretinitis, congenital cataracts, cerebral cortical atrophy as well as variable degrees of limb atrophy and skin scarring) (Table 33.4). Mortality rates have been reported to be as high as 30%. A total of 15% of cases develop herpes zoster by age four [43, 44].

Table 33.4 Characteristic findings of congenital varicella syndrome

Limb hypoplasia
Ocular abnormalities
Neurologic abnormalities
Fetal growth restriction
Gastrointestinal abnormalities
Skin lesions

In infected mothers, multiple cohort studies have reported that about 1% of fetuses will develop congenital varicella syndrome [4555]. The risk appears to be slightly higher (2%) if the patient is infected between 13 and 20 weeks [53].

Infants delivered within two weeks of maternal varicella infection are at risk of developing severe varicella infections. The highest risk (17–30%) occurs between five days prior to and two days after delivery as the infant is born without having acquired antibody from the mother [56]. The clinical course of neonatal varicella varies depending on the timing of exposure and can range from a mild rash and fever to a disseminated infection. Neonatal varicella that occurs in the first 1–12 days of life is most likely caused by transplacental transmission of VZV, whereas an infant who becomes infected between 12 and 28 days after birth most likely acquired VZV postnatally [57].

The diagnosis of varicella is usually clinical. Laboratory confirmation can be obtained by detecting viral DNA by PCR testing of skin scrapings from the base of a vesicle or through immunofluorescence detection of VZV antigen. Vesicular fluid can also be cultured; however the process is slow and less sensitive than direct detection techniques. Prenatal diagnosis can be made via PCR of amniotic fluid and/or ultrasound diagnosis [5863]. To avoid false negative results, amniocentesis should not be performed until one month after maternal infection [64]. Two studies have looked at laboratory prenatal diagnosis of varicella [46, 65]. Mouly et al. examined PCR testing of amniotic fluid in 107 fetuses and demonstrated an 8.4% transmission rate. Not all of the PCR positive infants had clinical manifestations at birth. Importantly, none of the infants who had a negative PCR went on to develop congenital varicella syndrome [46]. Similar results were reported by Kustermann et al. [65] supporting the idea that a negative PCR is reassuring.

At least five weeks should lapse between maternal infection and fetal ultrasound as imaging performed sooner has failed to detect deformities [60, 66, 67]. If associated anomalies are seen in the setting of a maternal infection, risk of fetal infection is high (Table 33.5). Pretorius et al. published a case series that described 37 maternal VZV infections. Five infants were diagnosed with congenital varicella syndrome, all of whom had sonographic abnormalities. The other 32 were unaffected and other than an isolated case of polyhydramnios, no other abnormalities were found [60].

Table 33.5 Ultrasound findings suggestive of congenital varicella syndrome

Hyperechoic foci in the liver, heart, brain and bowl Limb deformities and contractures
Hydrops Cardiac malformations
Fetal growth restriction Ventriculormegaly
Porencephaly Microcephaly

Postnatally, the diagnosis of congenital varicella requires a history of first or second trimester maternal varicella infection, fetal abnormalities consistent with congenital varicella, and evidence of intrauterine VZV infection. Intrauterine VZV infection can be demonstrated by detection of VZV DNA in the newborn, the presence of VZV IgM antibodies in cord blood, the appearance of clinical zoster early in infancy, and/or the persistence of VZV IgG for more than seven months after birth [44].

Non‐immune pregnant women who have been exposed to varicella should be treated with VariZIG, a purified immune globulin preparation made from human plasma containing high levels of anti‐VZV antibodies (immunoglobulin G). It should be administered within 4–10 days of exposure [68]. Early administration (within four days) may produce milder symptoms in those who go on to develop varicella [69, 70]. Prior to the use of immune globulins, rates of infection were approximately 70–89%. A Phase III, multi‐centered, three‐arm randomized, active controlled study of non‐immune pregnant patients who were treated either with VariZIG or its predecessor VZIG found that 29% of patients treated with VariZIG developed varicella – a significant reduction when compared to historical data [57, 70]. Study VZ‐009, an expanded access protocol that administered VariZIG to high‐risk patients including pregnant women showed a reduction the number of patient with VZV who developed VZV pneumonia suggesting that VariZIG helps to reduce the severity of chickenpox in exposed patients [71].

A cohort study by Enders suggests that intravenous immunoglobulin (IVIG) given to exposed women may prevent congenital varicella syndrome. There were nine cases of congenital varicella in this prospective study of 1373 women. None occurred in the infants of 97 varicella infected women who received post‐exposure prophylaxis with anti‐varicella zoster immunoglobulin [53].

When started within 24 hours of initial rash development, oral acyclovir reduces constitutional symptoms, total lesions and duration of lesion formation. Two pregnancy registries have not demonstrated an increased rate of congenital malformations in pregnant women treated with acyclovir [72, 73] so oral acyclovir or its prodrug valacyclovir should be considered in pregnancy if chickenpox lesions develop. Acyclovir is slowly and incompletely absorbed with bioavailability of about 15–30%; valacyclovir is the orally administered prodrug of acyclovir that overcomes the problem of poor oral bioavailability and exhibits improved pharmacokinetic properties. Historically, mortality rates were as high as 40% in pregnant women who developed VZV pneumonia [41]. While no randomized controlled trials have been performed, case reports and case series have suggested that treatment with intravenous acyclovir may reduce maternal morbidity and mortality [7476]. A review by Smego found a mortality rate of 14% in patients treated with intravenous acyclovir [75] and a recent NICHD/MFMU case–control analysis of 18 women with VZV pneumonia treated with acyclovir had no maternal deaths [77].

Intravenous acyclovir may also help prevent neonatal varicella. Huang et al. who showed that a combination of intravenous acyclovir and IVIG effectively prevented neonatal varicella in infants whose mothers had a varicella rash either seven days prior to or five days after delivery, whereas 50% of the infants receiving IVIG alone developed neonatal varicella [78].

Pre‐pregnancy vaccination is the best way to prevent fetal varicella infection. Highly effective, the varicella vaccine is an attenuated live virus vaccine that prevents ∼98% cases of varicella [79]. First approved in the US in 1995, between 2000 and 2010, varicella incidence declined by 82%. In the first 12 years that the vaccination was available, varicella‐related deaths decreased by 88% [80, 81]. Australian surveillance data also showed a significant reduction in rates of congenital and neonatal varicella (0.8–0.19% and 5.8–2% respectively) after the introduction of the vaccine [45].

In women contemplating pregnancy, varicella history should be elicited. Those with no clear history of a two‐dose vaccine or varicella infection should be vaccinated. It is recommended that women wait for one to three months post‐vaccination to conceive [8284]. Varicella vaccination should be deferred until the postpartum period in non‐immune patients as the vaccine is a live‐attenuated virus vaccine and there are theoretical concerns that transplacental viremia could result in fetal infection and congenital varicella syndrome. That said, post‐marketing surveillance studies of Varivax (Varicella Virus vaccine live [Oka/Merck]) demonstrated no cases of congenital varicella syndrome to women who were inadvertently vaccinated during pregnancy or within three months prior to conception [85, 86].

Ideally, women should be screened for varicella immune status prior to pregnancy. Those who don’t have a disease or vaccination history should be vaccinated prior to conception. Seronegative women who are exposed to varicella during pregnancy should be given VariZIG within 10 days of exposure. In women who develop varicella, oral acyclovir or oral valacyclovir are likely to reduce the severity of clinical disease. Pregnant women who develop varicella pneumonia should be given IV acyclovir. Additionally, in infants of women who had varicella around the time of delivery, consideration should be given to treating the newborn with both IVIG and IV acyclovir in order to prevent neonatal varicella syndrome. Finally, seronegative women should be vaccinated in the immediate postpartum period.


Jul 19, 2020 | Posted by in GYNECOLOGY | Comments Off on 33: Infections in pregnancy
Premium Wordpress Themes by UFO Themes