Mother-to-Child Transmission of Cytomegalovirus and Prevention of Congenital Infection

Abstract

Mother-to-child transmission (MTCT) of cytomegalovirus (CMV) occurs transplacentally (congenital infection), during birth and through breast milk, although the latter 2 modes of transmission are not associated with the central nervous system sequelae that occur with congenital infection. CMV persists indefinitely in its human host, and MTCT can occur if the mother was infected in the past or during the current pregnancy. The goal of efforts to prevent MTCT of CMV is to prevent congenital infection, an important cause of disability due to hearing loss, impaired vision, cognitive impairment, and neuromotor deficits. Vaccines for prevention of maternal and congenital CMV infection are being developed but will not likely be available for at least a decade. Rather than waiting for an effective vaccine to solve the problem, more effort must be devoted to defining the potential for public health measures to prevent congenital CMV infection by reducing rates of maternal infection during pregnancy.

Mother-to-child transmission (MTCT) plays a key role in maintaining the prevalence of cytomegalovirus (CMV) in humans. CMV is well adapted to persistence in its human hosts; a substantial portion of its large genome is devoted to gene products that can interact with, evade, or take advantage of the human innate and adaptive immune systems and inflammatory responses [1]. Initial CMV infection in a healthy adult or child is usually clinically silent but results in viremia that persists for weeks to months with shedding of virus in urine, saliva, vaginal fluid, and semen that can persist for months to years. In addition to prolonged viral replication, CMV persists in its human host through latent viral DNA with no detectable virus replication. Intermittent viral shedding in various body fluids can occur in healthy hosts years after initial infection. Impaired immune response leads to reactivation of latent infection and diminishes the host's ability to limit CMV replication. Newborns infected prenatally, patients undergoing organ transplantation, chemotherapy for cancer, or rheumatologic disease, with HIV or congenital immunodeficiency, all have difficulty controlling CMV infection and are at risk for serious multisystem CMV disease.

Epidemiology of CMV Infection

CMV is ubiquitous in human populations. Higher prevalence of infection and younger age at acquisition of CMV infection have been associated with low socioeconomic status, origin in a developing country, nonwhite race, and geographic region in the United States with higher prevalence in the South [2]. Differences in age-related prevalence of CMV infection are probably related to living circumstances, child-rearing practices, and social customs. Transmission of CMV from person to person requires contact with body fluids as occurs with breast-feeding, intimate (sexual) contact, or care of young children. High rates of CMV infection have been noted in children who attend daycare centers and their parents, in daycare workers, sexually active adolescents, and clients of clinics for sexually transmitted infections. In the United States and in Western Europe, a relatively high proportion of women are CMV sero negative when they reach reproductive age compared with countries in Asia and Africa in which almost all young women are CMV sero positive. Table 1 compares seroprevalence rates of CMV in women of childbearing age from different regions of the world.

Table 1.

Prevalence of CMV Infection in Women of Childbearing Age

Study Location	Population	N	Prevalence (%)
Ankara, Turkey [33][1]	women 15–49 years	745	99
Cotonou, Benin	pregnant women	211	97
Seoul, South Korea	prenatal clinic	575	96
Sendai, Japan	prenatal clinic	10,218	95
Sao Paulo, Brazil [34][2]	pregnant women, middle SES	427	67
	pregnant women, lower SES	179	84
Northern Italy	women, pregnant or hospital patients	12,568	77
Helsinki, Finland	Women, prenatal clinics	1,088	71
Birmingham, USA	prenatal, middle SES	12,140	54
	prenatal, lower SES	4,078	77
Grenoble, France	women, prenatal clinic	1,018	52

Reprinted with permission from Mocarski, ES, Shenk, S, Pass, RF. Cytomegaloviruses. In: Knipe, DM, & Howley, PM. Field's Virology. 5th ed. Philadelphia: Lippincott Williams & Wilkins, 2007:ch. 69.

Mother-To-Child Transmission of CMV Infection

Transplacental Transmission

CMV is transmitted from mother to fetus in approximately 35% of pregnancies in which a maternal primary infection occurs [3]. Transplacental transmission rates are lower, approximately 20%, with infection in the first trimester, and increase with advancing gestational age to approximately 75% with third trimester infection [4–6]. Maternal CMV infections are usually clinically unrecognized; studies that used screening for CMV antibody to detect primary infections were instrumental in improving our knowledge of them. Congenital CMV infection due to first trimester maternal primary infection is more likely than later gestational infection to cause fetal disease that is apparent in the newborn as signs of congenital infection and results in disability due to central nervous system damage, hearing loss, or impaired vision [7, 8]. CMV can also be transmitted from mother to fetus if the mother had CMV infection in the past and was immune at the time of conception (recurrent infection). Little is known about the biology of these infections or why maternal immunity from previous infection is unable to prevent all congenital infections. CMV infection is chronic; the human host never eliminates the virus, and persistent infection or reactivation of latent virus is possibly involved. There is evidence that reinfection with a CMV strain with slightly different envelope glycoprotein epitopes from the initial strain results in maternal reinfections that can lead to congenital infection [9]. If reinfection is the major cause for congenital infection in babies of mothers with prior immunity, it would mean that a successful strategy for prevention of primary maternal infection might also prevent reinfections. Rates of congenital CMV infection due to recurrent maternal infection vary widely among populations, ranging from approximately 0.1%–1% or higher; they are generally higher in populations in which the overall prevalence of CMV infection is high, and they are lower in populations with low CMV prevalence rates in women of childbearing age.

Intrapartum Transmission

CMV MTCT also occurs during birth and is due to presence of virus in the cervix or vagina. Cervicovaginal shedding of virus is common in women who are CMV antibody positive. Study of pregnant women that included testing for vaginal shedding at term and testing the newborn for CMV at birth and then again at intervals during the first few months after birth estimated that approximately 50% of babies who were not breastfed and were born to mothers who had positive vaginal CMV cultures at birth acquired the virus [10]. These babies usually begin to shed virus at 3–6 weeks of age. Active CMV infection with cervicovaginal shedding of virus is more common in women infected with HIV, especially those with poor control of HIV infection and low CD4 T cell counts, and CMV infection rates are high in their babies [11, 12].

Breast Milk Transmission

This mode of CMV transmission is by far the most common means of MTCT; it accounts for the majority of infants acquiring CMV in the first year of life in cultures in which most mothers are CMV antibody positive and breastfeed their infants [13]. Breast milk MTCT is arguably the most important means of spread of virus in terms of global impact on population prevalence of CMV. The epidemiologic impact of breast milk MTCT is amplified by the fact that infants who acquire CMV in the first year of life will shed virus in saliva and urine for years, and they have ample opportunity to transmit CMV to caregivers and other children. Using polymerase chain reaction, CMV DNA can be detected in the milk of more than 95% of CMV antibody positive mothers [14, 15]. Maternal immunity does not prevent virolactia nor does it prevent transmission of CMV through milk to the newborn. However, it seems to prevent CMV disease from mother's milk. Infants who are breastfed by CMV antibody positive mothers are CMV sero positive at birth from transplacental transfer of antibody, and, in general, an infant breastfed by its CMV-sero positive mother will not have any illness with CMV infection. However, if maternal, fetal, or newborn immune function is deficient or if the newborn is a very low birth weight and premature, disease in the infant could occur.

Clinical and Public Health Significance of CMV MTCT

Babies who acquire CMV intrapartum or from their mother's milk usually have no acute illness, and these infections are not known to cause the central nervous system and sensory disabilities associated with congenital infection. CMV MTCT to babies of HIV-positive mothers has been associated with increased risk of HIV transmission and with more rapid progression of HIV infection in the infant [16]. Very low birth weight newborns who do not have congenital CMV infection but acquire CMV during birth or from mother's milk can have acute illness associated with the onset of infection that is characterized by a sepsis-like picture, worsening of respiratory illness, hematological changes, abnormal liver function, and cholestasis [17, 18].

The public health significance of human CMV is largely due to infection during pregnancy with transmission of virus to the fetus. Congenital CMV infection is present in approximately 0.7% of newborns and seems to be the most common congenital infection in developed countries [19]. Congenital CMV infection is a leading cause of sensorineural hearing loss, and it is an important cause of developmental disability, cognitive impairment, cerebral palsy, and impaired vision. Approximately half of newborns with congenital CMV infection who have signs of intrauterine infection at birth will have disabilities as will approximately 5%–15% of those who are asymptomatic at birth [19]. The disease burden from congenital CMV infection is greater than that of Down's syndrome and spina bifida combined; it has been estimated that approximately 6000 children born in the United States each year will have central nervous system, auditory, visual, or motor disability due to congenital CMV infection [20].

Prevention of MTCT of CMV

Efforts to prevent congenital CMV infection are clearly merited. In reviewing priorities for vaccine development for the 21st century, the Institute of Medicine listed prevention of congenital CMV infection as a top priority based on healthcare costs savings and improvement in quality-adjusted life years [21]. CMV MTCT during vaginal birth or from breast-feeding is not associated with disease except possibly in special circumstances. Population-based efforts to interrupt these modes of transmission are not merited and would likely be harmful. The frequency and clinical significance of CMV MTCT to very low birth weight newborns varies widely among neonatal units, and there is no consensus on the approach to prevention of these infections. Prevention of congenital CMV infection can be considered in terms of prevention of maternal infection and prevention of transmission to the fetus if maternal infection has occurred.

Prevention of Maternal CMV Infection

Providing care to young children and intimacy with a partner or spouse involve contact with body fluids of another person and are the 2 best characterized risks for CMV infection in young women. These are also a normal part of family life and are activities that one could expect young women to engage in frequently. Thus, preventing maternal CMV infection by limiting exposure will be difficult, but it has been tried with some success. Results of small studies in the United States and in France suggest that providing CMV antibody negative pregnant women with information about CMV risk and methods of prevention could reduce the rate of maternal infection [22, 23]. The Centers for Disease Control and Prevention (CDC) in the United States has made specific recommendations for steps for pregnant women to take to prevent acquisition of CMV infection; these recommendations focus on avoidance of contact with body fluids from children, and the importance of hand washing and cleaning surfaces after child care activities associated with body fluids exposure [24]. To date, there have been no published studies of methods to prevent acquisition of CMV during intimate contact. Intuitively, recommending use of condoms, especially with new partners, seems logical because it has been effective with other infections that are transmitted by sexual contact. However, CMV is often present in saliva and could be transmitted by kissing; therefore, condoms might not prevent partner transmission of CMV. Another possible means of transmission of CMV infection to a pregnant woman is through assisted reproductive technology. Recommendations have been made for screening of sperm and ova donors with the aim of preventing this possible route of fetal infection [25, 26]. Aside from the occupational exposure experienced by daycare workers, most CMV transmission to pregnant women probably occurs in the context of family life, which will make public health efforts aimed at limiting exposure very challenging.

There is great interest in CMV vaccine development. A clinical trial of a CMV glycoprotein B vaccine reported evidence of efficacy for prevention of CMV infection in young mothers, although overall efficacy was only approximately 50%, and it appeared to decline significantly within a few years [7]. The results of this randomized, placebo-controlled, phase 2 clinical trial suggested that vaccine prevention of maternal infection is achievable, and a number of pharmaceutical companies known for vaccine development currently have active CMV vaccine development programs. However, a licensed CMV vaccine for prevention of maternal or congenital infection is not on the near horizon.

Preventing Maternal-Fetal Transmission of CMV

Passive immunization and antiviral treatment of pregnant women with primary CMV infection are being investigated for prevention of fetal infection and for prenatal treatment of the fetus known to be infected. A nonrandomized observational study in Italy in which pregnant women with primary CMV infection were able to accept or reject treatment with CMV hyperimmune globulin (HIG) reported a decrease in fetal infection rate to 16% compared with the 40% transmission rate observed in the women who were not treated [27]. A more recent randomized, placebo-controlled, observer-blind clinical trial of treatment with CMV HIG in women with primary CMV infection (also from Italy) was powered to detect a reduction of similar magnitude in fetal infection rate; it did not confirm efficacy for prevention of maternal-fetal transmission of CMV [28]. In addition, the latter study reported a trend toward an increased rate of premature delivery in women who received CMV HIG. In the absence of proven efficacy and given the possible risks seen in the randomized trial, CMV HIG should not be used for prevention of transmission of CMV outside of a clinical trial. Its use in the setting of proven fetal infection is less clear because there are no ongoing trials likely to provide the answer as to whether CMV HIG ameliorates disease among infected fetuses. It should clearly be considered experimental and should not be offered to women unless they have been thoroughly counseled about the lack of adequate data concerning risks and benefits of CMV HIG for therapeutic purposes. A large randomized, placebo-controlled, observer-blind clinical trial of CMV immune globulin for prevention of maternal-fetal transmission of CMV infection is currently underway (http://clinicaltrials.gov/ct2/show/NCT01376778).

Current Recommendations for Prevention of Maternal CMV Infection

In addition to the CDC's recommendations for preventing CMV infection in pregnant women, professional organizations have made recommendations for specific situations. The American College of Obstetricians and Gynecologists recommends counseling prenatal patients with young children or those who work with young children about the risk of CMV infection and methods for prevention, including hand washing and use of gloves with exposure to body fluids [29]. The American Academy of Pediatrics, the American Public Health Association, and the National Resource Center for Health and Safety in Child Care collaborated to publish Caring for Our Children, National Health and Safety Performance Standards: Guidelines for Out-of-Home Child Care Programs, which includes standards for staff education on increased risk of CMV infection and hygiene measures for reducing the risk of acquiring CMV [30]. The CDC has also made recommendations for preventing transmission of CMV to childcare workers [31].

Conclusions and Gaps in Knowledge

Although a vaccine that would stimulate lasting immunity and prevent CMV infection would solve the problem of congenital infection, none will be available for years. It makes sense to examine public health measures that might decrease rates of maternal infection with the goal of optimizing their effectiveness. Interventions that can decrease the risk of transmission of CMV to the fetus when primary maternal infection occurs during pregnancy are desperately needed now and will continue to be needed even if an effective vaccine is available. Table 2 lists current gaps in knowledge that merit study and could result in decreased rates of maternal and congenital CMV infection.

Table 2.

Gaps in Knowledge Related to Prevention of Maternal and Congenital Cytomegalovirus (CMV) Infection That Can Be Addressed With Well Designed Research Programs

Problem or Question	Studies Needed
1) What is the best strategy for a woman to prevent acquisition of CMV from her own child?	More rigorous testing of preventive messages in large study populations Determine the best approach to dissemination of this message
2) How can we increase the role of healthcare providers in prevention of maternal CMV infection?	Determine ways to incorporate CMV counseling and education into routine prenatal care Test the feasibility and impact of having pediatricians provide education on CMV prevention to young mothers and to youth
3) Child-to-child transmission of CMV in daycare centers	Develop and test strategies for preventing child-to-child transmission of CMV, especially in children from 1 to 4 years of age
4) High risk of CMV infection in daycare workers	Develop and test strategies for prevention of child-to-worker transmission of CMV in daycare centers Improve the performance of daycare centers in providing information on congenital CMV infection and occupational risk of infection to their workers
5) Maternal CMV infection from intimate partner	Develop and test strategies for prevention of CMV transmission from an intimate partner (discordant couple studies)
6) Can transmission of CMV to the fetus be prevented after primary CMV infection in the mother?	Determine whether passive immunization of pregnant women after primary CMV infection will prevent transmission to the fetus Determine whether antivirals can be used safely in pregnant women to prevent transmission of CMV to the fetus
7) Why do women who are immune to CMV before conception transmit CMV to the fetus, and can we determine when transmission occurs?	Develop strategies to determine the source of virus for these events Develop a test for a marker in maternal blood of fetal CMV infection
8) Can CMV vaccine be used to prevent maternal and congenital CMV infections?	Continue to address this question with phase 1, 2, and 3 studies of improved CMV vaccines