Healthline Blogs

Fruit of the Womb
Fruit of the Womb

Cytomegalovirus (CMV) Reprise

There is not a week that goes by in my practice that I do not have to address questions regarding cytomegalovirus (CMV) infections and pregnancy. Since we have also been discussing recurrent pregnancy loss in various contexts over the past year, I thought her questions were particularly timely and would be of interest to a good number of other readers. This is a good opportunity to reprise some of the information about CMV that we have provided in the past…

Anonymous has left a new comment on your post "Cytomegalovirus (CMV): Common and Confusing":

I have had reoccurring miscarriages and my OB did a TORCH blood test which showed my CMV levels were slightly elevated. I’m supposed to go back in two weeks for another test to recheck the levels. Could this virus have caused my miscarriages? Is there a period of time I should wait before trying to conceive again? If I do become pregnant again am I at risk of having a baby affected by CMV

To Anonymous: All of your questions are answered at the very end of this post. You can jump to there if you would like or read on for more information…

The TORCH test looks for antibodies in your blood specific for TOxoplasmosis, Rubella (German measles), Cytomegalovirus, and Herpes simplex virus. It is not unusual to include a screen for syphilis and human immunodeficiency virus (HIV) at the same time. All of these can cause infections of a baby during pregnancy (congenital infections) but by far, CMV is the most common. If antibodies to a specific organism(s) are present, this indicates you have been infected with that organism at some time in the past. Let me elaborate on that point a bit so that you have a better understanding of the tests when you discuss them with your doctor.

Usually, when we contract a virus infection like the flu, our immune systems react by first producing specific antibodies of the IgM class. These usually hang around only for no more than 2-4 weeks after the infection has been cleared. Shortly after IgM antibodies begin to be made, our bodies switch to the production of a second class of specific antibodies called IgG. IgG antibodies generally hang around for a long time after the infection is cleared and provide us with a source of 'permanent immunity' to the organism, helping to prevent reinfection, or decrease the severity of a secondary infection, with the same or similar organisms with which the antibodies might 'cross-react.' It is these IgG antibodies that also afford protective immunity to the baby because they can cross the placenta whereas IgM antibodies cannot.

We can use this information to help us to characterize the status of an infection. If neither IgM nor IgG is present, the individual has probably never been exposed to the organism of concern (or is too early in the course of the infection to have mounted any antibody response). If IgM is present and there is no IgG, then the infection is probably a 'primary' infection, indicating first time exposure to the organism, usually very early in its course. If both IgM and IgG are present, this also usually reflects a primary infection, but later in the course of the disease. And, if only IgG is present, then this indicates a state of permanent immunity established from an infection that occurred at some time in the past. In any of the first three circumstances, if an infection with a specific organism is suspected and could be of concern for a pregnancy, it is probably worth repeating the antibody titers in 4-6 weeks.

Things can be a little confusing with CMV infections. CMV grows very slowly and the incubation period from the time of exposure to onset of symptoms, or asymptomatic excretion of the virus, is on the order of 4-12 weeks. IgG antibodies usually can be detected by 1-2 weeks after the onset of symptomatic infections, but because of the long incubation period of CMV, this might be a month or more after actual exposure to the virus. The presence of IgM in the absence of IgG and in the presence of symptomatic disease (or a history of recent exposure to a known carrier) is highly presumptive of a true 'primary infection.' Similarly, the findings of both IgM and IgG with a significant rise (four-fold or more) in IgG titers (with or without a fall in IgM titers) on a follow-up screen 4-6 weeks later, usually (but not always in the case of CMV) indicates a recent primary infection. Also useful, the presence of IgG in the absence of IgM, is highly suggestive of a remote exposure to the virus, often greater than 6 months previously.

The rub with serology in classifying CMV infections comes in most often when IgM is present but IgG titers are relatively stable or mildly fluctuating. Unlike most common viral infections, CMV-specific IgM can sometimes persist 6-9 months following its appearance. And to make things even more confusing, IgM has been found to reappear on occasion with reactivation of latent CMV infections. In other words, except in the circumstances detailed above, we may not be able to use the presence of IgM in our counseling to tell patients that they have had a primary or recurrent infection during the pregnancy or if the infection might have occurred even prior to the current pregnancy.


CMV is the MOST common congenitally (fetal) and perinatally (newborn) acquired virus disease in humans and the single most important infectious cause of mental retardation and congenital deafness in the U.S. and other industrialized nations. CMV is a member of the herpes family and human CMV is restricted to humans with no known animal reservoir. At least 80-90% of all individuals are infected with CMV during their lives, but infection may occur in the absence of clinical symptoms or without recognizing that an illness is the result of CMV. As with other herpes viruses, once an individual is infected with CMV, 'recurrences' can result from periodic reactivation of virus replication at various sites in which the virus is latently harbored in the body. During periods of reactivation, the virus can be more readily transmitted to individuals who have not been previously infected. Although probably not that common, secondary infections can also occur with other strains of CMV.

Transmission of CMV can occur from exposure to just about any body fluid, most commonly via saliva, respiratory, and venereal routes or by contact with infected urine or breast milk. Exposure tends to occur at earlier ages in lower socioeconomic groups, promiscuous individuals, and children at day care centers. Antibodies to CMV (seropositivity), indicating a history of infection with the virus, in women during the common childbearing years (18-35) can be found in about 50% of those in middle and upper, and 90% in lower, socioeconomic groups. Among previously noninfected (seronegative) women, the chance of becoming infected with the virus (seroconversion) is about 1-3% per year, however, this is as high as 10-20%in women who work in day care settings and 50% in women with infected children under two years of age!

Congenital infections with CMV occur in 1-2% of ALL pregnancies, accounting for 40-50,000 cases per year, indicating the virus can cross the placenta with relative ease compared to other herpes and most other common virus infections. Transmission of CMV to the fetus can occur with both primary and recurrent infections despite maternal immunity and has been documented in consecutive pregnancies. However, congenital infection rarely results in a poor outcome unless the congenital infection is the result of a primary (first time) infection in the mother during the pregnancy. Overall, among seronegative women, there is about a 0.5-1% risk of a primary CMV infection in each pregnancy, although this is much higher in 'at risk' groups such as day care workers, health care providers, and women with young children. Primary maternal infections carry an overall transmission risk to the baby of 25-50% and recurrent infections about 2-3%. Maternal antibody to CMV is incompletely protective against transmission to the baby, but it does play a major role in reducing the severity of infection in both fetus and newborn.

A high percentage of primary maternal infections are asymptomatic, or simply confused with another illness, or even written off as normal symptoms of pregnancy, so usually the diagnosis is not considered. Indeed, perhaps, the hardest part of making the diagnosis of CMV infection during pregnancy in mother and baby is simply suspecting that it might be a problem. If a woman develops a prolonged (weeks) illness with fever, resembling infectious mononucleosis (caused by another herpes virus, Epstein-Barr virus (EBV)), and the latter cannot be confirmed by routine testing, then CMV should be high on the differential diagnosis. The ante goes up considerably if the mother is a young teen having her first baby (especially if she conceived shortly after becoming sexually active or changing partners), or if she works, or has another young child, in a high risk setting such as a daycare center.

Serologic testing, as we discussed at the outset, can often help to confirm a primary infection when it is obtained coincident with maternal symptoms. Virus detection by culture, immunofluorescent, or polymerase chain reaction (PCR) techniques, best done by sampling maternal urine, may be positive (and almost always will be for months after a primary infection), but this alone does not tell us if this is a primary symptomatic CMV infection or a recurrent infection accompanying another illness. Unfortunately, in most cases maternal infection with CMV is usually not suspected until the baby is found to be growth restricted or has subtle physical abnormalities, or a thickened placenta, to suggest it has either a chromosomal abnormality or a congenital infection, sometimes months after the maternal exposure. Under these circumstances, maternal serologic testing might not be helpful in establishing either the fetal diagnosis or the diagnosis of a primary maternal infection during pregnancy.

Congenital CMV infections associated with primary maternal infections early in pregnancy and accompanied by growth restriction and detectable abnormalities by fetal ultrasound, generally, have a very poor prognosis, but even then, the outcome is not entirely predictable. Establishing the presence of fetal infection, when fetal abnormalities are identified, requires an 'invasive' procedure. Performing a simple amniocentesis, and using a CMV-specific PCR or shell vial culture technique, can confirm fetal infection with CMV in nearly 100% of cases. (Remember, the amniotic fluid from midtrimester on is mostly fetal urine, and CMV is excreted in large amounts from the kidneys following congenital infection, and often for years afterwards, even in 'asymptomatic' cases). However, unless we have confirmatory maternal serologic information, or symptomatic infection confirmed to be the result of CMV during the pregnancy, we still may not know if the congenital infection is the result of primary or recurrent maternal disease. If the diagnosis of CMV is not suspected until late in pregnancy, or not until after the birth of the baby, detection of CMV in a urine sample taken from the baby within the first two weeks'(preferably, the first 48-72 hours) of life also suffices to confirm congenital infection. CMV is a VERY slow growing virus, so any detection of virus in this time frame most certainly represents intrauterine infection.

Congenital CMV infections are asymptomatic or unrecognized in about 90% of cases, including 85-90% of babies acquiring the virus as a consequence of primary maternal infection and 99% of those resulting from recurrent infections. Most of the asymptomatic infections pose no immediate life threat, but 10-15% of these babies are at risk for long-term complications such as sensorineural hearing loss, chorioretinitis, and dental abnormalities, usually apparent by two years' of age. Of these, the hearing loss is by far the most significant because delay in its detection can contribute to psychomotor retardation. More than 90% of congenitally infected infants, regardless of the severity of their infection, will shed infectious virus at birth, and may do so for 6 or more years longer, even in the presence of specific immunity.

About 5-10% of congenitally infected babies will have significant evidence of infection at birth. Approximately half of these will have classic 'cytomegalic inclusion disease (CMID)' (described below) and half will have atypical involvement. Virtually 100% of these infants excrete CMV at birth and will continue to do so for many years afterwards. Infants in this group have a 20-30% eventual mortality and account for 90% of the significant mental and psychomotor retardation associated with congenital CMV infections. The most severely affected children are usually the result of primary maternal infections during pregnancy and should be suspected in instances of unexplained intrauterine growth restriction and fetal death beyond 20 weeks' gestation.

The most common findings in babies with CMID include symmetrical growth restriction, multiple small skin hemorrhages (petechiae), enlargement of the liver and spleen (hepatosplenomegaly), jaundice, microcephaly (small head), abnormalities of the dental enamel, and chorioretinitis. They can have numerous associated birth defects involving virtually any organ system, a catalog of which is not necessary for our discussion today. Prenatal detection by ultrasound of fetal growth delay and gross or subtle, nonspecific physical abnormalities of the baby may be the only clues that a congenital CMV infection has occurred. Unfortunately, these often manifest themselves only weeks or even months after fetal infection has occurred because the virus is very slow growing, and it is relatively nondestructive, compared to other herpes viruses.


With the above information as background, let me answer our reader’s questions:

Could this virus have caused my miscarriages? It is unlikely. Although I did not address this specific point above, many people over many years have tried to implicate CMV in recurrent miscarriages and I have never seen an article that found good evidence to do so.

Is there a period of time I should wait before trying to conceive again? The TORCH screen and your CMV seropositivity should not preclude your getting pregnant again soon. You already have ‘protective immunity’ and you will probably have the virus for the rest of your life. However, if you have had recurring miscarriages, perhaps you should wait to conceive until you have had a more thorough evaluation for that.

If I do become pregnant again am I at risk of having a baby affected by CMV? Yes, you have a 2-3% chance of having a baby with a congenital CMV infection. But, the presence of your antibody to CMV will decrease the risk of serious complications from the infection. Most (99%) babies who get CMV as the result of a recurrent maternal infection are asymptomatic, but 10-15% of these babies are at risk for long-term complications such as sensorineural hearing loss, chorioretinitis, and dental abnormalities, usually apparent by two years' of age. Your baby can be screened within 48-72 hours of birth by simply checking his/her urine for CMV to determine if there was a congenital infection. Almost all babies with a congenital CMV infection will excrete the virus in their urine. If CMV is present, the baby simply needs to be followed more closely for the hearing and eye problems noted above to prevent correctible sources of developmental delay. These children usually do not have problems with growth or mental retardation.
  • 1
Was this article helpful? Yes No