OVERVIEW: What every practitioner needs to know

Are you sure your patient has cytomegalovirus? What are the typical findings for this disease?

The presentation of cytomegalovirus (CMV) infection in pediatric practice depends on the age of the child and timing of infection. The most important impact of CMV infection is caused by congenital CMV, which may be symptomatic or asymptomatic at birth. Acquired CMV infections are typically asymptomatic but can produce undifferentiated febrile syndromes and heterophile-negative mononucleosis. The clinical presentation of these forms of CMV infection is summarized below:

Symptomatic Congenital CMV Infection


Continue Reading





Intrauterine growth retardation

Sensorineural hearing loss (SNHL)

Neurodevelopmental disabilities: mental retardation, cerebral palsy, seizures

Asymptomatic Congenital CMV Infection

Usually the infant appears normal; hence, asymptomatic infections are missed.

Approximately 10% of infants will have sensorineural hearing loss that may not be evident at birth.

Subtle long-term neurodevelopmental disabilities may occur.

Acquired CMV Infection

CMV infection is commonly acquired postnatally from breastfeeding.

Breast milk–acquired infections are asymptomatic in healthy term infants.

Primary infections are often acquired by toddlers in group day care settings.

CMV is sexually transmitted in adolescents and young adults.

Usually asymptomatic in adults, CMV may cause “heterophile-negative” mononucleosis.

What other disease/condition shares some of these symptoms?

A number of so-called TORCH (toxoplasmosis, other agents, rubella, cytomegalovirus, herpes simplex) infections can mimic congenital CMV. They include congenital rubella syndrome, congenital toxoplasmosis, and congenital lymphocytic choriomeningitis virus infection.

In young children with primary CMV infection, symptoms may mimic those associated with many viral infections, such as fever, sore throat, and malaise.

In adolescents and young adults with primary CMV infection, a mononucleosis syndrome may be noted. It is generally indistinguishable from Epstein-Barr virus (EBV) mononucleosis, although it is more common for splenomegaly to be associated with CMV mononucleosis. Rash associated with administration of β-lactam antibiotics (particularly amoxicillin as well as related agents) may occur with CMV mononucleosis, as with EBV mononucleosis.

Factors associated with congenital cytomegalovirus transmission

CMV infection of the fetus likely occurs through a transplacental route of infection. Infection of the fetus may occur as a result of primary maternal infection or reinfection with a new strain of CMV. Infection of the fetus occurs more commonly in the context of primary maternal infection (~40%) than in the context of recurrent maternal infection (rate of transmission of CMV in women with preconception immunity is approximately 0.7%).

Preexisting maternal immunity may lessen the risk of neurodevelopmental sequelae in infants who are congenitally infected.

What laboratory studies should you request to help confirm the diagnosis? How should you interpret the results?

Many clinicians mistakenly believe that “TORCH titers” provide useful diagnostic information, but they are seldom of any value in the diagnosis of congenital CMV infection.

The cornerstone of diagnosis is the use of virologic studies

  • Urine or saliva culture

  • Polymerase chain reaction (PCR) studies of urine, saliva, and blood

  • CMV-specific IgM studies are occasionally positive

Timing of specimen acquisition is critical in interpreting studies; virologic studies must be obtained in the first 14 days of life to make the diagnosis of congenital infection, since the finding of CMV after this time point may reflect natal transmission by breast milk.

Other important laboratory studies include complete blood count, platelet count, and serum transaminase levels.

What laboratory studies should you request to help confirm the diagnosis of natal cytomegalovirus?

Natal CMV infections may be acquired in the birth canal during the process of vaginal delivery but are much more commonly acquired by breastfeeding. Acquistion of CMV from breast milk is of no clinical significance in healthy term babies, and diagnostic studies are not needed.

In low–birth weight premature infants, CMV acquired from breast milk or from blood transfusion (with blood not subjected to leukofiltration) can cause serious symptomatic disease and diagnostic studies are warranted:

  • Saliva, tracheal aspirates, urine for viral culture and PCR studies

  • Blood PCR for quantification of viral load

It may be difficult to exclude the diagnosis of congenital CMV in some infants.

What laboratory studies should you request to help confirm the diagnosis of cytomegalovirus infections in adolescents and young adults?

The vast majority of CMV infections in this age group are asymptomatic and diagnostic studies are not warranted or necessary.

CMV may be associated with “heterophile-negative” mononucleosis.

CMV IgG and IgM studies may occasionally be useful when a definitive diagnosis is desirable.

Role of viral culture and PCR in making this diagnosis is unclear.

There may be a role in using CMV serologic tests to define the CMV infection status of pregnant women, or of women considering pregnancy.

There is no clear-cut role for routine screening for CMV antibodies during pregnancy.

Women who work in group day care and mothers of young children attending group day care may be at high risk, and serologic screening and monitoring can be considered.

Health care providers are not at increased risk for acquiring CMV infection, and there is no role for screening or work reassignment based on serostatus.

Demonstration of seroconversion and/or appearance of CMV IgM during pregnancy may represent a high-risk situation.

In addition to seroconversion (IgG studies) and demonstration of IgM antibodies, determination of the CMV IgG avidity index may be an important study predicting risk of transmission to the fetus (high avidity correlates with decreased risk of transmission).

Would imaging studies be helpful? If so, which ones?

Imaging studies are of great importance in the diagnosis of congenital CMV involving the central nervous system (CNS) and help to predict prognosis.

Imaging may be performed using ultrasonography or computed tomography (CT) of the head.

Important findings in CNS imaging in the neonatal period include the following (Figure 1):

Figure 1.

CNS imaging: Congenital CMV ultrasonogram (upper panel) and MRI (lower panel) of neonate with congenital CMV infection demonstrating lissencephaly, ventriculomegaly, and periventricular calcifications (arrow).

  • Microcephaly

  • Intracranial calcifications (classically with periventricular distribution)

  • Lissencephaly

  • Polymicrogyria

  • Other structural anomalies

Subtle abnormalities may not be demonstrable by imaging until magnetic resonance imaging can be performed later in childhood.

The absence of intracranial calcifications by ultrasonography and/or CT does not preclude the possibility of CNS injury and long-term sequelae.

If you are able to confirm that the patient has cytomegalovirus infection, what treatment should be initiated?

Treatment of cytomegalovirus infections in pediatric practice

Symptomatic Congenital CMV Infection

The antiviral drug most commonly used in CMV infections in infants and children is ganciclovir. Ganciclovir is a nucleoside analog that inhibits viral DNA replication.

For symptomatic congenital CMV infection involving the CNS, a 6-week course of ganciclovir has been demonstrated to reduce the risk of SNHL and possibly other neurodevelopmental sequelae.

  • Recommended dose: 6 mg/kg every 12 hours

  • Recommended duration of therapy: 6 weeks

Valganciclovir (valine ester of ganciclovir) is less well studied than ganciclovir, but it appears to be an acceptable orally bioavailable therapy for treatment of symptomatic congenital CMV.

Optimal dose uncertain but 16 mg/kg every 12 hours appears to be appropriate

Recommended duration of therapy: 6 weeks (can be initiated as intravenous ganciclovir and switched to oral valganciclovir)

Long-term therapy with oral valganciclovir (~6 months) may further reduce the risk of SNHL and CNS sequelae, and this therapy is currently being evaluated in a controlled trial sponsored by the National Institutes of Health.

The lack of any data regarding long-term therapy (i.e., 6 months) with oral valganciclovir in the setting of symptomatic congenital CMV infection precludes definitive recommendations at this time; consultation with an expert in the treatment of perinatal viral infection is recommended.

Symptomatic Postnatally Acquired CMV Infection

Indications for ganciclovir therapy in postnatally acquired CMV infections (symptomatic disease in premature infants) are not clear.

Ganciclovir is warranted for severe disease, including pneumonia and severe hepatitis.

Ganciclovir can be considered for symptomatic infants with recalcitrant thrombocytopenia or high-grade viremia.

Duration of therapy is 14-21 days; longer if end-organ disease persists.

There is no clear role for the use of CMV immune globulin in this setting.

Other CMV Infections

Ganciclovir is indicated for prophylaxis and treatment of CMV infections in immunocompromised solid organ and hematopoietic stem cell transplant patients.

Licensed alternative agents for immune compromised patients include cidofovir and foscarnet; use of these agents is typically used in the setting of ganciclovir side effects or identification of ganciclovir resistance.

CMV immune globulin may be considered in pregnant patients in the setting of documented fetal infection (amniocentesis confirmation or ultrasonographic abnormalities with serologic and/organ virologic evidence of primary maternal infection).

There is no indication for antiviral therapy in children or adolescents with primary CMV infection, including CMV mononucleosis.

What are the adverse effects associated with each treatment option?

Ganciclovir therapy carries a significant risk of neutropenia

Almost two thirds of infants had neutropenia when treated with ganciclovir for congenital CMV infection.

Neutropenia is readily reversible with discontinuation of therapy.

If continued treatment is desirable in this setting, administration of granulocyte colony-stimulating factor can correct neutropenia and allow therapy to continue.

Ganciclovir can also be associated with bone marrow suppression of other cell lineages (i.e., anemia, thrombocytopenia).

Treatment with cidofovir or foscarnet can be associated with substantial toxicity, particularly nephotoxicity, with associated alterations in fluid and electrolyte homeostasis.

What are the possible outcomes of cytomegalovirus infection?

Cytomegalovirus infections are ubiquitous and virtually always asymptomatic in healthy individuals.

It is important to emphasize the common and latent nature of CMV with patients and families who generally will have never heard of the virus.

Acquisition of infection is lifelong and the analogy with other Herpesviridae (i.e., varicella zoster, herpes simplex) can be stressed.

CMV mononucleosis has a variable clinical course, just as EBV mononucleosis does, and although some restrictions are appropriate for some patients (e.g., decreased school/work workload, avoidance of contact sports if splenomegaly present), the prognosis for complete recovery is excellent.

In immunocompromised transplant patients (solid organ and hematopoietic stem cell), CMV can cause life-threatening end-organ disease, is a contributing factor in graft-versus-host disease, and increases the risk for opportunistic fungal infections and graft failure.

In the pregnant patient, the outcome of CMV infection for the fetus depends on maternal immune status before establishment of pregnancy.

For primary maternal infections, the risk of transmission to the fetus is approximately one third (32% in a recent meta-analysis), whereas for recurrent maternal infection, the risk of transmission is 1.4%.

In the infected fetus, the long-term outcome of infection depends on whether or not the infant demonstrates symptoms at birth.

Approximately 10% of congenitally infected infants have signs and symptoms of disease at birth.

Symptomatic infants have a 40%-90% risk of subsequent neurologic sequelae, including mental retardation, microcephaly, developmental delay, seizure disorders, and cerebral palsy.

Ninety percent of congenitally infected infants are asymptomatic at birth.

A range between 7% and 20% of asymptomatically congenitally infected infants has been reported for the risk of permanent sequelae, the most common sequela being sensorineural hearing loss.

Epidemiology of cytomegalovirus and disease trends

Seroprevalence is highest in the developing world, where infection is acquired early in life. There is no seasonal variation.

Prevalence in the developed world is higher in African Americans and those of lower socioeconomic status.

Infections in the developed world are common in group day care, where up to 70% of toddlers may shed virus in urine.

Congenital CMV infection rates track directly with seroprevalence in the population.

The higher the maternal seroprevalence of CMV antibodies in the population, the greater the rate of congenital CMV infection.

Most of these infections will be reinfections.

The biggest impact of congenital CMV with respect to childhood neurodevelopmental disability (particularly hearing loss) is in infants born to women with preconception antibodies to CMV.

This issue complicates vaccine development.

Newborn screening of saliva or dried blood spots may provide better insights into the current epidemiology of congenital infection.

Maternal serologic screening during pregnancy is controversial and not currently recommended by the American College of Obstetrics and Gynecology.

Host/pathogenic interactions and impact of genetics

CMV encodes genes that induce apoptosis, evade host immune responses, and modulate the cell cycle; these genes may play a role in the pathogenesis of infection.

There is no clear evidence for increased genetic susceptibility to severe CMV infection, but some suggestion that Toll receptor polymorphisms are associated with more severe disease after solid organ transplantation.

CMV may contribute to protracted stays in the intensive care unit for adult patients, may play a role in the pathogenesis of arterial disease and malignancies, and may be responsible for immunosenescence and impaired immunity in the elderly. These associations are controversial but may provide a rationale for universal vaccination against CMV, once a vaccine is available, and not just selective immunization of women of childbearing age.

Prospects for prevention of cytomegalovirus infection

Handwashing, scrupulous hygienic precautions, and avoidance of infectious secretions (urine, saliva, nasopharyngeal secretions) may protect against both primary infection and reinfection.

Hygienic precautions should be stressed with young women who are pregnant or may be pregnant.

CMV immune globulin for pregnant women with evidence of fetal infection may protect the fetus against CMV disease; this intervention is unproved but available.

CMV is sexually transmitted, and this is another source of acquisition of infection that patients can be educated about.

Leukofiltration of blood products has essentially eliminated transfusion-associated CMV infection and disease.

A number of vaccines against CMV are in clinical trials but none is licensed for use.

A recent study of a subunit glycoprotein vaccine demonstrated 50% protective efficacy in a clinical trial in young women of childbearing age.

What is the evidence?

Weller, TH. “The cytomegaloviruses: ubiquitous agents with protean clinical manifestations”. N Engl J Med. vol. 285. 1971. pp. 203-14, 267-74.

Cohen, JI, Corey, GR. “Cytomegalovirus infection in the normal host”. Medicine (Baltimore). vol. 64. 1985. pp. 100-14.

Horwitz, CA, Henle, W, Henle, G. “Clinical and laboratory evaluation of cytomegalovirus-induced mononucleosis in previously healthy individuals. Report of 82 cases”. Medicine (Baltimore). vol. 65. 1986. pp. 124-34.

Bravender, T. “Epstein-Barr virus, cytomegalovirus, and infectious mononucleosis”. Adolesc Med State Art Rev. vol. 21. 2010. pp. 251-64.

Cheeran, MC, Lokensgard, JR, Schleiss, MR. “Neuropathogenesis of congenital cytomegalovirus infection: disease mechanisms and prospects for intervention”. Clin Microbiol Rev. vol. 22. 2009. pp. 99-126.

Demmler-Harrison, GJ. “Congenital cytomegalovirus: public health action towards awareness, prevention, and treatment”. J Clin Virol. vol. 46. 2009. pp. S1-5.

Arvin, AM, Fast, P, Myers, M. “Vaccine development to prevent cytomegalovirus disease: report from the National Vaccine Advisory Committee”. Clin Infect Dis. vol. 39. 2004. pp. 233-9.

Kimberlin, DW, Lin, CY, Sánchez, PJ. “Effect of ganciclovir therapy on hearing in symptomatic congenital cytomegalovirus disease involving the central nervous system: a randomized, controlled trial”. J Pediatr. vol. 143. 2003. pp. 16-25.

Söderberg-Nauclér, C. “Does cytomegalovirus play a causative role in the development of various inflammatory diseases and cancer”. J Intern Med. vol. 259. 2006. pp. 219-46.

Osawa, R, Singh, N. “Cytomegalovirus infection in critically ill patients: a systematic review”. Crit Care. vol. 13. 2009. pp. R68

Fowler, KB, Stagno, S, Pass, RF. “The outcome of congenital cytomegalovirus infection in relation to maternal antibody status”. N Engl J Med. vol. 326. 1992. pp. 663-7.

Sung, H, Schleiss, MR. “Update on the current status of cytomegalovirus vaccines”. Exp Rev Vaccine. vol. 9. 2010. pp. 1303-14.

Schleiss, MR. “Congenital cytomegalovirus infection: molecular mechanisms mediating viral pathogenesis”. Infect Disord Drug Targets. vol. 11. 2011. pp. 449-65.