Chapter 247 Cytomegalovirus
Human cytomegalovirus (CMV) is widely distributed. Most CMV infections are inapparent, but the virus can cause a variety of clinical illnesses that range in severity from mild to fatal. CMV is the most common cause of congenital infection, which occasionally causes the syndrome of cytomegalic inclusion disease (hepatosplenomegaly, jaundice, petechia, purpura, and microcephaly) in neonates. In immunocompetent adults, CMV infection is occasionally characterized by a mononucleosis-like syndrome. In immunosuppressed persons, including transplant recipients and patients with AIDS, CMV pneumonitis, retinitis, and gastrointestinal disease are common and can be fatal.
Primary infection occurs in a seronegative, susceptible host. Recurrent infection represents reactivation of latent infection or reinfection of a seropositive immune host. Disease may result from primary or recurrent CMV infection, but the former is more commonly associated with severe disease.
CMV is the largest of the herpesviruses and has a diameter of 200 nm with a double-stranded DNA viral genome of 240 kb in a 64-nm core enclosed by an icosahedral capsid composed of 162 capsomers. The core is assembled in the nucleus of the host cells. The capsid is surrounded by a poorly defined amorphous tegument, which is itself surrounded by a loosely applied, lipid-containing envelope. The tegument contains the most immunogenic proteins of the virion, including targets of T-lymphocyte and antibody responses. The envelope is acquired during the budding process through the nuclear membrane into a cytoplasmic vacuole, which contains the protein components of the envelope. The envelope glycoproteins induce strong antibody responses, including neutralizing antibodies in the infected host. Mature viruses exit the cells by cell lysis (fibroblasts) or by poorly defined exocytic pathways. Routine serologic tests do not define specific serotypes. In contrast, restriction endonuclease analysis of CMV DNA shows that human strains are not genetically identical unless obtained from epidemiologically related cases.
Seroepidemiologic surveys demonstrate CMV infection in every population examined worldwide. The prevalence of infection increases with age and is higher in developing countries and among lower socioeconomic strata of the more developed nations.
Transmission sources of CMV include saliva, breast milk, cervical and vaginal secretions, urine, semen, tears, blood products, and organ allographs. The spread of CMV requires very close or intimate contact because it is very labile. Transmission occurs by direct person-to-person contact, but indirect transmission is possible via contaminated fomites.
The incidence of congenital CMV infection ranges from 0.2% to 2.2% (average 1%) of all live births, with the higher rates among populations with a lower economic standard of living. The risk for fetal infection is greatest with maternal primary CMV infection (30%) and much less likely with recurrent infection (<1%). In the USA, 1-4% of pregnant women acquire primary CMV infection, and as many as 8,000 newborns have neurodevelopmental sequelae associated with congenital CMV infection.
Perinatal transmission is common, accounting for an incidence of 10-60% through the 1st 6 mo of life. The most important perinatal sources of virus are genital tract secretions at delivery and breast milk. Among CMV-seropositive mothers, virus is detectable in breast milk in 96%, with postnatal transmission occurring in approximately 38% of infants, resulting in symptomatic infection in nearly half of very low birthweight infants. Infected infants may excrete virus for years in saliva and urine.
After the 1st year of life, the prevalence of infection depends on group activities, with child-care centers contributing to rapid spread of CMV among children. Infection rates of 50-80% during childhood are common. For children who are not exposed to other toddlers, the rate of infection increases very slowly throughout the 1st decade of life. A 2nd peak occurs in adolescence as a result of sexual transmission. Seronegative child-care workers and parents of young children shedding CMV have a 10-20% annual risk of acquiring CMV, in contrast to the risk of 1-3% per year for the general population.
Hospital workers are not at increased risk for acquiring CMV infection from patients. With the implementation of universal precautions, the risk of nosocomial transmission of CMV to health care workers is expected to be lower than the risk of acquiring the infection in the community.
CMV infection may be transmitted in transplanted organs (kidney, heart, bone marrow). Following transplantation, many patients excrete CMV as a result of infection acquired from the donor organ or from reactivation of latent infection caused by immunosuppression. Seronegative transplant recipients of organs from seropositive donors are at greatest risk for severe disease.
Nosocomial infection is a hazard of transfusion of blood and blood products. In a population with a 50% prevalence of CMV infection, the risk has been estimated at 2.7% per unit of whole blood. Leukocyte transfusions pose a much greater risk. Infection is usually asymptomatic, but even in well children and adults there is a risk for disease if the recipient is seronegative and receives multiple units. Immunocompromised patients and seronegative premature infants have a much higher (10-30%) risk for disease.
Clinical CMV disease generally results from a combination of altered cellular immunity, uncontrolled viral replication with increased virus burden, multiorgan involvement, and end-organ disease secondary to direct viral cytopathic effects. Increased levels of virus replication, as ascertained by genome copy numbers, are useful in identifying patients at risk for invasive disease and dissemination of infection. The presence of CMV in areas of inflammation increases the expression of soluble mediators such as cytokines and chemokines, leading to recruitment of inflammatory cells. The interactions between the virus and host inflammatory response seem to lead to persistent viral replication, viral gene expression, and dissemination.
The signs and symptoms of CMV infection vary with age, route of transmission, and immunocompetence of the patient. The infection is subclinical in most patients. In infants and young children, primary CMV infection occasionally causes pneumonitis, hepatomegaly, hepatitis, and petechial rashes. In older children, adolescents, and adults, CMV may cause a mononucleosis-like syndrome characterized by fatigue, malaise, myalgia, headache, fever, hepatosplenomegaly, elevated liver enzyme values, and atypical lymphocytosis. The course of CMV mononucleosis is generally mild, lasting 2-3 wk. Clinical presentations include occasionally persistent fever, overt hepatitis, and a morbilliform rash. Recurrent infections are asymptomatic in the immunocompetent host.
The risk of CMV disease is increased in immunocompromised persons, with both primary and recurrent infections (Chapter 171). Illness with a primary infection includes pneumonitis (most common), hepatitis, chorioretinitis, gastrointestinal disease, or fever with leukopenia as an isolated entity or as a manifestation of generalized disease, which may be fatal. The risk is greatest in bone marrow transplant recipients and in patients with AIDS. Pneumonia, retinitis, and involvement of the central nervous system and gastrointestinal tract are usually severe and progressive. Submucosal ulcerations can occur anywhere in the gastrointestinal tract and may lead to hemorrhage and perforation. Pancreatitis and cholecystitis may also occur.
Symptomatic congenital CMV infection was originally termed cytomegalic inclusion disease. Only 5% of all congenitally infected infants have severe cytomegalic inclusion disease, another 5% have mild involvement, and 90% are born with subclinical, but still chronic, CMV infection. The characteristic signs and symptoms of clinically manifested infections include intrauterine growth restriction, prematurity, hepatosplenomegaly and jaundice, blueberry muffin–like rash, thrombocytopenia and purpura, and microcephaly and intracranial calcifications. Other neurologic problems include chorioretinitis, sensorineural hearing loss, and mild increases in cerebrospinal fluid protein. Symptomatic newborns are usually easy to identify. Congenital infections that are symptomatic and most severe and those resulting in sequelae are more likely to be caused by primary rather than reactivated infections in pregnant women. Re-infection with a different strain of CMV can lead to symptomatic congenital infection. Asymptomatic congenital CMV infection is likely a leading cause of sensorineural hearing loss, which occurs in approximately 7-10% of all infants with congenital CMV infection, whether symptomatic at birth or not.
Infections resulting from exposure to CMV in the maternal genital tract at delivery or in breast milk occur despite the presence of maternally derived, passively acquired antibody. Approximately 6-12% of seropositive mothers transmit CMV by contaminated cervical-vaginal secretions and 40% by breast milk to their infants, who usually remain asymptomatic and do not exhibit sequelae. Occasionally, perinatally acquired CMV infection is associated with pneumonitis and a sepsis-like syndrome. Premature and ill full-term infants may have neurologic sequelae and psychomotor retardation. However, the risk for hearing loss, chorioretinitis, and microcephaly does not appear to be increased. Very low birthweight infants with transfusion-acquired or breast milk–acquired CMV infection have a much greater risk of morbidity.
Active CMV infection is best confirmed by virus isolation from urine, saliva, bronchoalveolar washings, breast milk, cervical secretions, buffy coat, and tissues obtained by biopsy. Rapid identification (within 24 hr) is routinely available with the centrifugation-enhanced rapid culture system based on the detection of CMV early antigens using monoclonal antibodies. Several methods are used for rapid quantitative detection of CMV antigens, and quantitative polymerase chain reaction (PCR) assays are also available. The presence of viral shedding and active infection does not distinguish between primary and recurrent infections. A primary infection is confirmed by seroconversion or the simultaneous detection of immunoglobulin (Ig) M and IgG antibodies with low functional avidity. A simple increase in antibody titers in initially seropositive patients must be interpreted with caution, because such an increase is occasionally observed years after primary infection. IgG antibodies persist for life. For the first weeks after primary infection, the functional avidity of IgG class antibodies is very low, rising to a peak in 4-5 mo. IgM antibodies can be demonstrated transiently in both symptomatic and asymptomatic infection at 4-16 wk, which is during the acute phase of symptomatic disease. IgM antibodies are occasionally found with these assays (0.2-1%) in patients with recurrent infection.
Recurrent infection is defined by the reappearance of viral excretion in a patient known to have been seropositive in the past. The distinction between reactivation of endogenous virus and re-infection with a different strain of CMV requires CMV-DNA analysis or the measurement of antibodies against strain-specific epitopes of CMV, such as glycoprotein H epitopes.
In immunocompromised persons, excretion of CMV, increases in IgG titers, and even the presence of IgM antibodies are common, greatly confounding the ability to distinguish primary and recurrent infections. Demonstrating viremia by buffy coat culture, detection of CMV antigenemia, or detection of CMV DNA by PCR implies active disease and worse prognosis regardless of whether the type of infection is primary, recurrent, or uncertain.
The definitive method for diagnosis of congenital CMV infection is virus isolation or demonstration of CMV DNA by PCR, which must be performed during the 1st 2 weeks of life because viral excretion afterwards may represent infection acquired at birth or shortly thereafter. Urine and saliva are the best specimens for culture and saliva, and cord blood is best for PCR. Infants with congenital CMV infection may excrete CMV in the urine for several years. An IgG antibody test is of little diagnostic value because a positive result also reflects maternal antibodies, although a negative result excludes the diagnosis of congenital CMV infection. Demonstration of stable or rising titers in serial specimens during the 1st year of life does not help, because acquired infection in the first few months of life is common. IgM tests lack sensitivity and specificity and are unreliable for diagnosis of congenital CMV infection.
IgM antibody tests and the measurement of CMV IgG avidity can identify women at high risk for transmitting CMV in utero. Fetal infection can be confirmed by viral isolation from amniotic fluid. The sensitivity of this method is excellent after the 22nd wk of gestation. The detection of viral genome by PCR in amniotic fluid is equally sensitive and specific; quantitative PCR demonstrating 105 genome equivalents per mL of amniotic fluid is a predictor of symptomatic congenital infection.
Ganciclovir, foscarnet, and cidofovir are inhibitors of viral DNA polymerase shown to be effective in CMV disease and are approved for use in the USA. Treatment is seldom indicated for immunocompetent persons but is recommended for immunocompromised persons and remains controversial for infants with symptomatic congenital infection.
The more severe the immunosuppression, such as that required after bone marrow transplantation, the more severe is the CMV disease. Ganciclovir combined with immune globulin, either standard intravenous immunoglobulin (IVIG) or hyperimmune CMV IVIG, has been used to treat life-threatening CMV infections in immunocompromised hosts (bone marrow, heart, and kidney transplant recipients and patients with AIDS). Two published regimens are: ganciclovir (7.5 mg/kg/day divided every 8 hr IV for 14 days) with CMV IVIG (400 mg/kg on days 1, 2, and 7, and 200 mg/kg on day 14); and ganciclovir (7.5 mg/kg/day divided every 8 hr IV for 20 days with IVIG 500 mg/kg every other day for 10 doses).
CMV retinitis and gastrointestinal disease appear to be clinically responsive to therapy but, like viral excretion, often recur on cessation. Toxicity with ganciclovir is frequent and often severe and includes neutropenia, thrombocytopenia, liver dysfunction, reduction in spermatogenesis, and gastrointestinal and renal abnormalities. Oral valganciclovir, the orally bioavailable prodrug of ganciclovir, causes less toxicity and appears to be as effective as IV ganciclovir. Foscarnet is an alternative antiviral agent, although there is limited information on its use in children. CMV prophylaxis with ganciclovir or acyclovir reduces the risk of morbidity in solid organ transplantation. Prophylactic treatment with valacyclovir in adults (900 mg PO once daily for 90 days) is a safe and effective regimen to prevent CMV disease in kidney and pancreas transplant recipients.
Patients with CMV mononucleosis usually recover fully, although some have protracted symptoms. Most immunocompromised patients also recover uneventfully, but many experience severe pneumonitis, with a high fatality rate if hypoxemia develops. CMV infection and disease may be fatal in individuals with increased susceptibility to infections, such as patients with AIDS.
A randomized controlled study with ganciclovir (6 mg/kg/dose every 12 hr IV for the 1st 6 wk of life) concluded that treatment both prevents hearing deterioration and improves or maintains normal hearing function at 6 mo of age, and may prevent hearing deterioration that occurs after 1 yr of age. Drug-related toxicity was common, with significant neutropenia developing in 63% of ganciclovir-treated patients, compared with 21% in the untreated group. The logistic obstacles of intravenous therapy for the 1st 6 wk of life, limited benefit, and adverse effects have limited enthusiasm for this regimen. A randomized study of oral valganciclovir, is under way to compare the efficiency of 42 days of treatment with 6 mo of treatment.
The use of CMV-free blood products, especially for premature newborns, and, whenever possible, the use of organs from CMV-free donors for transplantation represent important measures to prevent CMV infection and disease in patients at high risk.
Pregnant women who are CMV seropositive are at low risk of delivering a symptomatic newborn. If possible, pregnant women should undergo CMV serologic testing, especially if they provide care for young children who are potential CMV excreters. Pregnant women who are CMV seronegative should be counseled regarding good handwashing and other hygienic measures and avoidance of contact with oral secretions of others. Those with suspected recent CMV infection may undergo additional diagnostic evaluations to ascertain in utero transmission and fetal disease. An uncontrolled trial has shown that the use of CMV hyperimmune globulin in pregnant women with primary CMV can lessen the risk of transmission to the unborn baby and can even reduce the risk of disease in the infected fetus.
The use of IVIG or CMV IVIG for prophylaxis of infection in solid organ and bone marrow transplant recipients reduces the risk of symptomatic disease but does not prevent infection. The efficacy of prophylaxis is more striking when the hazard of primary CMV infection is greatest, such as in bone marrow transplantation. There is no consensus for a uniform prophylaxis regimen for CMV infection. Recommended regimens include either IVIG (1,000 mg/kg) or CMV IVIG (500 mg/kg) given as a single intravenous dose beginning within 72 hr of transplantation and once weekly thereafter until 90-120 days after transplantation.
The beneficial role of immunity is substantial, as illustrated by the fact that most cases of severe disease follow primary infection, especially with congenital infection, transfusion-acquired infection, and infection in transplant recipients.
The 1st vaccine tested in humans was developed from the live attenuated Towne strain of CMV, which proved immunogenic but did not prevent infection in renal transplant recipients or normal adult women. However, in renal transplant recipients the vaccine reduced the virulence of primary infection. Promising results have been obtained with a CMV vaccine consisting of recombinant envelope glycoprotein B (gB) combined with the adjuvant MF59. In a double-blind placebo-controlled trial of seronegative adult women results showed that after 3 doses this vaccine induced antibodies to gB, neutralizing antibodies, and cell-mediated immunity. A later study of 464 women randomly assigned to receive either this gB subunit vaccine or placebo showed prevention of maternal infection as measured by seroconversion to CMV proteins other than gB in about 50% of vaccines. One congenital infection occurred among infants of this immunized group, whereas 3 infections occurred in the placebo group.
Bale JFJr. Screening newborns for congenital cytomegalovirus infection. JAMA. 2010;303(14):1425-1426.
Bonaros NE, Kocher A, Dunkler D, et al. Comparison of combined prophylaxis of cytomegalovirus hyperimmune globulin plus ganciclovir versus cytomegalovirus hyperimmune globulin alone in high-risk heart transplant recipients. Transplantation. 2004;77:890-897.
Foulon I, Naessens A, Foulon W, et al. A 10-year prospective study of sensorineural hearing loss in children with congenital cytomegalovirus infection. J Pediatr. 2008;153:84-88.
Fowler KB, Stagno S, Pass RF. Maternal immunity and prevention of congenital cytomegalovirus infection. JAMA. 2003;289:1008-1011.
Goelz R, Hihn E, Hamprecht K, et al. Effects of different CMV-heat-inactivation-methods on growth factors in human breast milk. Pediatr Res. 2009;65:458-461.
Kanengisser-Pines B, Hazan Y, Pines G, et al. High cytomegalovirus IgG avidity is a reliable indicator of past infection in patients with positive IgM detected during the first trimester of pregnancy. J Perinat Med. 2009;37:15-18.
Kimberlin DW, Lin CY, Sanchez PJ, et al. Effect of ganciclovir therapy on hearing in symptomatic congenital cytomegalovirus disease involving the central nervous system: a randomized, controlled trial. J Pediatr. 2003;143:16-25.
Lazzarotto T, Guerra B, Lanari M, et al. New advances in the diagnosis of congenital cytomegalovirus infection. J Clin Virol. 2008;41:192-197.
Nassetta L, Kimberlin D, Whitley R. Treatment of congenital cytomegalovirus infection: implications for future therapeutic strategies. J Antimicrob Chemother. 2009;63:862-867.
Nigro G, Adler SP, La Torre R, et al. Passive immunization during pregnancy for congenital cytomegalovirus infection. N Engl J Med. 2005;353:1350-1362.
Pass RF. Congenital cytomegalovirus infection: screening and treatment. J Pediatr. 2010;157(2):179-180.
Pass RF, Zhang C, Evans A, et al. Vaccine prevention of maternal cytomegalovirus infection. N Engl J Med. 2009;360:1191-1199.
Revello MG, Gerna G. Diagnosis and management of human cytomegalovirus infection in the mother, fetus, and newborn infant. Clin Microbiol Rev. 2002;15:680-715.
Ross SA, Fowler KB, Ashrith G, et al. Hearing loss in children with congenital cytomegalovirus infection born to mothers with preexisting immunity. J Pediatr. 2006;148:332-336.
Stagno S, Britt W. Cytomegalovirus. In: Remington JS, Klein JO, editors. Infectious diseases of the fetus and newborn infant. ed 6. Philadelphia: WB Saunders; 2005:739-781.
Vaudry W, Ettenger R, Jara P, et al. Valganciclovir dosing according to body surface area and renal function in pediatric solid organ transplant recipients. Am J Transplant. 2009;9:636-643.
Weller TH. The cytomegaloviruses: ubiquitous agents with protean clinical manifestations: I. N Engl J Med. 1971;285:203-214.
Weller TH. The cytomegaloviruses: ubiquitous agents with protean clinical manifestations: II. N Engl J Med. 1971;285:267-274.