Congenital cytomegalovirus infection

SYMPOSIUM: IMMUNITY AND INFECTION

Congenital cytomegalovirus infection

Introduction Human cytomegalovirus (CMV) is one of the most frequent congenital infections in humans. It is the commonest noninherited cause of childhood sensorineural hearing loss (SNHL) as well as a significant cause of neurodevelopmental delay. It is under-diagnosed because the majority of maternal infections during gestation are asymptomatic. Moreover, many newborns are also asymptomatic at birth but may manifest signs later in life and retrospective diagnosis is difficult.

Amelia Joseph Nikunj Mahida Gemma Clark William Irving Shiu Soo

Natural history CMV is an enveloped, double-stranded DNA b herpesvirus. Primary entry is usually via a mucosal site, followed by a viraemic phase where the virus infects a wide range of human tissues and subsequent secretion of the virus in bodily fluids such as saliva, urine, breast milk and genital secretions. The primary infection is usually asymptomatic in the immunocompetent host, but may produce an infectious mononucleosis-like syndrome in around 10% of older children or adults. Infection causes well described features in human cells such as cytomegaly, intranuclear inclusion bodies and multinucleated giant cells. In immunocompetent individuals this is followed by an immune response, clearance of the viraemia and subsequent viral latency. In the latent phase, the virus either stops replicating or undergoes low level replication at an undetectable level inside blood monocytes, tissue macrophages and bone marrow stem cells. Intermittently, the virus reactivates from these sites, leading to viraemia and viral shedding in bodily secretions.

Abstract Congenital cytomegalovirus is the most common intrauterine infection and the leading non-genetic cause of sensorineural hearing loss. Worldwide, the birth prevalence is estimated at 7 per 1000 with the highest rates seen in developing countries. The highest intrauterine transmission rates and risk of neurodevelopmental sequelae are associated with primary maternal infections. Transmission occurs less frequently after non primary maternal infections due to reactivation or reinfection. 10e15% of infected infants are symptomatic at birth, with neurological symptoms present in two-thirds. Infants who are asymptomatic at birth may go on to develop late neurodevelopmental sequelae, with sensorineural hearing loss being the commonest late consequence. Prenatal, neonatal and retrospective diagnosis can be challenging. Early treatment of symptomatic neonates with the antiviral drug valganciclovir can reduce the long-term neurodevelopmental sequelae. Universal or targeted screening for congenital CMV is not currently advocated. The development of an effective vaccine appears to be some years away. This review highlights the important considerations for clinicians regarding the diagnosis, investigation and management of children with possible or confirmed congenital CMV infection.

Keywords congenital infection; cytomegalovirus; screening; sensorineural hearing loss; valganciclovir

Epidemiology Seroprevalence of CMV increases with age and is closely related to the socioeconomic levels within a community. In developed countries 50% of women of child-bearing age are seropositive. In developing countries CMV is often acquired earlier in life due to higher breastfeeding rates and crowded living conditions, and up to 90% of childbearing aged women may be seropositive. The incidence of congenital CMV infection parallels maternal CMV prevalence. Acquisition of CMV during pregnancy often occurs through contact with young children who may shed the virus in their urine and saliva, or through sexual transmission from a partner. The reported rates of congenital CMV infection in developed countries are between 0.6% and 0.7% of live births, with rates in developing countries of between 1% and 5%.

ganciclovir;

Amelia Joseph BMedSci BMBS FRCPath PGCert MedEd is Consultant Microbiologist in the Department of Clinical Microbiology, Nottingham University Hospitals NHS Trust, Nottingham, UK. Conflicts of interest: none declared. Nikunj Mahida MBChB MSc FRCPath is Consultant Microbiologist in the Department of Clinical Microbiology, Nottingham University Hospitals NHS Trust, Nottingham, UK. Conflicts of interest: none declared.

Routes of transmission Vertical transmission of CMV can occur through three routes: intrauterine, intrapartum and post-natal. Intrauterine transmission is the most important route because it leads to congenital infection and its subsequent complications. Intrauterine transmission occurs through transplacental maternal leucocyte translocation, or direct infection of the placenta and amniotic fluid. Congenital CMV may result from either a primary maternal infection, reactivation of the latent maternal virus, or reinfection with a different viral strain during pregnancy. Primary maternal infection occurs in 1e4% of seronegative pregnant women. Transmission rates vary from 30 to 40% in the first trimester to up to approximately 75% in the third trimester. However, transmission to the fetus in the first trimester is associated with the greatest risk of severe fetal infection and subsequent

Gemma Clark BSc (Hons) PgDip MSc PhD is Clinical Scientist in the Department of Clinical Microbiology, Nottingham University Hospitals NHS Trust, Nottingham, UK. Conflicts of interest: none declared. William Irving MBBCh FRCPath PhD is Professor of Virology & Consultant Virologist in the Department of Clinical Microbiology, Nottingham University Hospitals NHS Trust, Nottingham, UK. Conflicts of interest: none declared. Shiu Soo MA PhD MB BChir FRCPath is Consultant Microbiologist in the Department of Clinical Microbiology, Nottingham University Hospitals NHS Trust, Nottingham, UK. Conflicts of interest: none declared.

PAEDIATRICS AND CHILD HEALTH 28:6

277

Ó 2018 Elsevier Ltd. All rights reserved.


developmental sequelae. Congenital infection after non-primary infection in the mother has significantly lower rates (1e3%) of fetal transmission and sequelae. Intrapartum transmission occurs through exposure to the virus in the maternal genital tract. Around 10% of seropositive mothers shed CMV in the genital tract at the time of delivery, with about 50% of exposed neonates acquiring infection through this route. Post-natal transmission occurs primarily through viral shedding in breastmilk and in oral secretions, and along with intra-partum transmission, usually causes asymptomatic infection in term neonates. In premature neonates very early acquisition of CMV may lead to more significant symptoms.

Clinical findings and laboratory abnormalities in symptomatic congenital CMV infection

Pathogenesis Studies of fetuses and neonates with congenital CMV show that the virus infects numerous cell types, with associated inflammatory infiltrate and organ damage. Brain cells of many different types show inclusions with focal necrosis. Epithelial cells of the semi-circular canals, vestibulae, cochlear and other ear structures are affected. Cytomegalic cells and focal necrosis can be seen in the retina, liver, lung and kidney. In addition to direct fetal effects, villitis and vascular necrosis can impair placental function. Neonates with congenital CMV infection are less able to control the infection due to immaturity of the immune system. This contributes to the commonly progressive nature of SNHL. Infants infected early in life may continue to shed virus in their urine and other secretions, acting as reservoirs of the virus.

Clinical finding

Frequency (%)

Petechiae Neurologic, one or more of the following: Microcephaly Lethargy/hypotonia Poor suck Seizures Jaundice Hepatosplenomegaly Small for gestational age (weight 120mg/dL

76 68 53 27 19 7 67 60 50 34 83 77 53 81 69 51 46

With permission from Boppana SB, Pass RF, Britt WJ, et al. Symptomatic congenital cytomegalovirus infection: neonatal morbidity and mortality. Pediatr Infect Dis 1992; 11:93.

Clinical manifestations Table 1

Symptomatic congenital CMV infection Table 1 summarises the common clinical and laboratory findings recorded in a review of 106 infants with symptomatic congenital CMV. 10e15% of congenitally infected neonates are symptomatic at birth, and jaundice, petechiae and hepatosplenomegaly are the most common clinical signs. Prematurity, small for gestational age growth parameters, thrombocytopenia and anaemia are frequently present. Other features include chorioretinitis, hepatitis, pneumonitis, colitis, and bone or dental abnormalities. Some degree of neurological abnormality is present in around two-thirds of symptomatic neonates. Neonates with symptomatic congenital CMV at birth should be considered high risk for long-term sequelae even if initial clinical findings are mild. Studies have shown that between 40 and 90% of symptomatic newborns develop long-term neurodevelopmental sequelae. SNHL occurs in approximately 35%, visual impairment in 22e58%, and cognitive deficits in up to two-thirds. Computer tomography, ultrasound and magnetic resonance techniques can provide evidence of CNS involvement with abnormal CNS imaging being predictive of neurological sequelae. Mortality rates in symptomatic, congenitally affected infants have been reported as ranging from 4% to as high as 20e30%.

than 25% of congenitally infected fetuses, therefore negative scans can only suggest reduced risk, but cannot exclude congenital CMV infection. Intrauterine growth restriction, abnormal amniotic fluid volume (usually oligo-rather than polyhydramnios), hyperechogenic bowel, pleural effusions and liver calcifications may be visible. Microcephaly, cerebral ventriculomegaly and intracranial calcifications are the commonest neurological abnormalities detected on prenatal ultrasound scanning. Cerebral ultrasound abnormalities are strongly associated with a poor prognosis in relation to neurological and cognitive development. Placental pathology, for example placental enlargement or infarcts, may also be visible on ultrasound. Differential diagnosis There are a number of congenital infections causing differential diagnostic problems including toxoplasmosis, rubella, parvovirus B19 and syphilis. Toxoplasmosis may cause cerebral calcifications, chorioretinitis and usually a macular, rather than petechial, rash. Rubella may present with petechiae, bone defects and SNHL. Parvovirus B19 can cause hepatomegaly and anaemia. Early congenital syphilis may cause hepatosplenomegaly, bony changes and lymphadenopathy. Neonatal infections may also mimic the symptoms of congenital CMV. Disseminated herpes simplex infection, enteroviral infection and bacterial sepsis may have similar clinical or biochemical abnormalities. Non-infectious conditions including a large number of metabolic disorders may also produce similar features.

Radiological features Prenatal ultrasound scanning may detect structural or growth abnormalities caused by CMV infection. However, the sensitivity and specificity can be poor as these abnormalities are common to many other congenital conditions. Ultrasound findings are present in less


278



Asymptomatic congenital CMV infection The remaining 85e90% of congenitally infected neonates are asymptomatic at birth (Figure 1). Given that maternal infection is also likely to be asymptomatic, this makes the detection and diagnosis of these neonates particularly difficult. Lower rates of neurodevelopmental sequelae are reported in asymptomatic children compared to symptomatic children. Approximately 5e15% of asymptomatic neonates go on to develop some longterm sequelae with SNHL being the most common manifestation. The long-term risk of cognitive and developmental delay has not been studied systematically; however some studies have shown a degree of cognitive delay or impaired functional abilities on long-term follow-up of asymptomatic infants.

Maternal diagnosis The diagnosis of primary maternal CMV infection is based on serology. This is best demonstrated by seroconversion from CMV IgG negative to IgG positive, when a baseline serum sample from either the earliest antenatal visit or prior to conception is available. The detection of CMV IgM in isolation is not helpful because it may be found in both primary infections as well as reactivation of CMV, and is also prone to non-specific crossreactivity. It can also remain elevated for several months following primary infection. The detection of both CMV IgM and IgG antibodies may indicate a recent infection but it does not allow the ability to distinguish between primary and non-primary infections. CMV IgG avidity testing allows some differentiation of timing of infection. Detection of low avidity CMV IgG with IgM indicates recent primary CMV infection within the previous 3e4 months. The presence of high avidity CMV IgG is evidence of an infection older than 6 months.

Perinatal and postnatal infection in premature neonates In term babies post-natal acquisition of CMV (for example through breastfeeding) has little significance. In contrast, premature neonates may demonstrate worsening respiratory status, neutropaenia or a sepsis-like presentation. For premature infants acquiring infection postnatally it remains unclear whether the virus causes any long-term neurodevelopmental sequelae independent from prematurity itself. CMV can also cause symptoms in premature infants if acquired through blood transfusion; the use of leucocyte-depleted, irradiated, or CMV-IgG negative donor blood products reduces this risk.

Prenatal diagnosis of fetal cytomegalovirus infection The potential for severe sequelae in the fetus through primary maternal CMV infection may justify invasive testing. It may also be considered in reactivation if there are sonographic findings to suggest CMV abnormalities in the fetus. Amniocentesis and CMV PCR is the preferred method for confirming fetal CMV infection, and if required should be performed after 21 weeks of gestation and at least 6 weeks following maternal infection. Prior to 20 weeks gestation amniocentesis is associated with a lower sensitivity and there is a 6 week lag time from maternal infection to transmission through the placenta, viral replication in the fetal kidney and excretion into amniotic fluid. Nucleic acid detection assays, such as real-time PCR, are the most sensitive method for the detection of CMV DNA in amniotic fluid. Whilst false positives can occur from contamination of the amniotic fluid with maternal tissues, high CMV viral loads are pathognomonic. Fetal cord blood sampling does not increase sensitivity or specificity of detecting symptomatic CMV infection. In infected fetuses, 2 to 4 weekly ultrasound examinations can help detect symptomatic infections and define prognosis.

Sensorineural hearing loss (SNHL) Hearing loss due to congenital CMV infection can be unilateral or bilateral. Children with normal hearing at birth can develop hearing loss later, and the progressive nature of hearing loss is suggestive of active, persistent infection within the endolabyrinth. Congenital CMV can lead to late-onset or a progressive decline in hearing, usually within the first two years of life, but it has also been observed in school aged children. It is estimated about 21% of all hearing loss at birth and around 25% of all hearing loss at 4 years of age is attributable to congenital CMV infection.

Diagnosis

Postnatal diagnosis The currently accepted standard for diagnosing congenital CMV infection in newborns is virus detection by PCR amplification in the urine in the first 3 weeks of life. However collection of urine from neonates can be difficult. A study comparing saliva and urine CMV PCR in newborns found that sampling difficulties prevented one third of urine samples being analysed, whereas saliva samples were more easily obtained. A more recent prospective multicentre trial found both liquid and dried saliva to have high rates of sensitivity and specificity. These recent advances have demonstrated that saliva PCR is both sensitive and specific and may be considered as the investigation of choice for diagnosing congenital CMV infection in newborns.

Testing for CMV infection is usually performed in pregnancy if there is clinical suspicion of a maternal illness such as mononucleosis or flu-like illness, or when a fetal abnormality suggestive of congenital CMV infection is detected on prenatal ultrasound examination.

Primary Maternal CMV Infection 30-40% Fetal Infection 10-15% Symptomatic at birth 50-60% Long-term Sequelae

Retrospective diagnosis Older children presenting with neurodevelopmental symptoms or SNHL may undergo investigation to retrospectively diagnose congenital CMV infection. Dried blood spots (DBS) are collected routinely after birth in a number of countries including the United Kingdom. CMV DNA has been shown to be detectable from such

85-90% Asymptomatic at birth 5-15% Late Sequelae

Figure 1 Outcomes after primary maternal CMV infection.


279



congenital CMV disease. There are currently no specific recommendations for the use of CMV-specific immunoglobulin. Disappointingly, the first randomised-controlled trial of perinatal administration of CMV-specific immunoglobulin to prevent congenital CMV following primary maternal infection during pregnancy found no significant reduction in the rates of CMV transmission when compared to a placebo. Further randomized controlled trials are ongoing, which may clarify the role of prophylactic CMV-specific immunoglobulin.

cards in children with SNHL up to 18 years of age. However, four large studies have shown that sensitivities vary from 34% to 100% for CMV DNA detection from DBS. Technical factors such as the size of the DBS, DNA extraction methods and the fact that not all newborns are viraemic at birth may account for the wide range of sensitivities. Therefore, a positive CMV PCR result from a DBS taken in the first 3 weeks of life is diagnostic of congenital CMV infection, but a negative result cannot reliably exclude it.

Management Monitoring Close monitoring during ganciclovir or valganciclovir therapy is required due to the risk of adverse effects. Haematological parameters should be monitored regularly due to bone marrow suppressive effects, and it is recommended that absolute neutrophil counts should be followed weekly for 6 weeks, then at week 8 and then monthly for the duration of therapy. Neutropaenia is common (63% with ganciclovir and 38% with valganciclovir) and cessation of treatment may be required until counts recover. Thrombocytopenia and anaemia can also occur. Liver function tests and creatinine clearance should be monitored, and dose reduction may be required in renal impairment. Whole blood viral loads should be monitored weekly to assess treatment efficacy. Blood viral loads usually fall dramatically (between 1 and 2 logs) on treatment. Urine and saliva usually have higher loads initially and tend to fall even more convincingly (by around 3 to 4 logs). Therapeutic drug level monitoring can be performed regularly to ensure treatment efficacy and lower the risk of adverse effects. Emergence of resistance can occur and resistance testing should be considered in those neonates with persistently high viral loads not responding to antivirals.

Who to treat Antiviral treatment is currently only recommended for symptomatic newborns with CNS disease or severe symptomatic focal organ disease. There is no current evidence for the effectiveness of antiviral treatment in CMV infected infants without CNS disease or disseminated life-threatening organ dysfunction due to congenital CMV. Treatment has been shown to be effective when commenced within the first 30 days of life. Antiviral therapy and outcomes The nucleoside analogue, ganciclovir and its prodrug, valganciclovir are the preferred antiviral agents for use in the treatment of symptomatic CMV disease. A phase III randomised trial assessed the outcome of symptomatic neonates randomised to either 6 weeks of intravenous ganciclovir 6 mg/kg every 12 hours commencing within the first month of life or no treatment. Treatment was shown to prevent hearing deterioration at 6 months and 1 year of life; there were also short-term improvements in weight gain, head circumference and resolution of liver abnormalities. Follow-up showed reduced developmental delays at 6 and 12 months compared to untreated infants. Valganciclovir is recommended at a dose of 16 mg/kg twice daily, as an off-license indication, and should be given for no longer than 6 months when the primary outcome is improvement of audiological or developmental outcomes. Although there have been no head-to-head trials comparing the clinical end-points and efficacy of valganciclovir with ganciclovir, pharmacokinetic studies have shown similar parameters between the two drugs. Given that the administration of 6 weeks of IV ganciclovir is likely to require the placement of a central venous line, the risk/benefit analysis between prolonged central line use and potential added benefit of IV ganciclovir over oral valganciclovir should be considered.

Long-term follow-up Initial ophthalmological examination should be performed at diagnosis to assess for retinal scarring in children, and symptomatic neonates should have annual ophthalmology assessments until the age of 5. Consensus guidelines on the management of congenital CMV recommend hearing assessments every 6 months until the child is 3 years of age, and then annually until adolescence. Neurodevelopmental assessment beginning at the first year of life may be helpful in some children, and should be employed on a case-by-case basis. Support for parents and families is available from a voluntary organisation, The Congenital CMV Association.

Prenatal use of CMV-specific immunoglobulin As there is a lower risk of fetal transmission in mothers who have pre-existing CMV-specific antibodies, there is a rationale for passive immunization of mothers with primary infection using CMV-specific immunoglobulin. Whilst there is some limited evidence to suggest that CMV-specific immunoglobulin may be effective in inhibiting transmission in primary CMV infection, a 2011 Cochrane review highlighted the lack of high quality research in this area.

Prevention Behavioural measures Prevention of CMV infection during pregnancy is challenging due to the epidemiology of CMV infection. In the absence of an effective vaccine, prevention of maternal infection relies on behavioural measures that limit exposure to the virus. Pregnant women with exposure to young children may be frequently exposed to the virus in urine and other secretions, and the Centers for Disease Control and prevention (CDC) has produced advice on measures to reduced CMV exposure in pregnant

What’s new? New consensus recommendations published in 2017 advocate the use of oral valganciclovir for the treatment of symptomatic


280



women focused around hand hygiene. CMV infection rates are not increased in hospital workers who care for children, suggesting that standard hygiene precautions may be effective in preventing transmission.

Bialas KM, Permar SR. The march towards a vaccine for congenital CMV: rationale and models. PLoS Pathog 2016; 12: e1005355. https://doi.org/10.1371/journal.ppat.1005355. Boppana SB. Saliva PCR assay for CMV screening in newborns. N Engl J Med 2011; 364: 2111e8. Dollard SC, Grosse SD, Ross DS. New estimates of the prevalence of neurological and sensory sequelae and mortality associated with congenital cytomegalovirus infection. Rev Med Virol 2007; 17: 355e63. Griffiths PD. Strategies to prevent CMV infection in the neonate. Semin Neonatol 2002; 7: 293e9. Kadambari S, Luck S, Davis A, et al. Clinically targeted screening for congenital CMV e potential for integration into the National Hearing Screening Programme. Acta Paediatr, 2013; https://doi.org/10. 1111/apa.12335. Manicklal S, Emery VC, Lazzarotto T, Boppana SB, Gupta RK. The “silent” global burden of congenital cytomegalovirus. Clin Microbiol Rev 2013; 26: 86e102. Pass RF. Cytomegalovirus. In: Long SS, Pickering LK, Prober CG, eds. Principles and practice of pediatric infectious diseases. 5th ed. Elsevier Saunders, 2018; 1073e80. Pass RF, Zhang C, Evans A, et al. Vaccine prevention of maternal cytomegalovirus infection. N Engl J Med 2009; 360: 1191e9. Rawlinson WD, Boppana SB, Fowler KB, et al. Congenital cytomegalovirus infection in pregnancy and the neonate: consensus recommendations for prevention, diagnosis, and therapy. Lancet 2017; 17: e177e88. Revello MG, Lazzarotto T, Guerra B, et al. A randomized trial of hyperimmune globulin to prevent congenital cytomegalovirus. N Engl J Med 2014; 370: 1316e26. Swanson EC, Schleiss MR. Congenital cytomegalovirus infection: new prospects for prevention and therapy. Pediatr Clin N Am 2013; 60: 3355e4349. Syggelou A, Iacovidou N, Kloudas S, Christoni Z, Papaevangelou V. Congenital cytomegalovirus infection. Ann N Y Acad Sci 2010; 1205: 144e7.

Prenatal screening Routine serological screening of pregnant women for CMV serological status is not recommended in the UK. Firstly, although seronegative women at risk of primary CMV infection would be identified, seropositive women are still susceptible to reactivation or reinfection. Secondly, there is no vaccine or other approved post-exposure prophylaxis including CMV specific antibodies available to susceptible pregnant women. Thirdly, although maternal seroconversion may be identified through the regular serological follow-up, this will not always lead to fetal transmission or long-term sequelae. However, there is some evidence that counselling seronegative women regarding risk reduction decreased the rate of primary infections and proponents argue this supports ascertaining serological status early in pregnancy. Postnatal screening Testing for congenital CMV is not currently included in the newborn screening programme in the UK. There are two suggested approaches in the UK: universal screening as part of the pre-existing NHS Newborn Bloodspot Screen Programme (NBSP), or clinically targeted screening as part of the NHS Newborn Hearing Screening Programme (NHSP). Universal screening could be integrated by the addition of CMV DNA detection by PCR on DBS. Asymptomatic congenitally-infected infants identified could then receive enhanced follow-up for SNHL. This approach has significant cost implications. A clinically targeted approach is more cost-effective; newborns that have SNHL identified through the NHSP could be referred for CMV PCR on a salivary sample. If this could be performed within 3 weeks of birth then a diagnosis of symptomatic congenital CMV could be made. However this would not identify those asymptomatic infants who later develop progressive hearing loss. Vaccination There is currently no available licensed CMV vaccine. The United States Institute of Medicine of the National Academy of Science has identified the development of a CMV vaccine as a top priority for the 21st century and several investigatory vaccines are currently in clinical trials. Previous results from a CMV vaccine trial in child-bearing-aged women showed a reduction in primary infections during pregnancy, but it is likely to be a number of years before a licensed vaccine is available. A phase 3 trial of the ASP0113 CMV vaccine in haematopoietic stem cell recipients has recently concluded and results are expected in 2018. A

Practice points C

C

C

C

FURTHER READING Adler SP, Finney JW, Manganello AM, Best AM. Prevention of child-tomother transmission of cytomegalovirus by changing behaviors: a randomized controlled trial. Pediatr Infect Dis J 1996; 15: 240. Atkinson C, Walter S, Sharland M, et al. Use of stored dried blood spots for retrospective diagnosis of congenital CMV. J Med Virol 2009; 81: 1394e8.


C

C

281

Congenital cytomegalovirus is the most common intrauterine infection and the leading non-genetic cause of sensorineural hearing loss. Transmission to the fetus can occur after primary maternal infection, or after reinfection or reactivation in the mother. Diagnosis can be made when virus is detected in saliva or urine within the first 21 days of life, or prenatally in amniotic fluid. Early treatment of symptomatic neonates with the antiviral drugs ganciclovir or valganciclovir reduces the long-term neurodevelopmental sequelae. Potential future advances include implementation of screening programmes and vaccine development. There is currently a lack of high quality evidence to support the use of CMV-specific immunoglobulin to prevent maternal transmission of CMV following primary infection.
