 |
|
 |
|
 |
 |
|
 |
 |
 |
 |
 |
 |
 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
PEDIATRICS Vol. 107 No. 5 May 2001, pp. 1227-1228
A Significant
Cause of Congenital Disease
To the Editor.
In November 1999 a comprehensive report by us on a Swedish
long-term prospective study of congenital cytomegalovirus (CMV) infection was published in the Scandinavian Journal of Infectious Diseases.1 The main observation The Swedish study1 was based on observations made in
infants born in 1977-1985. Congenital infection was identified from
newborn virologic screening at the General Hospital in Malmö,
Sweden. Maternal infections The large ongoing study from Alabama has provided very valuable
information of the spectrum of CMV disease and damage. The authors
describe 2 groups of mother-infant pairs studied. The first series of
infants were born 1972-19906,7 and the second
1991-1997.5 Based on findings in their first study, the
authors believed that symptomatic congenital CMV infection almost
always was the result of a primary maternal infection. However, the
results in the second study clearly documented that symptomatic
congenital CMV infection after a recurrent maternal infection occurs
more frequently than was thought previously. The authors do not have any explanation for the discrepancy between the findings in the 2 series. They propose that improved ability to categorize maternal infections with the help of a well-organized serum bank, including prepregnancy samples from a greater proportion of mothers, might be an
explanation.5
These unanimous data on the impact of secondary maternal CMV infection
obtained in several countries and populations make it possible to go on
to another question. What type of secondary maternal infection is
responsible for congenital infection? Reactivated infection or
reinfection with a new strain or both of them? Neutralization test,
performed in paired preconceptional and delivery (or cord) sera against
the infant's own CMV strain, would possibly solve this problem. The
answer has implications for the strategy of future vaccination.
Hopefully, only primary maternal infection and reinfection with a new
strain are able to cause infant disease. In that case, the goal would
be to prevent new infections from occurring during pregnancy by
vaccination with a broad-spectrum vaccine before fertile age.
Eradication of wild virus will, however, be a long-term project because
reactivated CMV is often transmitted vertically from mothers to infants
already during the first year of life, and later during infancy
horizontally especially in societies with low hygiene. The virus
remains in latent or reactivated form for the rest of life.
It would also be desirable to explore the impact of congenital and
maternal primary/secondary CMV infection in certain infant diseases
such as sensorineural deafness and cerebral palsy. Retrospective diagnosis of congenital infection is possible as late as 15-20 years
after birth by CMV-DNA analysis performed on blood drops sampled at
birth and stored in filter paper.9,10
already reported in
19842was that secondary (recurrent) maternal CMV
infection seems to be an important cause of congenital CMV disease.
Data from England (1983)3 and Belgium (1999)4
indicate a similar situation as in Sweden. While our report (1999) was
in press, a US report from Alabama (1999)5 was published
in the present journal with the same signification. The new US data
were contrary to those reported by the same group as late as in
1992.6,7 The new unanimous findings have weighty
consequences.
primary or secondary (reactivated latent
or reinfection with a new strain)were categorized with conventional serological tests and criteria. Sera were regularly sampled from the
mothers during pregnancy and at delivery; often preconceptional sera
were also available. A few years after starting, we detected the first
case of secondary maternal infection causing infant neurologic
disease.8 Some years later it was evident that this case
was not a unique one; 4 of the 5 infants shown to have symptoms from
the central nervous system (CNS) were born to mothers with confirmed or
presumed secondary infections.2 In our recent report we
presented the whole material of 16 474 virologically screened
newborns, 76 (0.5%) of them with confirmed congenital CMV infection.
Among 60 followed up for at least 7 years, 11 had symptoms from CNS.
Investigation of the mothers showed that symptomatic congenital
infection occurred in 5/27 (19%) primary, 6/23 (26%) secondary, and
unidentified maternal infections occurred in 0/10 (or 4/23 confirmed
primary, 1/4 presumed primary, 1/12 confirmed secondary, and 5/11
presumed secondary infections). It is noticeable that the diagnosis of
confirmed secondary infection requires access to a preconceptional
serum sample. To exclude technical and interpretational mistakes, sera from several mothers with presumed secondary infection were tested also
in other laboratories, in all cases with the same results as in ours.
* Department of Clinical Microbiology
Section of Clinical Virology
Department of Paediatrics
University Hospital
Malmö General Hospital
20502 Malmö, Sweden
Department of Logopedics, Phoniatrics and Audiology
University Hospital
University of Lund
Lund, Sweden
REFERENCES
In Reply.
We appreciate the comments by Dr Ahlfors and colleagues regarding our report.1 We also acknowledge the contributions of Dr Ahlfors and colleagues in Sweden to our understanding of the natural history of congenital CMV infection.2-4 The findings from our recent report have confirmed the results of studies from Sweden and England that children with congenital CMV infection born to mothers who were CMV-seropositive before pregnancy are also at risk for long-term neurologic complications.1-5 In our earlier studies, we did not observe symptomatic congenital CMV infection in children born to women with recurrent or secondary CMV infection.6 Although we do not have a clear explanation for this discrepancy, these findings can be explained at least in part by an improved ability to classify maternal CMV immune status in our recent studies. However, the exact contribution of the recurrent maternal infection to an adverse neurologic outcome in children with congenital CMV infection is not clear. This information can be provided only by studies where most maternal infections can be unequivocally classified as either primary or recurrent. In our recent report, only children who were born to women with confirmed recurrent CMV infection were included because most CMV-IgM antibody assays have only about 70% to 80% sensitivity. The relatively low sensitivity of IgM assays may also result in misclassification of a primary infection as a recurrent infection because of the absence of IgM antibodies. In their most recent publication by Ahlfors and colleagues, only 2 of the 8 children in the recurrent infection group with an adverse neurologic outcome were born to women with confirmed recurrent CMV infection and only 1 of these 2 had clinical abnormalities at birth.4 In view of the interest in the development of CMV vaccines to prevent the damage associated with congenital CMV infection, we agree with Dr Ahlfors and colleagues that it is important to determine whether the congenital infection in children of immune mothers is the result of a reactivation or reinfection with a new virus. The focus of ongoing studies in our laboratory is to address this issue of reactivation versus reinfection.
We would like to point out that the term "symptomatic congenital CMV infection" was used differently in our studies than the studies by Ahlfors et al.1,2,4,6 We have used this term to describe only those infants with clinical abnormalities suggestive of congenital infection during the newborn period and not to describe the children with neurologic sequelae on follow-up. In contrast, Ahlfors et al have classified all children with CNS involvement at any time during their follow-up as those with "symptomatic infection." Thus, it is important to recognize this difference when the results from various studies are being compared. It was documented that the presence of clinical abnormalities during the newborn period is an important predictor for the development of long-term neurologic sequelae in children with congenital CMV infection.7-10 Although Ahlfors et al have not observed an association between clinical abnormalities at birth and sequelae in their most recent study, several previous studies from this laboratory and others have documented that children with newborn findings are at much greater risk for developing long-term neurologic complications than those who are asymptomatic at birth.7-10 In a study of a large number of children with symptomatic congenital CMV infection, we could document the presence of CNS involvement in more than 70% of children.11
Department of Pediatrics
University of Alabama at Birmingham
Birmingham, AL 35294-0011
REFERENCES
In Reply.
Dr Boppana correctly remarks that abnormalities during the newborn period are an important prediction for the development of long-term neurologic sequelae in children with congenital CMV infection. However, the possibility of making such a prediction depends on if these symptoms are of CNS or non-CNS character. In our prospective study1 the only detected infectious symptoms at birth were of mild to moderate non-CNS character. I cite page 452 of this study:
It is difficult to predict the outcome in infected infants who lack neonatal symptoms or show only non-CNS symptoms. In agreement with Bale et al,2 we found no relation between a negative outcome (ie, permanent neurologic disease) and the occurrence of transient neonatal features such as premature birth, hepatomegaly, splenomegaly, or petechiae. Possibly the mother's age, number of sex partners, how long she and the infant's father lived together before conception, etc, may affect the character of her own CMV infection and, consequently, the infant's infection. In the citation above, reference is also made to Swedish and English studies3,4 performed on occasional cases of severe congenital CMV infection detected in the routine service: Even in infants with neonatal signs of cerebral infection, the subsequent course of events might be problem-free or better than expected.
Department of Clinical Microbiology
Section of Clinical Virology
University Hospital
Malmö General Hospital
20502 Malmö, Sweden
REFERENCES
This article has been cited by other articles:
 |
|
 |
|
  S. Lipitz, R. Achiron, Y. Zalel, E. Mendelson, M. Tepperberg, and R. Gamzu Outcome of Pregnancies With Vertical Transmission of Primary Cytomegalovirus Infection Obstet. Gynecol., September 1, 2002; 100(3): 428 - 433. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
||
|
||
Home | My Pediatrics | Journal Information | Current Issue | Past Issues | Subscriptions & Services | Contact Us Click here for faster international access © 2001 American Academy of Pediatrics. All rights reserved. |
||
|
||
