664 
Congenital Cytomegalovirus Infection in Infants Born to HIV-infected Mothers, Prevalence and Risk Factors: ANRS French Perinatal EPF Cohort CO1/10/11
G Guibert1, J Warszawski1,2, J Le Chenadec1, Y Benmebarek3, J P Teglas1, S Blanche4, L Mandelbrot5,6, R Tubiana7, Christine Rouzioux*4,8, M Leruez-Ville4,8, and The ANRS French Perinatal Cohort EPF
1INSERM U822, Le Kremlin Bicetre, France; 2Univ Paris Sud, Le Kremlin-Bicetre, France; 3Hosp Bicetre, Le Kremlin-Bicetre, France; 4Hosp Necker, Paris, France; 5Hosp Louis Mourier, Colombes, France; 6Univ Paris 7 Diderot, Paris, France; 7Hosp Pitie-Salpetriere, Paris, France; and 8Univ Paris 5, France
Background: In developed countries, 0.3% to 0.5% of all newborns are
congenitally infected by cytomegalovirus (CMV), thereby incurring the risk of
sensorineural hearing loss or mental retardation. Few results about congenital
CMV infection in infants born to HIV-infected women have been reported. We
aimed to estimate the prevalence of congenital CMV infection and perinatal risk
factors among children enrolled in the ANRS French Perinatal Cohort (EPF), a
national multi-center prospective cohort of mother-to-child HIV transmission.
Methods: All infants born between 1993 and 2004 in EPF sites and screened for CMV were included in the analysis. As part of the standard
follow-up of infants born alive during that period in participant maternities,
a urine sample was obtained within the first 10 days of life. From 1993 to
2001, these urine samples were tested for CMV using rapid viral culture, and since
2001, using real-time polymerase chain reaction (RT-PCR).
Results: Between 1993 and 2004, 4995 of the 7878 newborns included in
EPF were screened for CMV. The prevalence of CMV infection was 2.4% (119
positive tests; 95% confidence interval [CI] 2.0 to 2.8%), and 51.6% were born to mothers with a documented and
positive serological CMV test at delivery: time from seroconversion was
unknown for most of them; it occurred during pregnancy in at least 0.7% (n =
19) of them. As expected, congenital CMV infection
tended to be more frequent in cases of maternal seroconversion (2 of 19 [10.5%] vs 10 of 540 [1.9%]; p = 0.06). The prevalence of congenital CMV infection was significantly higher
in HIV-infected newborns (10.2% [13 of 128], 95%CI 4.9 to 15.5%) than in uninfected ones (2.2% [106 of 4867], 95%CI
1.8 to 2.6%; p <0.01).
Among children born at term and HIV-uninfected, maternal young age, maternal
low CD4 T cell count near delivery (<200 CD4 cells/mm3),
gemellity, and male gender were factors independently associated with
congenital CMV infection in a logistic regression. Congenital CMV infection was
not associated with the type of ART administered during pregnancy.
Conclusions: The
prevalence of congenital CMV infection was high in children born to HIV-
infected mothers, with a high risk of co-infections among HIV-infected infants.
|
