Abstract 1588 Poster Session II, Sunday, 5/2 (poster 121)

Introduction. During a four year period (1991-1995) in a randomized study of the efficacy of intravenous immunoglobulin (IVIG) in preventing infections in low birthweight infants (<1500 grams), the incidence, morbidity and mortality of cytomegalovirus (CMV) infection was studied.

Methods. On admission, infants were assigned to groups determined by mother's CMV serostatus, Serial urine viral cultures and/or PCR for CMV were performed on the infants born to seropositive mothers. Infants born to seronegative mothers had at least one viral culture and/or PCR on admission and/or discharge. Infants were followed until 70 days or discharge. Viral cultures were obtained on fresh breast milk when available. The administration of CMV negative blood was practiced during the study period

Results. The incidence of CMV seropositive mothers was 66%. A total of 199 infants were enrolled: 6 were eliminated due to failure to determine maternal CMV status or withdraw from study. 131 infants were born to seropositive mothers (68%) and 62 infants born to seronegative mothers (32%). There were no differences between the CMV seropositive and seronegative groups for birthweight, gestation age, mean number of hospital days and mean number of sepsis evaluations. Eleven infants (6%) were identified to have CMV infection: one infant had a positive urine culture at birth (congenital infection rate=0.8%) and 10 had a positive urine at average 32.2 days of life (range 16-57 days)(acquired infection rate=5.2%). Eight infants with acquired infection (6.1%) were born to CMV seropositive mothers and 2 (3.2%) to seronegative mothers. There were no deaths among the CMV infected infants. Four fresh breast milk cultures from 4 mothers were performed. Breast milk was cultured and fed to mother's own infant on average 27 days of life. Two of four cultures were positive. Follow up urine cultures were obtained on one infant after 5 days and the other after 23 days. None of these infants who received breast milk were infected with CMV. There was also no significant difference detected between infants who received IVIG (3/92,3.4%) than those who did not receive IVIG (7/90,7.8%) and acquisition of CMV.

Conclusion. The congenital CMV infection rate in low birthweight infants appears similar to those born near term. Postnatal acquisition of CMV can occur as early as 16 days of age. IVIG did not influence the rate of postnatal infection. Infants born to both CMV seronegative and seropositive mothers appear at risk for infection.