Does umbilical cord blood polymerase chain reaction positivity indicate in utero (pre-labor) HIV infection?

Abstract
To compare risk factors for infants whose cord blood was positive for HIV DNA with those who were cord blood-negative but found to be HIV DNA-positive in early infancy. In 1994, infants born to HIV-infected women were enrolled in a study in Blantyre, Malawi. Birth weight and transmission risk factors from cord blood-positive infants were compared with cord blood-negative/HIV-positive infants on their first postnatal visit (4–7 weeks of age). Testing for HIV DNA on cord and peripheral blood was performed by polymerase chain reaction. Of 249 HIV-infected infants (overall transmission rate, 26%), 83 (33%) were cord blood-positive and 166 were initially cord blood-negative. The mean birth weight was 2.1% (59 g) lighter in cord blood-positive infants than initially cord blood-negative infants; initially cord blood-negative infants were 2.8% (80 g) lighter than uninfected infants born to HIV-infected women. There were no significant differences in the risk factors for infection between HIV-infected cord blood-positive and -negative infants; when transmission was increased, both HIV-infected cord blood-positive and -negative infants contributed to the increase in a similar proportion. It was concluded that umbilical cord blood positivity for HIV DNA did not identify a subset of in utero HIV-infected infants and suggested that HIV-infected cord blood-positive and -negative infants have similar timing and routes of HIV infection.