No abstract

No abstract

Accurate and timely diagnosis of infection status in infants born to women infected with human immunodeficiency virus (HIV) is of paramount importance. The comparative accuracy of five diagnostic decision rules was evaluated in 208 HIV-exposed infants (32 infected, 176 uninfected) based on laboratory testing during the first 6 months of life. Diagnostic rules A and B, which required single blood samples analyzed by culture and polymerase chain reaction (PCR) (rule A) or culture, PCR, and p24 antigen detection (rule B) were more prone to incorrect diagnoses than were rules requiring 2 blood samples analyzed by a single assay (rule C) or combinations of culture and PCR (rules D and E). Rule D, which used PCR as the initial test, established the most useful algorithm: a positive PCR result followed by a positive culture in the second sample confirmed infected status, while two consecutive negative PCR results reconfirmed as negative at 6 months of age established uninfected status.