Thrombocytopoenia is a well-described marker of human immunedeficiency viru
s type 1 (HIV) infection of children. After documenting thrombocytopoenia i
n a HIV-exposed neonate, who was subsequently shown not to be infected, we
evaluated thrombocytopoenia (platelets < 150 X 10(9)/l) as a marker for HIV
exposure in newborns. This is, to our knowledge, the first study of thromb
ocytopoenia in MV-exposed neonates. A retrospective hospital-based descript
ive study was performed over an 18-month period (July 1997-December 1998).
Patients were recruited either through identification of known HIV-positive
mothers or neonates with thrombocytopoenia who were then screened for HIV
exposure. For eligibility, neonates with either HIV antibodies or direct ev
idence of infection (p24 antigenaemia or HIV RNA detected by polymerase cha
in reaction after I month of age) were included. Thirty-four HIV-exposed (H
IV-ELISA positive) neonates were identified, of whom 16 (47 per cent) had t
hrombocytopenia. In 12 (35 per cent), no underlying cause other than HIV wa
s identified. Nine thrombocytopoenic babies were infected and seven uninfec
ted. Unexplained thrombocytopoenia, while an imperfect marker, should alert
the physician to the possibility of HIV exposure. It does not necessarily
imply HIV infection. A prospective study is recommended to evaluate further
the sensitivity and specificity of this marker.