BABIES who somehow rid themselves of HIV infection shortly after birth have
intrigued AIDS researchers for about ten years. If the secret behind these
infants鈥� ability to shake off the virus could be uncovered, it might lead to new
drugs or vaccines. But now a team of researchers in the US has come to a
startling conclusion about these so-called transient HIV infections鈥攖hey
never actually happened.
Instead, in every case they have examined, they believe that sloppy testing
probably resulted in an uninfected newborn being misdiagnosed as HIV-positive.
Months later, when the test was repeated, the child appeared to have fought off
the virus (This Week, 8 April 1995, p 6).
鈥淥ur initial goal was to prove transient infections were real and then get a
clue to how these people cleared the infection,鈥� says Lisa Frenkel of the
University of Washington in Seattle. Frenkel and her colleagues started by
trying to confirm previous cases of transient infection.
Advertisement
Diagnosing HIV infection in infants is notoriously hard. The standard adult
test for anti-HIV antibodies doesn鈥檛 work, because maternal antibodies persist
in the blood for over a year after birth. So doctors use other techniques such
as looking for viral proteins directly, or by using the polymerase chain
reaction (PCR) to amplify one viral gene in a sample so there are millions of
copies that can be detected.
In the current issue of Science (vol 280, p 1073), the researchers
report their analysis of 42 instances of transient infection in infants and one
unusual case in a new mother. In each case they discovered something
suspicious.
In six cases they analysed human instead of viral genes and found the
HIV-positive samples belonged to another child. In 17 instances, the virus in
the babies鈥� blood was unrelated to the mother鈥檚 infection, which suggested the
sample was infected in the lab. In 20 cases, they couldn鈥檛 detect a full set of
viral genes in the blood sample. That suggests the original tests were false
positives or had been contaminated during PCR.
The researchers conclude that there are problems with these complicated tests
for infants and their mothers. 鈥淎nyone who has used them knows how problematic
they can be,鈥� says Frenkel. In this study they went to great lengths to avoid
contamination. For instance, samples from mothers and infants were analysed by
different people in different buildings a mile apart.
Frenkel adds that it is still possible there are real cases of transient
infections. 鈥淭his just emphasises the need to document them carefully,鈥� she
says.