Recommendations of the U.S. Public Health Service Task Force on the Use of Zidovudine to Reduce Perinatal Transmission of Human Immuno deficiency Virus

Vol. 43 / No. RR-11

MMWR

they are already in labor, when the full ZDV regimen used in ACTG Protocol 076 cannot be administered. Moreover, many pregnant women are not aware that they are
HIV infected, are not tested before or during pregnancy, and remain undiagnosed. As
a result, they do not receive information about therapy that could reduce the risk for
HIV transmission to their infants.
Potential Long-Term Adverse Effects of ZDV Administered During
Pregnancy
The long-term effects of ZDV treatment during pregnancy solely to reduce perinatal
transmission or of fetal and neonatal exposure to ZDV are not known. ZDV is a nucleoside analog that inhibits HIV replication by interfering with HIV RNA-dependent DNA
polymerase. ZDV triphosphate also can inhibit human cellular DNA polymerases, but
only at concentrations much higher than those required to inhibit HIV polymerase.
However, gamma DNA polymerase, which is required for mitochondrial replication,
may be inhibited by ZDV at concentrations nearer to those that can be achieved in
vivo.
Concerns related to the potential long-term toxicity of nucleoside analogs include
potential mutagenic and carcinogenic effects, possible effects on tissues with high
mitochondrial content (such as hepatic and cardiac tissue), possible teratogenicity,
and possible effects on the reproductive system.
ZDV has been shown to be a mutagen in vitro, and, in a mammalian in vitro cell
transformation assay, ZDV was positive at concentrations of 0.5 tg/mL (8). Noninvasive squamous epithelial vaginal tumors were produced after 19-21 months of
continuous dosing in 12% of mice administered a dosage equivalent to three times the
estimated human exposure at the recommended therapeutic dosage. Similar findings
were observed in 3% of rats that received 24 times the recommended therapeutic dosage. Carcinogenicity studies in rodents, however, may not be predictive of human
experience.
In humans, an increased incidence of non-Hodgkin's lymphoma has been reported
in HIV-infected men receiving ZDV, but this increase probably reflects longer survival
despite severe immunodeficiency rather than a direct effect of ZDV (9). The potential
for carcinogenesis should be further assessed through continued follow-up of children who were exposed to ZDV in utero.
Myopathy and cardiomyopathy have been associated with ZDV therapy. In an individual patient, the effects secondary to ZDV are often difficult to distinguish from
those of HIV infection. A prospective study of HIV-infected children demonstrated no
effect of ZDV therapy on cardiac function (10).
Reproductivity/fertility studies in animals have demonstrated no adverse effects of
ZDV on either the fertility of male or female rats or the reproductive capacity of their
offspring (11 ). ZDV administered to mice early in gestation was associated with an
embryotoxic effect and fetal resorptions; however, ZDV administered at or beyond
midgestation had no detectable effect on the fetus (12,13).