Formula-fed
infants born to women who started ART during pregnancy under routine programme
conditions in Botswana, without frequent viral load monitoring or specialised
care, were fourteen times less likely to become infected with HIV compared to
infants born to mothers who got zidovudine and/or single-dose nevirapine.
Scott
Dryden-Peterson and colleagues report the findings in a prospective observational study
published in the advance online edition of the Journal of Acquired Immune Deficiency Syndromes.
The
observed transmission rate in mothers on ART (0.4% 95% CI: 0-2.2%) is among the
lowest reported to date, “supporting the effectiveness of ART for PMTCT outside
the context of a clinical trial.” And, in accordance with Botswana’s treatment guidelines the
women on ART had lower CD4 cell counts than those getting zidovudine and/or
single-dose nevirapine. Yet, lower CD4 cell counts are strongly associated with
vertical transmission.
The
2010 World Health Organization (WHO) guidelines for prevention of
mother-to-child transmission recommend that pregnant women who need treatment
for their own health (those with CD4 cell counts under 350 cells/mm3
or WHO clinical stage 3 or 4), start ART.
For
those pregnant women not yet in need of treatment, WHO recommends either ART or
zidovudine with single-dose nevirapine if zidovudine has been used for under
four weeks, followed by infant prophylaxis.
Observational
evidence suggests that maternal ART may lead to greater reductions in MTCT in
women at all CD4 cell counts. However, a recent randomised trial, the Kesho
Bora study, did not find a significant difference between ART and zidovudine
among women with CD4 cell counts over 350 cells/mm3.
The
authors note that few studies have looked at comparing the effectiveness of
these strategies within a resource-poor programme setting.
Given
that government and programme managers are having to decide which strategy to
use for PMTCT the authors chose to compare rates of HIV infection among infants
born to mothers taking ART or zidovudine in the Botswana national programme.
Between
February 2009 and April 2010 the authors enrolled consenting HIV-infected women
who had delivered live-born infants on the maternity wards of Scottish
Livingstone (a district referral hospital) and the Princess Marina
Hospital (the largest
national hospital). They prospectively followed the infants (93% of whom were
formula-fed), 60% of whose mothers took ART (258 infants) and 40% of whom took
zidovudine (170 infants) during pregnancy.
The
Botswana
national programme provides free HIV treatment including PMTCT interventions to
all its citizens. The programme also provides free infant formula. In 2009 93%
of the 32% of pregnant women who were HIV-infected took part in the PMTCT
programme.
Botswana national
guidelines at the time of the study recommended three-drug antiretroviral therapy
when the CD4 count fell below 250 cells/mm3. Pregnant women with CD4
counts above this level received AZT prophylaxis during pregnancy and
single-dose nevirapine at the time of delivery.
Infants
were followed from birth until six months of age. HIV infant testing was done
at one month and repeated at six months in breast-fed infants. Positive results
were confirmed by repeat testing. Infants who did not return for testing were
followed-up in their homes.
Among
the 415 (97%) infants for whom final HIV status could be determined, ten
infants (2.5%) became HIV-infected; nine (5.5%, 95% CI: 2.6-10.2%) in the
zidovudine group and one (0.4%, 95% CI: 0-2.2%) in the ART group.
HIV
transmission in the zidovudine group was also analysed according to the new WHO-recommended threshold for maternal antiretroviral treatment, in order to assess
the programmatic performance of giving zidovudine prophylaxis according to current WHO guidelines.
This analysis found that the
relative risk of transmission in women with CD4 counts at or above 350 was similar to the overall
rate (RR: 13.3, 95% CI: 1.6-112, p=0.007), indicating that even in women with
CD4 counts above the recommended level for treatment, zidovudine prophylaxis was associated with a much higher rate of vertical transmission than was seen in women with CD4 counts below 250 treated with triple-drug ART.
Infant
free survival until six months of age was greater in the ART group than in the
zidovudine group, 95.7% and 90.4%, p=0.040, respectively.
The
authors note their findings “do not support the equivalemce of zidovudine and
ART for prevention of MTCT”. Five of the nine infant infections in the
zidovudine group happened in mothers with CD4 cell counts equal to or greater
than 350 cell/mm3.
While
late start of ART was associated with an increased risk of MTCT, it did not
differ significantly between the two groups.
The
authors highlight that providing zidovudine beyond four to six weeks did not
lead to further reductions in viral load or to a greater proportion of women
with an undetectable viral load at delivery. Both of which, they note, are
important predictors of MTCT risk.
The
study raises operational challenges, note the authors. Delays in CD4 testing
and getting women onto ART resulted in nine women (8.3%) not getting ART before
delivery; one subsequently transmitted HIV to her infant. Only 22.7% (5) of
women who had been on zidovudine for under four weeks got single-dose
nevirapine.
However,
the authors note the primary reason in both groups for a shortened time on ART
during pregnancy was prematurity rather than a delay in starting ART. They add
“nearly one third of the newborns in this cohort were premature or low
birthweight, emphasizing the importance of improving access to neonatal
services in parallel with PMTCT programmes.”
The
authors acknowledge the non-randomised design and earlier start of ART “limits
a conclusive assessment of ART versus zidovudine.”
However,
the authors conclude “the findings of this study indicate that a strategy to provide
ART for all HIV-infected women, as is currently being piloted in Botswana, could
nearly eliminate infant HIV infection.”