Implementing new recommendations for prevention of
mother-to-child transmission (PMTCT) using antiretroviral drugs has resulted in a
significant reduction in new infant HIV infections in Zambia, a study of one
district, presented at the 19th Conference on Retroviruses and
Opportunistic Infections in Seattle, shows.
However the impact of improved regimens is being diluted by
high-levels of loss to follow-up among mothers receiving triple-drug
combination therapy, highlighting the need for improved mechanisms to keep
mothers in care after childbirth and to improve adherence to treatment.
A large review of pregnant women in South Africa presented
at the conference also found high rates of loss to follow-up within three
months of entering care.
The World Health Organization issued new
recommendations on the use of antiretroviral drugs for prevention of
mother-to-child transmission in 2009, encouraging countries to adopt triple-drug
antiretroviral therapy during pregnancy and breastfeeding even if the mother is
not eligible for antiretroviral treatment for her own health, wherever this is
affordable ('Option B').
The alternative option ('Option A') is to use a more limited
regimen of zidovudine from week 14 of pregnancy, single-dose nevirapine for
mother and infant around the time of delivery, and zidovudine with 3TC during delivery and for one week after
delivery.
Although clinical trials have shown that triple-drug therapy
is highly effective in reducing the risk of mother-to-child transmission, there
is limited information on the real-life effectiveness of the various
recommended options for PMTCT.
Researchers from the Center for Infectious Diseases
Research, Zambia, designed a controlled observational cohort study in the rural
Kafue district, around 60 miles outside the Zambian capital Lusaka.
The study compared the performance of universal triple-drug
therapy for all pregnant and breastfeeding women at four sites with five
control sites where women not eligible for antiretroviral therapy for their own
health received a modified version of the WHO Option A regimen.
The study was designed to reproduce real-world conditions as
far as ethically possible; women were seen every two to four weeks until
delivery, and infants were tested for HIV DNA by PCR test at 6 weeks, 6 months
and 12 months after birth. Staff at participating centres attempted to contact
any mothers who missed appointments three times before mothers were classified
as lost to follow-up.
The study measured two outcomes:
- Infant HIV infection, or death
- Infant HIV infection, death or maternal loss to
follow-up (“programme failure”). This outcome was designed to capture the
extent to which loss to follow-up differed between study sites, and whether
this was attributable to the type of PMTCT regimen being provided.
One hundred and forty-three women were recruited at the
intervention sites, and 141 at the control sites. Women at the control sites
had significantly higher CD4 cell counts at baseline (440 cells/mm3
vs 345 at the intervention sites, p<0.001) and were somewhat more likely to
be asymptomatic, having WHO stage 1 HIV disease (p = 0.05).
Women at the control sites were more likely to stop
breastfeeding in the first year after delivery; approximately half of women at
the control sites were still breastfeeding after one year, compared to
approximately 85% in the intervention group (who were still receiving
triple-drug ART while breastfeeding) (p<0.001). The majority of infant
infections at the control sites occurred during 24 and 60 weeks after delivery.
Only one infection occurred in the intervention group,
detectable by HIV DNA testing at week 6. At week 6 there was no significant
difference in the risk of mother-to-child transmission or death, but by month 6
the risk of transmission or death was significantly greater among women at the
control sites (13.5% vs 3%, relative risk 3.67, 95% confidence interval 1.50-9.02). This difference was sustained at 12 months (19.5% vs 5.8%, RR 3.44, 95%
CI 1.56-7.59).
However the inclusion of loss to follow-up as part of the
combined outcome measure of programme success resulted in a less favourable
assessment of the performance of triple-drug therapy as PMTCT at months 6 and
12. In neither case was there a significant difference in performance, due to
higher rates of loss to follow-up at the intervention sites compared to the
control sites at these time points.
Despite the superior effectiveness of triple-drug therapy in
preventing new infections and death, the investigators concluded that patient
loss to follow-up poses a significant threat to the effectiveness of PMTCT
programmes, and called for greater attention to tailored interventions that can
improve retention in care and adherence to treatment.
A second study, presented by Landon Myer of the University
of Cape Town on behalf of the IeDEA collaborative epidemiology group, reviewed
rates of loss to follow-up among women with HIV who initiated antiretroviral
therapy at six large public sector HIV clinics in South Africa between 2002 and
2009. The study reviewed 29,653 women who started triple combination ART based
on South African national guidelines during the period.
Although pregnant women who received ART were significantly less
likely to die during the first year of treatment compared to non-pregnant women
(3% vs 9%, p<0.001), pregnant women were significantly more likely to be
lost to follow-up (19% vs 11%, p<0.001), and this trend was consistent
across all sites.
Loss to follow-up was more common in women under 30 and in
women with higher CD4 cell counts in pregnant women who started ART during
pregnancy.
Landon Myer said that it was still unclear whether loss to
follow-up was occurring post-partum or before delivery, and that more research
was needed into the reasons for loss to follow-up, as well as the design of
interventions to reduce loss to follow-up rates in pregnant women on ART,
especially as guidelines expand the population of women eligible to begin ART
during pregnancy.