Reducing maternal
viral load during pregnancy is the most important factor in preventing
mother-to-child transmission (MTCT) of HIV, results of a large study published
in The Pediatric Infectious Disease Journal confirm. Other risk factors for MTCT included insufficient food, a recent sexually transmitted infection (STI) and a history of diseases associated with immune suppression.

The research was conducted in Malawi and involved
women who did not take antiretroviral therapy (ART) until the onset of delivery. The transmission rate was 0.5% among women with a viral load below 1000
copies/ml, compared to 7.5% among women with a viral load above 10,000

“Our study
confirms that with maternal viral load suppression, most instances of perinatal
MTCT can be avoided,” comment the authors. “These findings can help inform
decisions to target populations for programs seeking to improve coverage of ART
during pregnancy, and may support efforts to recognize and treat maternal TB,
STIs and other infection prior to and during pregnancy, and efforts to improve
maternal nutritional status antenatally, to achieve worldwide elimination of
pediatric HIV infection and improve outcomes for HIV-infected pregnant women.”

An estimated
150,000 children aged 15 years and younger were newly infected with HIV in
2015, mostly due to MTCT. Transmissions from mother to child can occur in
utero, at the time of delivery or due to breastfeeding. ART during pregnancy
and delivery dramatically reduces the risk of MTCT. However, despite massive
increases in ART coverage, in 2014 an estimated 25% of pregnant women with known
HIV infection did not receive ART.

Moreover, as a large number
of HIV-positive women do not receive ART during pregnancy, it is important to
identify the full spectrum of risk factors for MTCT. Investigators from
the Breastfeeding, Antiretrovirals and Nutrition (BAN) study conducted a
retrospective analysis of the records of 2,275 HIV-positive pregnant women. 

Recruitment and
follow-up took place between 2004 and 2010. Participants were required to be at
least 14 years of age, ART naïve, planning to breastfeed and to
have a CD4 cell count of at least 250 cells/mm3. The women did not
receive any ART until the onset of labour, when the mothers and infants were
provided with a single dose of nevirapine. All the mothers also received
twice-daily zidovudine and lamivudine from the onset of labour and for seven
days after giving birth, with all infants receiving twice-daily weight-adjusted
zidovudine and lamivudine for seven days.

The infants were
tested for HIV via DNA PCR at birth and again at weeks one, two and four

Approximately a
third of women had a CD4 cell count below 350 cells/mm3 and 63% had
a viral load above 10,000 copies/ml during pregnancy. Only 35% had more than a
primary education. Anaemia during pregnancy, food insecurity and poor nutrition
were all common and observed in over 40% of participants.

There were 119
cases of MTCT, with 115 occurring in-utero, the remaining four cases in the
first two weeks of life.

Women transmitting
HIV to their infants had significantly higher median viral loads during
pregnancy than women who did not transmit (61,830 copies/ml vs. 16,529
copies/ml, p 0.001).

Of the 200 women
who had a viral load below 1000 copies/ml during pregnancy, there was only one
case of MTCT, a  transmission rate of
0.5%. Of note, there was a history of maternal TB infection in this case. The corresponding rate for women with viral loads between 1,000 copies/ml
to 10,0000 copies/ml was 1.4%, increasing to 7.5% for women with a viral load over
10,000 copies/ml.

Women passing on
HIV to their babies were also more like than non-transmitting women to have a
CD4 cell count below 350 cells/mm3 (36% vs. 30%).

Shortage of food
was more common among transmitting mothers than mothers who did not transmit
(51% vs. 41%). A history of TB was more prevalent among transmitters than
non-transmitting (6.7% vs. 2.8%), as was history of herpes zoster (14% vs. 4%). Transmitters were less likely to have any formal post-primary education than non-transmitters (25% vs. 35%).

After adjusting
for potential confounders, viral load during pregnancy was the strongest risk
factor for MTCT. Compared to a viral load above 10,000 copies/ml, a viral load
below 1,000 copies/ml was associated with a 90% reduction in the risk of MTCT
(aOR = 0.1; 95% CI, 0.01-0.4), and a viral load between 1,000 and 10,000
associated with an 80% reduction in the odds of transmission (aOR = 0.02; 95%
CI, 1.2-2.6).

Other significant
risk factors were:

  • shortage of food (aOR = 1.8; 95% CI, 1.2-2.6),
  • a
    self-reported STI in the previous twelve months (aOR  = 1.9; 95% CI, 1.0004-3.7),
  • a history of
    herpes zoster (aOR = 3.0; 95% CI, 1.6-5.6) and
  • a history of TB (aOR = 2.5; 95%
    CI, 1.1-5.7).

The investigators note that while food insecurity may be a cause of nutritional deficiencies, previous randomised trials of nutritional supplements have
not proven effective. It’s possible that food shortages in this study functioned as an
indicator of social determinants of health.

They speculate that histories of TB and herpes zoster may have caused immunosuppression and brief, unobserved increases in viral loads. STIs may be associated with genital shedding of HIV.

“Clearly, every
effort must be made to begin ART and reduce viral load in HIV-infected pregnant
women,” conclude the authors. “Our data also suggest that food insecurity,
history of diseases indicating immunosuppression, and history of other STIs may
be risk factors for perinatal MTCT.”