Sixfold increased risk of infant HIV infection and high rates of low birth weight among women with HIV/hepatitis B co-infection

Ground Picture/

Women with high levels of hepatitis B viraemia (>106IU/ml) who have co-infection with HIV had a more than sixfold increased risk of having infants with HIV infection compared to women with HIV alone or co-infected with low levels of hepatitis B viraemia, Dr Debika Bhattarcharya of the University of California, Los Angeles, told participants last week at the Conference on Retroviruses and Opportunistic Infections (CROI 2019) in Seattle.

In this post hoc analysis of HPTN 046 (a randomised controlled trial evaluating six months of infant nevirapine vs placebo for prevention of mother-to-child HIV transmission undertaken in sub-Saharan Africa between 2007 and 2010), high levels of hepatitis B viraemia also increased the risk of having infants with poor outcomes including low birth weight.

Hepatitis B virus (HBV) is relatively common in sub-Saharan Africa with 3-9% of pregnant women living with HIV affected. Co-infection is linked to lower CD4 cell counts as well as increased HIV viral load.



The presence of virus in the blood.


hepatitis B virus (HBV)

The hepatitis B virus can be spread through sexual contact, sharing of contaminated needles and syringes, needlestick injuries and during childbirth. Hepatitis B infection may be either short-lived and rapidly cleared in less than six months by the immune system (acute infection) or lifelong (chronic). The infection can lead to serious illnesses such as cirrhosis and liver cancer. A vaccine is available to prevent the infection.


Something the immune system can recognise as 'foreign' and attack.


A pill or liquid which looks and tastes exactly like a real drug, but contains no active substance.

multivariate analysis

An extension of multivariable analysis that is used to model two or more outcomes at the same time.

Hepatitis B viral load is associated with hepatitis B transmission during childbirth, as well as poor long-term outcomes in adults including cirrhosis and liver cancer.

There is conflicting data about the effect of maternal HBV on maternal and infant outcomes despite the global burden of hepatitis. Data on HIV co-infection is sparse.

Dr Bhattarcharya and her colleagues examined the effect of maternal HBV infection on infant and maternal outcomes in HIV/HBV co-infection. They looked at a study (HPTN 046) conducted at a time when many women were not on antiretroviral therapy, so as to be able to isolate the effects of uncontrolled HBV viraemia on these outcomes. They hypothesised that maternal HBV infection would be associated with adverse maternal and infant outcomes in this cohort, many of whom presented late to care.

Maternal samples were retrospectively tested for hepatitis B surface antigen and if positive tested for hepatitis B 'e'-antigen at study entry and hepatitis B viral load at delivery. High and low hepatitis B viral load was defined as ≥106IU/ml and ≤106IU/ml, respectively.

Of the 2016 mothers with 2041 infants included in the analysis, 88 mothers  (4.3%) had HIV/HBV co-infection. Their 88 infants were evenly distributed between having received nevirapine or placebo.

The age for all women was similar with a median of 27 (IQR: 23-31) years.

As seen in other cohorts CD4 counts at entry were lower among women with high levels of hepatitis B viraemia (320 cells/mm3) compared to women with HIV alone (490 cells/mm3) or HIV and low levels of hepatitis B viraemia (434 cells/mm3), p < 0.007.

In multivariate analysis, adjusted for maternal CD4 cell count, age and maternal antiretroviral therapy, infants born to mothers with high levels of hepatitis B viraemia were more likely to be of a low birth weight (30%, 3/10) compared to those with HIV alone (10%, 194/1953) or HIV and low levels of hepatitis B viraemia (6%, 5/78), p = 0.03.

Examining maternal HBV viraemia and the threshold for low birth weight further, the authors tested and compared different models. Viral loads above 105IU/ml were associated with low birth weight.

Similarly, the likelihood of HIV infection among those with high levels of hepatitis B viraemia was greater (20%, 2/10) compared to HIV alone (4%, 53/1953) and low levels of HBV viraemia (0%, 0/78), respectively, p 0.01.  A shorter time to HIV infection was also linked with a high hepatitis B viral load.

Whereas low levels of hepatitis B viraemia were not associated with infant HIV infection, the increased risk of infant HIV infection among women with high levels was close to sevenfold: hazard ratio (HR): 6.75, p.004.

The critical point Dr Bhattacharya stressed is high levels of hepatitis B viraemia. Typically hepatitis B viral loads are not looked at but only hepatitis B surface antigen status, she added.

Hepatitis B viral load had no effect on congenital malformation, infant mortality or maternal outcomes.

Dr Bhattacharya concluded hepatitis B replication among pregnant women with HIV/HBV co-infection increases the risk of poor infant outcomes and vertical transmission. Reducing maternal hepatitis B viral load in this population before giving birth has benefits beyond prevention of perinatal hepatitis B transmission.


Bhattacharya D et al. Maternal HBV viremia is associated with adverse infant outcomes in HIV/HBV women. Conference on Retroviruses and Opportunistic Infections, Seattle, abstract 41, 2019.

View the abstract on the conference website.

Watch the webcast of this presentation on the conference website.

Update: Following the conference presentation, this study was published in a peer-reviewed journal:

Bhattacharya D et al. Maternal HBV Viremia and Association With Adverse Infant Outcomes in Women Living With HIV and HBV. Pediatric Infectious Diseases Journal, online ahead of print, November 2020.

DOI: 10.1097/INF.0000000000002980