Maternal death rate five times higher in women with HIV, South African audit shows

This article is more than 12 years old. Click here for more recent articles on this topic

Maternal mortality ratios in Johannesburg, South Africa in HIV-infected women are more than six times higher than in HIV-negative women despite integration of antiretroviral treatment into prenatal services, reported Vivian Black and colleagues in a five-year audit published in the August 2009 issue of Obstetrics and Gynecology.

If the United Nation’s Fifth Millennium Development Goal (MDG) of reducing maternal mortality by 20% by 2015 is to be reached the causes of maternal mortality as well as of preventable contributing factors need to be clearly understood.

The global maternal mortality ratio of 400 per 100,000 live births as estimated by the World Health Organization does not reveal the considerable regional variations.


mother-to-child transmission (MTCT)

Transmission of HIV from a mother to her unborn child in the womb or during birth, or to infants via breast milk. Also known as vertical transmission.


Social attitudes that suggest that having a particular illness or being in a particular situation is something to be ashamed of. Stigma can be questioned and challenged.


Any lung infection that causes inflammation. The infecting organism may be bacteria (such as Streptococcus pneumoniae), a virus (such as influenza), a fungus (such as Pneumocystis pneumonia or PCP) or something else. The disease is sometimes characterised by where the infection was acquired: in the community, in hospital or in a nursing home.


A healthcare professional’s recommendation that a person sees another medical specialist or service.

middle income countries

The World Bank classifies countries according to their income: low, lower-middle, upper-middle and high. There are around 50 lower-middle income countries (mostly in Africa and Asia) and around 60 upper-middle income countries (in Africa, Eastern Europe, Asia, Latin America and the Caribbean).

In 2005, South Africa – a middle-income country  –a ratio close to the global average but considerably higher than countries of similar gross domestic product per person, for example, Portugal and Brazil which had a ratio of 11 and 110 respectively.

Since 1998 HIV, an indirect cause of maternal mortality, has been the leading contributor to maternal mortality in South Africa, reversing previous declines seen in maternal mortality rates.

An audit of maternal deaths for the period 1996 to 1998 in Durban, South Africa showed facility-based maternal mortality rates for women with HIV to be 323 per 100,000 compared to 148 per 100,000 for those not infected – over two times higher. Co-infection with tuberculosis had a considerable impact on outcomes.

The authors noted that HIV prevalence among women attending antenatal clinics has remained steady at between 28% and 33% over the past four years.

The authors reviewed maternal deaths at a tertiary level facility in Johannesburg, Gauteng Province, for the five-year period from 2003 to 2007. Variables of interest included: deaths due to HIV, the patterns of these deaths and changes over time. Antiretroviral therapy became available in 2004 and was integrated into an existing programme for prevention of mother to child transmission (PMTCT) in the prenatal clinic of the hospital at the facility. The authors assess its impact on maternal mortality.

HIV testing and counselling is offered at the first prenatal visit and CD4 cell counts done for those who test positive. Eligibility for antiretroviral treatment is based upon a CD4 cell count <200 cells/mm3or WHO clinical stage 4. Women receive the standardized ART regimen of stavudine, lamivudine and nevirapine, along with cotrimoxazole prophylaxis. Single-dose nevirapine (for both mother and infant) was given for prevention of mother to child transmission during the period under review, prior to the updating of South African guidelines.

Patient case files, birth registers, death certificates and mortality summaries were reviewed. Maternal death was defined as death of a woman at the facility during pregnancy or within 42 days of childbirth. No information was available for women who died at home or at another facility. Cause of death was determined through multidisciplinary clinical case discussions. Annual maternal mortality ratios were calculated and disaggregated by HIV status.

For the period 2003 to 2007 a total of 108 women died and had a mean age of 28.7 years. It was the first pregnancy for eleven percent, a third of the proportion of all women in their first pregnancy delivering at the hospital.

HIV test results were available for 72 % (76); almost 80% were HIV positive. The median CD4 cell count for 53 of HIV infected women who had the test was 72 cells/mm³ (interquartile range: 29-194 cells/mm³).

Only two of the HIV-infected women had begun antiretroviral therapy. The authors note this clearly demonstrates that missed opportunities for starting treatment persist.

Most deaths were associated with advanced HIV disease, the most common causes being tuberculosis (36%) and pneumonia (20%). Median CD4 count in women whose death was due to an HIV-related illness was 50 cells/mm³ compared to a median of 335 cells/mm3 in HIV-infected women who died of non-related HIV causes.

The authors argue that most of the HIV-related deaths could have been avoided if ART and cotrimoxazole prophylaxis had been started. In HIV-negative women or those of unknown status deaths were overwhelmingly due to obstetric causes with hypertension accounting for over 50%.

While the number of deaths over the five year period ranged from 15 to 25, the number of live births remained constant at around 7,000 per year.

The authors note that while coverage of HIV testing increased each year (women in 2007 were 3.4 times more likely to have a known HIV status (95% CI 3.2-3.6) than those in 2003) HIV testing and follow-up after diagnosis remained the most significant programmatic weakness.

The authors suggest that systematic evaluation of the processes of HIV testing and care for pregnant women could be useful and might include: “assessment of performance against predefined criteria and agreed targets, for example, routine provider-initiated HIV testing, provision of CD4 results at the second prenatal visit, a target time of three weeks from first visit to antiretroviral treatment initiation and active follow-up processes to ensure that women with advanced HIV disease are retained in care”. The authors highlight the use of mobile phone technologies as an effective means of follow-up in this setting.

While the numbers of women who began ART increased over time, with coverage in 2007 estimated at 59.2%, the maternal mortality ratio for HIV-infected woman was over six times higher than the ratio in HIV-negative women (776 versus 124 per 100,000), or 6.2 (95% CI 3.6 to 11.4).

In total, 44.3% (95% CI 30.8 to 54.8%) of deaths were due to HIV. The mortality rate among HIV-infected women who died of non HIV-related causes was 171 out of 100,000, similar to that of HIV-negative women.

The authors note the importance of expansion of ART to two primary health facilities close to the hospital led by nurses and midwives. Advantages include ensuring high level coverage, bringing HIV care closer to the patient and potentially avoiding unnecessary referrals to tertiary care. They stress that integration of ART into prenatal care will help secure both the health of women as well as prevent transmission to newborns. Others factors the authors cite as barriers to uptake of prevention of mother to child transmission are: the framing of PMTCT itself as a paediatric issue, weak health systems, poor communication between health workers and pregnant women as well as the fear of stigma.

Nearly three-quarters of all deaths occurred in the week after childbirth, and the authors stress the importance of strengthening postnatal health services.

The authors note that statistically, maternal deaths are rare, which makes understanding the effects of new interventions difficult. Comparison of facility-based deaths with other institutions may not be feasible since a disproportionately higher number of difficult pregnancies are referred to them increasing the probabilities of a higher mortality rate. The authors note too that their findings may be the result of changes over time and not necessarily the introduction of ART.

They conclude, “Although it was not possible to demonstrate that the integration of antiretroviral treatment within prenatal care services has reduced maternal deaths, provision of antiretroviral treatment into prenatal care remains an important strategy for reducing maternal deaths in high HIV-burden countries. Interventions, such as clinical audits, are required to target weaknesses in HIV treatment and care within maternal health services.”


Black V et al. Effect of human immunodeficiency virus treatment on maternal mortality at a tertiary center in South Africa: A 5-year audit. Obstetrics and Gynecology 114 (2), 292-299, 2009 (full text freely available here).