Main Article Content

Maternal outcomes following introduction of antiretroviral therapy in the public sector


E Bera

Abstract

Background. Antiretroviral therapy (ART) has been available in the public sector since 2004, but limited published data exist on maternal outcomes in South Africa following its introduction.
Objectives. A prospective study of maternal outcomes after the introduction of ART at Frere Hospital, East London, E Cape.
Methods. Pregnant women with indications for lifelong ART were commenced on treatment, and followed up until 6 weeks after delivery. Baseline demographic details were collected, including details on the gestation and mode of delivery. The mothers and their infants were then referred to local ART centres for continued care. Maternal outcomes measured were maternal death, and predefined maternal morbidity. Results were analysed using Epi-info software version 3.3.2 (2005).
Results. Data on 385 women are presented. The women initiated ART at a median age of 28 years. Median gestation at commencement of ART was 30 weeks. The median CD4 count and HIV-1 RNA viral load (VL) were 173 cells/μl and 4.56 log10 copies/ml, respectively. Fifty-five (14.3%) and 19 women (4.9%) had World Health Organization (WHO) stage 3 and stage 4 disease, respectively. Twenty-five (6.5%) were concurrently on treatment for tuberculosis (TB) while 10% had other co-infections – meningitis, hepatitis, pneumonia and urinary tract infections. Immune reconstitution inflammatory syndrome (IRIS) occurred in 7.0% of cases, and pre-eclampsia developed in 7.5%. Median gestation at delivery was 39 weeks. Seven maternal deaths (1.8%) occurred in this cohort. All 7 women died in the postpartum period. Five women died within 5 weeks of commencing ART. Two women died of Pneumocystis jirovecii pneumonia (PCP) and another 2 died from liver failure – 1 death was presumed to be from lactic acidosis related to stavudine (d4T) toxicity. The 3 remaining deaths were due to PCP IRIS, meningitis and TB, respectively. The strongest predictor of maternal death was WHO stage 4 disease (p=0.0006). Maternal plasma VL >5.00 log10 copies/ml, pre-eclampsia, IRIS, concurrent ART and tuberculosis treatment, and CD4 count <100 cells/μl were not significantly predictive of maternal death on multivariate analysis.

Conclusion. Despite the availability of ART in the public sector, maternal mortality remains a concern. Reasons include late entry of pregnant women into care, as well as suboptimal management of women initiated on ART who develop opportunistic infections. There is an urgent need to develop national ART guidelines specifically for pregnant women.

 


Journal Identifiers


eISSN: 2305-8862
print ISSN: 0038-2329