Start main page content

Soweto study informs global network on technique to discover why children die

- Wits University

A Wits-led pilot study in Soweto has shown that minimally invasive tissue sampling (MITS) can accurately determine cause of stillbirth in South African women.

The observational pilot study at Chris Hani Baragwanath Academic Hospital informed the rollout of MITS in low- and middle-income (LMIC) countries globally via CHAMPS – the Child Health and Mortality Prevention Surveillance Network.

The absence of data that reliably characterise specific causes of death in children under-five has severely hampered progress in preventing childhood mortality.

The Soweto Stillbirths Study

Despite approximately 2.6 million stillbirths occurring globally each year – 95% in LMICs – data and knowledge on the biological causes of these deaths in LMICs is limited.  

Currently, causes of death in children under-5 in LMICs are determined by vital registration data and verbal autopsies. However, these methods do not attribute death from infection-related illnesses (such as pneumonia, sepsis, and diarrhoea) to a specific pathogen.

Professor Shabir Madhi, who is the Director of the Medical Research Council Respiratory and Meningeal Pathogens Research Unit (RMPRU) at Wits, led a pilot study to evaluate the effectiveness of MITS in determining the cause of stillbirths in South African women.

Professor Shabir Madhi

The Soweto stillbirths study investigated the utility of the minimally invasive tissue sampling technique – along with examination of the placenta and clinical history – in attributing stillbirth in a South African LMIC setting. The technique was also used to understand specific causes of death in very young children.

“Minimally invasive tissue sampling can pinpoint pathological causes of infant death, the findings of which can then inform policy, and more effective interventions to mitigate child mortality,” says Madhi.

MITS identifies leading causes of Soweto stillbirths

Sampling of brain, lung, and liver tissue using biopsy needles, blood and cerebrospinal fluid collection, as well as an examination of the placenta revealed a cause of stillbirth for 117 of the 129 cases (90.7%).

The leading underlying causes of stillbirth were maternal hypertensive disorders [high blood pressure]; placental separation and haemorrhage; and chorioamnionitis [inflammation of the fetal membranes due to a bacterial infection].

The leading immediate causes of foetal death were antepartum hypoxia [abnormally low levels of oxygen in the blood not long before birth] and foetal infection due to E.coli, enterococcus, and Group B Streptococcus bacterial infection.

“In addition to indicating the potential of our approach in ascertaining granular causes of stillbirth in LMIC settings, the findings provide insight into the pathogenies of stillbirths in settings such as ours. This included the potentially under-appreciated dominant role of foetal infection as the cause of stillbirth in 37.2% of cases,” says Madhi, who, along with Dr Jayani Pathirana and Dr Vicky Baillie in the Wits RMPRU co-authored a related paper unravelling specific causes of neonatal mortality using MITS.

Hospital-acquired multidrug-resistant bacterial infections cause neonatal deaths

This prospective, observation pilot study enrolled neonatal deaths at Chris Hani Baragwanath Academic Hospital. According to the paper, MITS included needle core-biopsy sampling for histopathology [the study of changes in tissues caused by disease] of brain, lung, and liver tissue. Microbiological culture and/or molecular tests were performed on lung, liver, blood, cerebrospinal fluid, and stool samples. The ’underlying’ and ‘immediate’ causes of death were determined for each by an international panel of 12-15 medical specialists.

Infections were the immediate or underlying cause of death in 57.5% of all 153 neonatal deaths in this study. Overall, 74.4% of 90 infection-related deaths were hospital-acquired, mainly due to multidrug-resistant bacteria. Group B streptococcus (GBS) was the most common pathogen among deaths with ‘infections’ as the underlying cause.

“An overwhelming find from our study was the dominant but under-emphasised role of hospital-acquired infections to in-facility neonatal deaths, especially in late neonatal death [deaths between 7 and 27 days]. The observation on the contributions of hospital-acquired infection as a cause of neonatal deaths in this study have underpinned a re-evaluation and introduction of strategies aimed at improving infection control practices at the hospital, the success of which will be evaluated through the ongoing CHAMPS programme here in Soweto,” says Madhi.

Illuminating child mortality: Discovering why children die

Rather than provide conclusive data on the causes of stillbirths in the Soweto stillbirths pilot study, the main objective was to evaluate the utility of MITS as a diagnostic tool. The Soweto stillbirths pilot study preceded the multicentre Child Health and Mortality Prevention Surveillance (CHAMPS) programme now underway.

The Bill and Melinda Gates Foundation fund CHAMPS, which aims to generate mortality data in a way that is designed to enable stakeholders at local, national, regional, and global levels to prioritise the most effective strategies and invest in the most needed tools.

This objective responds to Sustainable Development Goal 3.2: The elimination of all preventable under-5 mortality and stillbirths by 2030 and reducing global under-5 mortality to a maximum of 25 deaths per 1000 live births (from its current rate of 44 deaths per 1000 live births).

Targeting the specific causes of child death will require more precise and robust data than have previously been available.

A paper on the Soweto study was part of 13-article CHAMPS supplement published in the Clinical Infectious Diseases journal in October 2019, and co-authored by Madhi, Dr Pratima L. Raghunathan, Director for Science, CHAMPS network, Emory Global Health Institute, and Dr Robert F. Breiman, Executive Director, CHAMPS, and Director, Emory Global Health Institute.

The supplement, entitled The Child Health and Mortality Prevention Surveillance (CHAMPS) Network: Foundational methods, comprises articles that feature a variety of methods to describe why children die in the parts of the world where children die too often.

According to the supplement, current understanding of the causes of under-5 childhood deaths in LMICs relies heavily on country-level vital registration data and verbal autopsies. The CHAMPS network aims to systematically describe causes of child death and stillbirth in LMICs using minimally invasive tissue sampling.

The articles in the supplement introduce a range of methods to identify cause of death in children under 5 years. These data are crucial to locally target interventions more effectively and rapidly reduce under-5 mortality.