Gram-negative bacteria drove most infection-related newborn deaths in post-mortem sampling

Design

Prospective, multicentre, observational post-mortem surveillance study

Population

2,609 neonatal deaths at CHAMPS sites in six African countries and Bangladesh, 2016-2023

Primary outcome

Pathogen-specific causes in the causal pathway to death

Effect

Infection in 44%; Gram-negative in 74% of these; Klebsiella 41.7%, Acinetobacter 25.7%

CHAMPS is a surveillance network that investigates child deaths using clinical records, verbal autopsy and minimally invasive tissue sampling: blood, cerebrospinal fluid, swabs and needle biopsies of brain, lung and liver, tested by culture, multiplex PCR and histopathology. A panel then assigns causes. This report, published in May 2026, covers 2,609 neonatal deaths at sites in Ethiopia, Kenya, Mali, Mozambique, Sierra Leone, South Africa and Bangladesh.

Infection was in the causal pathway in 1,147 deaths (44%) and the underlying cause in 432 (17%). Gram-negative bacteria were implicated in 74% of infection-related deaths. Klebsiella pneumoniae was the commonest (42%), then Acinetobacter baumannii (26%), E. coli (10%) and group B streptococcus (6%). A third were polymicrobial. The pattern differed between presumed community-acquired and hospital-acquired infections. The panel judged more than 80% of these deaths avertable.

The pathogens that kill are largely the organisms of neonatal units, many of them resistant. The authors call for empirical regimens to be reviewed against this picture, and for prevention to focus on infection control in facilities.

• Check that your unit's empirical regimen for neonatal sepsis covers the Klebsiella and Acinetobacter isolates in your own antibiogram.
• Send blood cultures before antibiotics in every sick newborn, so that therapy can be narrowed or escalated.
• Treat hand hygiene, line care and equipment cleaning in neonatal units as part of sepsis treatment, not an add-on.
• Report clusters of Klebsiella or Acinetobacter in a neonatal unit to infection control promptly.

Why it matters

It shows that the organisms killing newborns in these settings are largely hospital Gram-negatives, which older empirical regimens may not cover.

Don't overread it

The single south Asian site was in Bangladesh; Indian pathogen patterns and resistance may differ.

The statistics, in plain English

These are proportions of deaths, not risks, so they describe what kills rather than how likely infection is. The sites were chosen for high child mortality, and Bangladesh was the only south Asian site, so the mix may differ in Indian units.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.