Half-century study of E. coli infections in newborns reveals unexpected drivers of infection

New research from the University of Birmingham, in collaboration with the Amsterdam University Medical Centre (UMC) have analysed almost 50 years of bacterial samples to better understand how E. coli causes severe infections in newborn babies, revealing surprising insights into how these dangerous infections evolve over time.

The study, published in The Lancet Microbe, examined 1,790 E. coli isolates collected between 1975 and 2021, predominantly from neonates (the first four weeks of life) and infants under one year old with serious bloodstream or cerebrospinal fluid infections. 

Long-term genomic studies of E. coli infections in newborns are rare, making this dataset — spanning more than four decades — the first and most comprehensive collections of its kind.

Researchers found that antimicrobial resistance (AMR) played almost no role in shaping the population of bacteria causing these infections, in contrast to what is typically seen in infections affecting adults.

Instead, the study suggests that interactions between the bacteria and the human immune system drive changes in which strains become dominant over time.

Speaking about the study’s findings, Professor Alan McNally, Professor in Microbial Evolutionary Genomics and Head of School of Infection, Inflammation and Immunology, said: “The findings of our study have been a great surprise to us. Over 40 years of invasive E. coli infections in neonates and infants, there has been virtually zero influence of antimicrobial resistance. This suggests the dynamics that influence the E. coli causing these invasive infections in the very young are very different to those causing invasive infections in adults.”

“Instead of antimicrobial resistance being the driver of successful E. coli, in neonates it is frequent changes of important surface structures such as capsules that drive E. coli success. This has very important implications in how we can treat these severe infections in this at-risk age group and in how we might design effective vaccines."

The research also challenges the long-held assumption that K1 capsule strains dominate neonatal infections, which has influenced efforts to develop vaccines.

The study found that K1 strains accounted for around 58% of cases, suggesting vaccines targeting only these strains may protect only around half of infections.

The researchers say the findings highlight the importance of long-term genomic surveillance of pathogens worldwide to better understand how dangerous bacteria evolve. In addition, it is notable that the dataset was from the Netherlands, a country which has one of the lowest uses of antibiotics in human medicine in Europe. 

Willem van Schaik, Professor of Microbiology and Infection at the University of Birmingham and Head of Research for the School of Infection, Inflammation and Immunology, adds "Our study illustrates that antimicrobial stewardship can deliver long-term benefits by retaining a population of drug-susceptible bacteria. When these do cause infections, they will be easier to treat than infections caused by multidrug-resistant strains that are found to be more common elsewhere"

By understanding how these bacteria change over time, researchers hope the findings will help guide the development of more effective vaccines and improve protection for newborns most at risk of life-threatening bacterial infections.

This study was delivered through the National Institute for Health and Care Research (NIHR) Birmingham Biomedical Research Centre, where Professor Alan McNally co-leads the Infection and Acute care research theme – aiming to find better tests and treatments for infections, and designing better care pathways to improve patient outcomes.