The number of cases of childhood diarrhoea attributable to pathogens (bacteria, parasites, viruses or other infections) have been substantially underestimated and may be nearly twice as high as previous analysis suggests, according to new research published in The Lancet.
The analysis of over 10000 samples from Bangladesh, India, Pakistan, The Gambia, Kenya, Mali, and Mozambique finds that Shigella and rotavirus were the most common infections among children under 5 years old, followed by adenovirus, enterotoxin-producing E coli (ETEC), Cryptosporidium, and Campylobacter. While an oral vaccine for rotavirus exists, the authors say that the findings highlight the need for prioritisation of Shigella and ETEC vaccines.
Worldwide, diarrhoea remains the second leading cause of death in children under 5 years old, and is associated with approximately half a million deaths per year. Management of diarrhoea usually involves supportive care with rehydration and zinc supplementation. Antibiotics are not usually recommended unless there are signs of dysentery (severe diarrhoea with the presence of blood and mucus in the stools) or if cholera is suspected.
The findings come from a re-analysis of samples from the Global Enteric Multicenter Study (GEMS). Previous estimates of the infectious causes of diarrhoea were based on a variety of different detection methods, but this study, for the first time, uses a molecular diagnostic testing method called quantitative real-time PCR (qPCR) to test for 32 pathogens.
Researchers re-analysed stool samples from 10608 children with and without diarrhoea obtained from regions in seven countries in Asia (Bangladesh, India and Pakistan) and Africa (The Gambia, Kenya, Mali, and Mozambique).
The original GEMS study1, published in 2013, estimated that 51.5% of childhood diarrhoea cases could be attributed to pathogens but the new re-analysis finds the proportion is much higher at 89.3%. The original study identified four major pathogens: rotavirus, Shigella spp, Cryptosporidium spp, and heat-stable enterotoxin-producing E coli (ST-ETEC). This re-analysis reaffirmed these four and added two others: adenovirus 40/41 and Campylobacter jejuni/coli. Together, these six pathogens accounted for 77.8% of all diarrhoea. Among the children who had a diarrhoea-causing pathogen, about half had more than one infection (figure 4), highlighting the challenges of treating multiple infections.
"Childhood diarrhoea remains an enormous problem, made more confusing by the long list of possible infections, the difficulties in diagnosis, and the large number of undiagnosed cases," says lead author Professor Eric R Houpt, University of Virginia, Charlottesville, USA.
"We developed sensitive quantitative methods that levelled the diagnostic playing field and closed the diagnostic gap. Sadly, the "healthy" control children carried on average four infections, so determining the cause of diarrhoea is tricky. We found that when an infection replicates and reaches a certain threshold, then diarrhea happens. Our study was able to identify what that threshold is, pathogen by pathogen. In so doing, we concluded that six infections constitute the vast majority of disease, and these should be heavily prioritised. An oral vaccine for rotavirus exists, and development of vaccines for Shigella and ETEC are underway. Cryptosporidium, Campylobacter, and adenovirus now need a lot more attention."2
Other pathogens including norovirus GII, sapovirus, astrovirus, H. pylori, Vibrio cholerae, typical enteropathogenic E coli (tEPEC), Salmonella spp, Entamoeba histolytica, and Cyclospora cayetanensis were also associated with diarrhoea but were less common (figure 1). Several other pathogens were not associated with diarrhoea.
The authors say that the study sites were carefully selected to be broadly representative of countries with moderate to high child mortality, but some differences in the hierarchy of pathogens were observed between countries (table 2).
The authors point to some limitations, including that the analysis may underestimate pathogens that are shed with high frequency, so longitudinal studies will be needed to further understand these pathogens. Additionally, although the findings should provide evidence to prioritise vaccines for certain pathogens, vaccine development relies on subtyping of infections which was not provided in this study. Finally, the authors warn that qPCR does not assess antimicrobial resistance, and therefore the technology should be used in conjunction with conventional culture methods to detect antibiotic resistant pathogens.
Writing in a linked Comment, Dr Karen Keddy, Centre for Enteric Diseases, National Institute for Communicable Diseases, Johannesburg, South Africa and co-authors write: "These results imply that prioritising vaccine development for these six pathogens could lead to a substantial decrease in diarrhoea burden among children younger than 5 years over the next few decades, as has been seen for rotavirus. Follow-up studies will indicate which molecular diagnostics complement traditional methods." They add: "Liu and colleagues raised a concern that the burden of watery diarrhoea associated with Shigella spp is greater than had been previously recognised. Treatment recommendations for infection with Shigella spp are skewed towards dysentery, emphasising blood in the stool as an indication for treatment and, therefore, we might currently be undertreating patients with shigellosis."
The study was funded by the Bill & Melinda Gates Foundation.
Article: Use of quantitative molecular diagnostic methods to identify causes of diarrhoea in children: a reanalysis of the GEMS case-control study, Prof Eric R Houpt, MD et al., The Lancet, doi: 10.1016/S0140-6736(16)31529-X, published online 22 September 2016.
Comment: GEMS extend understanding of childhood diarrhoea, Karen H Keddy, Anthony M Smith, Nicola A Page, The Lancet, doi: 10.1016/S0140-6736(16)31664-6, published online 22 September 2016.