BMC publishes series of IDDO papers about the causes of non-malarial febrile illnesses globally

In order to map the main causes of fever in all malaria-endemic regions, researchers associated with the WorldWide Antimalarial Resistance Network (WWARN); the Infectious Diseases Data Observatory (IDDO)at the University of Oxford, the Foundation for Innovative New Diagnostics (FIND),London School of Hygiene & Tropical Medicine (LSHTM),University of Otago, University of Hong Kong, Fundação Oswaldo Cruz (Fiocruz), and the MORU Tropical Health Network - Lao-Oxford-Mahosot Hospital-Wellcome Trust Research Unit (LOMWRU) and Cambodia-Oxford Medical Research Unit (COMRU) screened more than 100,000 articles published between 1980 and 2015 to provide the most comprehensive data available to date. 

Febrile illness is the most common reason for childhood healthcare visits globally, with hundreds of millions of cases of fever in 0 to four-year-olds presenting at health facilities every year. Historically, in malaria-endemic countries it was assumed that malaria was the cause of fever, but with the advent of rapid diagnostic tests (RDTs) for malaria, combined with intensified malaria control activities over the last decade, the incidence rate of malaria has been substantially reduced. Health workers and researchers have noted that once malaria is ruled out, it is difficult to diagnose febrile illness due to limited diagnostic tools, laboratory facilities and the scarcity of comprehensive surveillance networks in low-resource settings. Thus, patients with fever are often misdiagnosed and given inappropriate treatments, such as antibiotics in the absence of a confirmed diagnosis, which in addition to mistreating individual patients may contribute to antimicrobial resistance.

Today, the ongoing pandemic of COVID-19, a disease with a potential presentation of non-specific febrile illness, further highlights the need for accurate diagnostics, strong surveillance networks and standardized data to efficiently handle this disease alongside addressing the multitudes of causes of potentially co-existing NMFIs.

In light of this background, these reviews present a comprehensive assimilation of the published literature on potential aetiologies of NMFIs in malaria-endemic areas. Data from the reviews show variation in reporting on infectious causes of fever across different geographies and age groups.

The results of these extensive literature reviews have been incorporated into an open access, online database that feeds into an interactive map that can filter data by country, microorganism type, patient age, sample type, pathogen family, genus and species, study year and clinical disorder. The interactive map has been designed to support scientists, researchers and those involved in developing policy and medical guidelines. In addition, the mapping team welcome new data that can be added in to the map as research moves ahead. To get in touch email info@iddo.org.

Series co-ordinator, IDDO Director Professor Philippe Guérin, said: “The uptake of malaria rapid diagnostic tests has revealed a major problem lurking in the background ­-that children and adults presenting with fever and malaria-like symptoms often prove not to have malaria.

“Unfortunately, actionable tests are not widely available to guide clinicians on how to manage the hundreds of millions of cases of non-malarial fever and ultimately there’s a high incidence of under-diagnosis. More work is needed in the priority areas of diagnostics research and development, and improved surveillance.”

Sabine Dittrich, Head of Malaria and Fever, FIND, said: “It is great to see FIND’s longstanding efforts to map the causes of non-malarial fevers around the world help guide the use and development of diagnostics which in turn will provide more evidence based care to patients around the world” 

In Southern Asia and South-eastern Asia, 29,558 records from 19 countries were screened and 2,410 met the selection criteria. Commonly reported bacterial causes were (in descending order of frequency) Salmonella Typhi, Escherichia coli, Staphylococcus aureus, Klebsiellapneumoniae, and coagulase negative Staphylococcus. While dengue virus, chikungunya virus, Japanese encephalitis virus, hepatitis B virus, and hepatitis C virus were common viral infections. The findings emphasise the need to standardise the reporting of aetiological studies to develop effective, evidence-based fever management guidelines and improve surveillance.

Lead author Poojan Shrestha said: “Comprehensive standardised reporting and sharing of data, ideally in real-time, through collaborative international effort is imperative to tackle the formidable burden of mortality and morbidity posed by non-malarial febrile illnesses, especially in low-resource settings.”

Work for the sub-Saharan Africa paper screened16,523 records from 48 African countries of which 1,083 met selection criteria. Staphylococcus aureus, non-typhoidal Salmonella, and E. coli were commonly reported bacterial infections, and yellow fever virus, Rift Valley fever virus, and Ebola virus disease were the most commonly reported viral infections. Dengue virus infection, previously not thought to be widespread in Africa, was reported in 54 of the articles. The paper concludes that as the threat of antimicrobial resistance looms, knowledge of the distribution of infectious agents causing fever, development of new diagnostic tools, and improved antimicrobial stewardship should be priories.

Dr Heidi Hopkins, Associate Professor at London School of Hygiene & Tropical Medicine, said: “This review provides a foundation from which to work toward a better understanding of causes of fever in Africa. Moving ahead, prospective studies and surveillance, with harmonized reporting of fever aetiology data, will help to answer many remaining questions and improve our ability to care for patients with both mild and severe febrile illnesses.”

The Latin American systematic review included a total of 625 publications from across 34 countries. The most frequently reported virus was dengue virus, followed by other arboviruses and hantavirus. For bacteria, Staphylococcus spp., Rickettsia spp., and Leptospira spp. were frequently reported. The review found that the cause of fever in Latin America was predominantly attributed to viruses or bacteria, with dengue being the most frequently reported pathogen. The paper highlighted the need for standardised protocols for reporting the causes of fever to help ensure better comparability of results and improved interpretation; to improve existing epidemiological surveillance networks and prioritise the development of diagnostic tools available in the region.

Andre Siqueira, Infectious Diseases Consultant and Public Health Researcher at Instituto Nacional de Infectologia Evandro Chagas Fiocruz, said: “The Latin America harbours a wide diversity of socioeconomic and environmental scenarios, which are receptive to a diverse array of infective agents, as depicted by the explosive emergence of zika virus in the region. Our review highlights the need for harmonized protocols for establishing the aetiology of febrile disease, what can enable health systems to define better syndrome-based case-management guidelines and improved surveillance protocols to detect and respond to prevalent or emerging pathogens.”        

Microbiology Investigation Criteria for Reporting Objectively (MICRO) framework

At the time of the review, guidelines did not exist to address the structure and reporting parameters of an epidemiological study reporting microbiological data. To address this challenge, the MICRO framework was developed by a working group of clinical and laboratory microbiologists, infectious disease physicians, epidemiologists and mathematical modellers working in the UK and various LMICs. This checklist will support the establishment of reporting standards, reduce heterogeneity of data and facilitate the collation and analysis of the data to inform policy, guidelines and appropriate treatment responses. Download the MICRO framework checklist.

Lead author Associate Professor Paul Turner, Director of the Cambodia Oxford Medical Research Unit, said: "Microbiology and antibiotic susceptibility data is often difficult to interpret unambiguously due to vague or incomplete descriptions of methodology. Use of the MICRO framework will enable researchers to report and use such data with greater clarity and confidence moving forwards".

Read more about Febrile illness mapping - much of the world without data and without evidence-based treatments.

Find out more about IDDO