Fluid Expansion as Supportive Therapy Trial (FEAST)
Study acronym
FEAST
Background
Severe infections such as malaria and sepsis remain important causes of childhood death in sub-Saharan Africa. Many deaths occur within the first 24 hours after hospital admission.
In high-resource settings, children with severe infection and shock commonly receive rapid intravenous fluid boluses to improve blood circulation. Before FEAST, there was limited evidence about the safety and effectiveness of this practice in African hospitals without intensive-care facilities.
Study purpose
FEAST investigated whether rapid intravenous fluid administration could reduce early mortality among critically ill African children with severe febrile illness and clinical evidence of impaired circulation.
The trial also compared albumin and saline as fluids for resuscitation.
Study design
FEAST was an open-label, multicentre, randomised controlled trial conducted in six hospitals across Uganda, Kenya and Tanzania.
Children without severe hypotension were randomly assigned to receive:
- A bolus of 5% human albumin solution;
- A bolus of 0.9% saline solution; or
- Maintenance fluids without a fluid bolus.
Most children assigned to a bolus received 20 ml/kg over one hour. A second bolus could be administered when signs of impaired circulation persisted.
Children with severe hypotension were assigned to either albumin or saline because withholding a bolus from this group was considered inappropriate.
Study population
The trial enrolled 3,170 children aged over 60 days and under 12 years. Participants had severe febrile illness, impaired consciousness or respiratory distress, and clinical evidence of impaired circulation.
Malaria was the most common diagnosis, although bacterial infections and other serious illnesses were also represented. Children with severe malnutrition, gastroenteritis, burns, trauma or substantial blood loss were excluded. UCL Innovative Clinical Trials Unit
Study sites
The trial was conducted at:
- Mbale Regional Referral Hospital, Uganda
- Soroti Regional Referral Hospital, Uganda
- Mulago Hospital, Kampala, Uganda
- St Mary’s Hospital Lacor, Uganda
- Kilifi District Hospital, Kenya
- Teule Hospital, Tanzania
Mbale Regional Referral Hospital’s participation made FEAST an important part of MCRI’s history of collaborative paediatric emergency-care research.
Study leadership
Professor Kathryn Maitland was the Chief Investigator.
Key Ugandan investigators included Professor Sarah Kiguli, Dr Robert Opoka, Dr Charles Engoru and Professor Peter Olupot-Olupot. The study was implemented through the FEAST Trial Group and supported by collaborating investigators and clinical teams across the three participating countries.
Sponsor and funder
Imperial College London sponsored the trial.
The United Kingdom Medical Research Council funded the study through the MRC–Department for International Development concordat under grant G0801439. Baxter Healthcare donated the resuscitation fluids but did not participate in the study’s design, analysis or publication decisions. FEAST follow-up analysis
Primary outcome
The primary outcome was death from any cause within 48 hours after randomisation.
Secondary outcomes included mortality at four weeks, neurological complications, adverse events and clinical responses to the assigned fluid strategy.
Principal findings
The study found that children who received albumin or saline fluid boluses had higher mortality than children who received maintenance fluids without a bolus.
By 48 hours, mortality was approximately 11% in both bolus groups compared with 7% in the maintenance-fluid group. Fluid boluses therefore produced an absolute increase in mortality of approximately three deaths for every 100 children treated.
The increased mortality occurred with both albumin and saline, with no evidence that either bolus fluid was safer. The trial was stopped early in January 2011 following advice from the independent Data and Safety Monitoring Committee. New England Journal of Medicine article
Importance of the study
The FEAST findings challenged the assumption that rapid fluid resuscitation was beneficial for all children with severe infection and impaired circulation.
The results demonstrated that fluid-management practices developed in well-resourced intensive-care settings could not automatically be transferred to resource-limited hospitals. Subsequent analysis found a 3.3% absolute increase in mortality by 48 hours and a 3.4% increase by 28 days among children receiving boluses. Published follow-up analysis
The results do not apply to children who require fluid replacement because of diarrhoeal dehydration, burns, trauma or major blood loss, as these groups were excluded from the trial.
Study period
The trial began in January 2009 and was stopped early in January 2011 because of evidence of harm in the fluid-bolus groups.
The principal findings were published in the New England Journal of Medicine in June 2011. PubMed record
Trial registration
The trial was registered as ISRCTN69856593 on 15 December 2008.
Ethics and regulatory approvals
Approvals were obtained from:
- Imperial College Research Ethics Committee
- Mbale Research Ethics Committee
- Uganda National Council for Science and Technology
- KEMRI Scientific and Ethics Review Unit
- Kenya Poisons and Pharmaceuticals Board
- Tanzania Food and Drugs Authority
Study status
FEAST is a completed study. Recruitment was closed in January 2011, and the principal results and subsequent analyses have been published.