The FEAST trial: A Practice-Changing Study from a Resource-Limited Area

Published - Written by Jennifer Zeis

Fluid boluses for shock are an international standard of practice. For a febrile child in shock first-hour fluid resuscitation (60 ml/kg of isotonic fluid within 15 minutes of shock diagnosis) is recommended by the American College of Critical Care Medicine, while the World Health Organization further specifies that a fluid bolus be used in children if advanced shock is accompanied by weak, fast pulse plus cold extremities.

But in some hospitals in sub-Saharan Africa, these boluses are not a part of the local standard of care. And in many hospitals in that region, intensive care facilities aren’t available, and triage and emergency care are largely ignored by child-survival programs. Also, most children in these hospitals do not receive specific fluid management aside from blood transfusion or maintenance fluids.

So, Kathryn Maitland of the Kenya Medical Research Institute (KEMRI)–Wellcome Trust Research Programme and colleagues set out to evaluate the administration of fluid boluses for shock with the intent of generating practical data for these resource-poor settings where malaria is endemic.

The results were not what the researchers had expected. The saline and albumin boluses, given to children in sub-Saharan Africa with severe febrile illness and impaired perfusion, were doing more harm than good.

In the Fluid Expansion as Supportive Therapy (FEAST) trial, published recently in NEJM, Maitland et al. randomized 3,141 children over 60 days old with severe febrile illness and impaired perfusion to one of three arms: 20 to 40 ml/kg of 5% albumin, 0.9% saline boluses, or no bolus at all at hospital admission in six sites located in Uganda, Kenya, or Tanzania. All children received maintenance fluids and the standard of care recommended by guidelines. The primary endpoint was 48-hour mortality.

Forty-eight-hour mortality was 10.6% and 10.5% in the albumin-bolus and saline-bolus groups, respectively, and 7.3% in the control group. Four-week mortality rates were 12.2%, 12.0%, and 8.7%, respectively. Most deaths (87%) occurred before 24 hours. At a scheduled interim data review in January 2011, with data available from 2,995 children, a data and safety monitoring committee recommended stopping enrollment because it was unlikely the bolus arms would prove to be superior.

“We could not identify any subgroup in which fluid resuscitation was beneficial,” Maitland et. al write. “This is remarkable given that many of the baseline characteristics of the children in this study are considered to be important criteria for bolus-fluid therapy.”

In his editorial, John Myburgh says he expects that these results will have an immediate impact. This trial is an “important contribution to the literature,” he writes, and the current practice of bolus-fluid resuscitation among these patients in resource-poor settings must stop. “Given that 2 million children die from [febrile illness due to medical causes and impaired perfusion or compensated shock] each year in sub-Saharan Africa, the potential impact is enormous.”

However, the lessons for developed countries are less clear. “The children enrolled in the FEAST trial were very ill, often with very low hemoglobin levels.  Many were chronically malnourished. Whether a similar study in children in a developed country would have the same results remains unknown,” says NEJM Deputy Editor Dr. Julie Ingelfinger. “The results of this study certainly will lead clinicians in resource-rich areas to use greater caution and heightened patient monitoring when bolus fluid resuscitation with either crystalloid or colloids is used in patients with compensated shock who have no clinical fluid deficit.”

Both Ingelfinger and Myburgh say that more research is needed before the lessons from FEAST are extended to other conditions. The children in this trial had shock with associated hypotension, said Myburgh, and extrapolating these results to children with other hypotensive conditions, such as severe dehydration from gastroenteritis and malnutrition, burns, and surgery, wouldn’t be justified.

Carrying out a study such as FEAST, considering the circumstances in the centers where it was performed and the organization and dedication it required from investigators, is notable, said Ingelfinger. “The investigators deserve praise for this work.”

Were you surprised by the results of the FEAST trial? Are they applicable to your place of practice?