Pancreatic enzymes and bile acids alone and in combination for acutely ill, severely malnourished children: a multi-country double-blind randomised placebo-controlled trial.

Background: Translocation of bacteria across the intestinal barrier has been postulated to contribute to mortality among severely malnourished children. Pancreatic enzymes (PE) and bile acids (BA) have anti-bacterial properties in the small intestine, but severe malnutrition is associated with impaired exocrine pancreatic and hepatobiliary functions. We evaluated whether ancillary treatment with PE and BA improves survival in hospitalised, acutely ill, severely malnourished children. Methods: We conducted a multicentre, adaptive, 2 × 2 factorial, randomised, double-blinded controlled trial of PE (3000 IU lipase/kg, twice per day) and BA (ursodeoxycholic acid, 10 mg/kg twice per day) administered orally for 21 days in hospitalised children, age 2-59 months, admitted to hospital with an acute, non-traumatic illness and severe malnutrition. The trial was carried out in six sites in Bangladesh, Kenya, Malawi and Uganda with recruitment occurring between June 20, 2021, and October 17, 2022. The primary outcome was mortality within 60 days of enrolment using an intention-to-treat analysis. Secondary outcomes included the incidence of Serious Adverse Events (SAEs); grade 3 or 4 toxicity events before day 21; number of days with diarrhoea; use of second- or third-line antibiotics during the admission; duration of initial hospitalisation; and changes in anthropometry from enrolment to follow-up at 21 and 60 days after start of the intervention. This trial is registered with www.clinicaltrials.gov, number NCT04542473. Findings: We enrolled 429 children (median age 10.4 months (interquartile range (IQR) 4.8-18.0)). Participants were randomised to PE (n = 213, 50%) or PE-placebo (n = 216, 50%) and to BA (n = 212, 49%) or BA-placebo (n = 217, 51%). Mortality by day 60 did not differ between children assigned to PE compared to PE-placebo (35 children (16%) vs 36 (17%)), when pooling across allocation to BA or BA-placebo (Hazard ratio (HR) 1.02 (95% CI: 0.64-1.62), P = 0.94). Also, 39 children (18%) assigned to BA died compared to 32 (15%) with BA-placebo, when pooling across allocation to PE or PE-placebo (HR 1.27 (95% CI: 0.79-2.02), P = 0.32). There was no evidence of interaction between the PE and BA interventions (HR 1.16, 95% CI: 0.46-2.96; p-value = 0.75). The trial was stopped early based on futility. There were no significant differences in secondary outcomes, other than children assigned to BA spent fewer days on second-line antibiotics than those receiving BA-placebo. Interpretation: Pancreatic enzymes and/or ursodeoxycholic acid do not reduce mortality, serious adverse events, or improve other clinical outcomes in hospitalised, acutely ill, severely malnourished children. Future trials should consider testing a package of interventions that include multiple domains to reduce mortality in this vulnerable population. Funding: The Gates Foundation, Wellcome.