Importance: Children born very preterm are at risk for long-term neurodevelopmental sequelae. Prophylactic high-dose recombinant human erythropoietin (rhEpo) shortly after birth has not been shown to improve cognitive, motor, and behavioral development at 2 and 5 years.

Objective: To investigate whether early high-dose rhEpo is associated with better executive functions and processing speed-late-maturing cognitive functions-in school-aged children born very preterm.

Design, setting, and participants: This single-center cohort study was a prospective, observational follow-up study of a multicenter neonatal clinical trial; 365 children born very preterm (mean gestational age, 29.3 weeks [range, 26.0-31.9 weeks]) who had been enrolled in the Swiss EPO Neuroprotection Trial at birth between 2005 and 2012, and who were included in the primary outcome analyses at 2 years, were eligible to be recruited for the EpoKids study between 2017 and 2021 when they were at school age. Term-born children were additionally recruited and included in a control group. Data were analyzed between May and September 2022.

Exposure: Administration of rhEpo (3000 IU/kg) or placebo (saline, 0.9%) intravenously 3 times within the first 2 days of life as part of the Swiss EPO Neuroprotection Trial.

Main outcome and measures: A comprehensive neuropsychological test battery assessed executive functions and processing speed, and parents reported on their child's executive functions in everyday life to test the hypothesis that early high-dose rhEpo administration is associated with better cognitive outcomes at school age.

Results: In the EpoKids study, 214 children born very preterm (58.6% of 365 children in eligible cohort) were assessed at a mean age of 10.4 years (range, 6.9-13.4 years); 117 (54.7%) were boys. There was no evidence that the 117 children who had received rhEpo differed from the 97 children who had received placebo in any of the 15 executive function and processing speed tests, nor in parent-rated executive functions (estimates ranged from -0.138 to 0.084, all 95% CIs included 0). Irrespective of rhEpo or placebo allocation, children born very preterm scored lower on 11 of 15 executive function and processing speed tests than term-born peers (estimates ranged from 0.112 to 0.255, 95% CIs did not include 0).

Conclusion and relevance: This study found no evidence for a positive association between prophylactic early high-dose rhEpo administration and long-term neurodevelopmental outcomes after very preterm birth. These results suggest that a comprehensive approach, including pharmacological and nonpharmacological prevention and intervention strategies, is needed to support these children's neurodevelopmental outcome.