Intermittent hypoxia and caffeine in infants born preterm: the ICAF Randomized Clinical Trial

Clinical Snapshot

Key Findings

• P Value: p<0.001 at 41 weeks, p<0.01 at 34 weeks

• Effect Size: At 41 weeks: caffeine 26.6 (18.5, 38.2) vs placebo 73.0 (51.3, 103.7) seconds/hour

• Primary Outcome: Seconds/hour of oxygen saturation <90% significantly reduced with caffeine at all postmenstrual ages

• Nnt Or Sensitivity: Approximately 64% reduction in intermittent hypoxia burden at 41 weeks

• Confidence Interval: 95% CI provided for median values with interquartile ranges

Clinical Application

Highly feasible - caffeine is readily available, well-tolerated, and routinely used in neonatal intensive care Directly applicable to Australian NICUs; caffeine is PBS-listed for apnoea of prematurity and widely used in tertiary neonatal centres Very preterm infants <31 weeks gestation in stable condition on room air between 32-36 weeks postmenstrual age

Abstract

OBJECTIVE: To determine whether extending caffeine therapy through 43 weeks' postmenstrual age (PMA) decreases intermittent hypoxia (IH) in convalescing preterm infants. Secondary objectives were to assess caffeine effects on changes in inflammation-related plasma biomarkers and brain MRI. DESIGN: Multicentre masked randomised trial. SETTING: 16 US hospitals. PATIENTS: Infants at <30 weeks + 6 days gestational age on caffeine between 32 weeks and 36+5 days PMA in room air with routine caffeine discontinuation prior to 36 weeks +6 days. INTERVENTION: Randomisation to caffeine or placebo and treated through 42 completed weeks. Pulse oximetry was recorded from enrolment through 1 week after stopping study drug. Blood for 12 inflammation-related biomarkers obtained at enrolment and 38 weeks' PMA and brain imaging after enrolment or <3 days of randomisation, and study end. MAIN OUTCOME MEASURE: Seconds/hour of oxygen saturation <90% from randomisation to study end. RESULTS: Randomised 160 subjects, 78 placebo, 82 caffeine. IH was less at every PMA with caffeine treatment from 34 (172.7 (123.4, 241.7); 84.7 (64.4, 111.4, p<0.01) through 41 weeks (73.0 (51.3, 103.7); 26.6 (18.5, 38.2, p<0.001). Adjusted TNF-α levels were 23% lower at follow-up in the caffeine group compared with placebo (p<0.02), without other biomarker differences. Paired brain imaging found no significant differences. CONCLUSIONS: Extended caffeine reduced the burden of IH in very preterm infants and may reduce inflammation. Further study is needed to determine if this effect of caffeine is associated with reduced risk of adverse outcomes. TRIAL REGISTRATION NUMBER: NCT03321734.

References

1. 1.Eichenwald, E., Corwin, M., McEntire, B., Knoblach, S., Limperopoulos, C., Kapse, K., Kerr, S., Heeren, T. C., Ikponmwonba, C., & Hunt, C. E. (2026). Intermittent hypoxia and caffeine in infants born preterm: the ICAF Randomized Clinical Trial. Archives of Disease in Childhood. Fetal and Neonatal Edition. https://doi.org/10.1542/peds.2015-3757