Keeping the Caffeine Going: Can a Familiar Medicine Steady a Premature Baby's Breathing Longer?

What a new clinical trial called ICAF found when premature babies stayed on caffeine until around their due date

A new study called the ICAF trial tested whether keeping very premature babies on caffeine — a medicine used in nearly every newborn intensive care unit to steady fragile breathing — for several weeks longer than usual would reduce the brief drops in their blood-oxygen levels. The answer was clear: babies who stayed on caffeine until about their due date spent roughly 60% less time with low oxygen than babies given a placebo, and the medicine appeared just as safe as it has always been [1]. What the study could not yet answer is the question parents most want answered — whether fewer of these oxygen dips means healthier development later on.

Why This Question Matters

Babies born many weeks early often have a breathing-control system that is not fully wired up. The part of the brain that reminds the body to keep breathing steadily is still maturing, so these infants pause their breathing or breathe shallowly far more often than full-term babies. Many of these episodes are brief and silent — the baby's oxygen level dips for a few seconds and then recovers, often without anyone noticing and without an alarm sounding. Doctors call these repeated dips "intermittent hypoxia." Caffeine is the standard medicine that helps prevent them, and it has been one of the most important advances in newborn care over the past several decades.

How We Got Here

Caffeine became a cornerstone of premature-baby care because of a landmark trial known as CAP. That study gave caffeine or a placebo to more than 2,000 premature babies and found that caffeine not only reduced pauses in breathing but, remarkably, improved the chances of surviving without a lasting disability [2]. When researchers checked back on those same children years later, the benefit held up — they did better at age 5 [3] and still showed differences in movement and school performance at age 11 [4]. Findings like these turned caffeine into a medicine given to almost every very premature baby.

But there was a loose end. Studies that recorded babies' oxygen levels continuously showed that their breathing does not fully settle into the steady, term-baby pattern until around 42 to 43 weeks — roughly two to three weeks past a typical due date [5]. And there were hints that these lingering oxygen dips might matter: in one large analysis, the premature babies who had the most low-oxygen episodes in their early months were somewhat more likely to have developmental problems later [6]. Here was the puzzle. Doctors usually stop caffeine once a baby seems to have outgrown obvious breathing pauses, often at around 34 weeks, following standard pediatric guidance [7]. That means the medicine is typically withdrawn weeks before the breathing system has actually finished maturing — leaving the quiet, hidden oxygen dips untreated during exactly the period they remain most common.

The ICAF researchers had already taken first steps toward this question. In an earlier study, ordinary doses of caffeine reduced the oxygen dips for a few weeks but the effect faded [8]; in a second study, a higher dose pushed the benefit a little further, to about 38 weeks [9]. ICAF was designed to find out whether caffeine, continued all the way to around the due date, could keep those dips suppressed through the whole vulnerable window.

What the Researchers Did

The trial enrolled babies born before about 30 weeks of pregnancy who were already on caffeine, breathing room air, and doing well — 160 babies in all, treated at 16 hospitals across the United States [1]. When the time came to stop their usual caffeine, each baby was randomly assigned, by chance and without the family or bedside team knowing which, to either keep receiving caffeine or to receive a look-alike placebo. The babies stayed on their assigned treatment until nearly three weeks past a full-term due date. Throughout, a special oxygen monitor recorded their oxygen levels continuously — even after they went home — so the researchers could measure exactly how much time each baby spent with low oxygen. Saliva tests confirmed that the caffeine babies really were getting a proper dose, including at home.

What They Found

Week after week, the babies on caffeine spent far less time with low oxygen. From about 34 weeks through 41 weeks, the caffeine group's time with oxygen below a key threshold was around 60% lower than the placebo group's, and the difference was large enough to be statistically convincing despite the study being smaller than planned [1]. The trial had to stop enrolling earlier than intended because its research funding ran out, which made it harder to detect smaller effects — yet the benefit on oxygen levels was strong enough to show through clearly.

The researchers also looked for clues about why this might matter. They measured markers of inflammation in the blood and found that one of them, called TNF-α, dropped more in the caffeine babies [1] — a hint that fewer oxygen dips might mean less inflammation, which other research links to brain injury in premature infants [10], [11]. Brain MRI scans, however, showed no differences between the two groups, though only a small number of babies had usable scans. There were also some unexpected practical observations: fewer caffeine babies needed to be put back on extra oxygen, and they tended to go home sooner, although they gained weight a little more slowly. Importantly, the caffeine appeared just as safe as always — no meaningful increase in fast heart rate, reflux, or other concerning events.

One detail worth understanding is why the study measured oxygen dips rather than something more obviously meaningful, like how the children turned out years later. Long-term studies take many years and large numbers of babies to complete. Measuring the oxygen dips first is a faster, practical way to find out whether the medicine does what it is supposed to do at all — a necessary stepping stone before committing the time and resources to a much bigger study of development. ICAF cleared that first hurdle convincingly: caffeine plainly reduced the dips. That is what makes the larger, longer study worth doing now. It is also worth knowing that the babies wore a research-grade oxygen monitor that recorded continuously, even at home, which is far more detailed than the routine monitoring most babies receive — so the study could see dips that ordinary bedside checks would miss entirely.

What This Means for Families

If your baby is on caffeine in the NICU, this study is reassuring on two counts: caffeine continues to look very safe, and keeping it going longer can meaningfully reduce the hidden oxygen dips that are so common as premature babies mature. For a baby who is having a lot of these dips near the time of going home, a care team might reasonably discuss continuing caffeine for a few extra weeks as a low-risk way to help steady breathing.

It also helps to understand what kind of medicine caffeine is in this setting. It is not a sedative or a strong drug with a long list of risks; at the doses used in newborn units it has decades of safety experience behind it, and ICAF adds to that reassurance by finding no meaningful increase in side effects even when the medicine was continued for extra weeks. For many families, knowing that the option being discussed is a familiar, well-tolerated medicine — rather than something experimental or risky — makes the conversation with the care team easier.

It is just as important to be honest about what the study did not show. Spending less time with low oxygen is encouraging, but the researchers could not yet prove that it leads to better thinking, movement, or learning down the road. That is the real goal, and it remains an open question — one that scientists who study caffeine describe as the central uncertainty in this field [12]. A separate recent trial in slightly more mature premature babies found that extra caffeine helped their breathing but did not actually get them home faster, because going home depended more on learning to feed than on breathing [13]. So families should see ICAF as a promising and reassuring step, not a finished answer. The slower weight gain seen with caffeine is expected and temporary, with babies catching up afterward.

What Comes Next

The natural next step, and the one the researchers themselves call for, is a larger trial that follows babies for years to see whether reducing these oxygen dips actually improves how they grow, think, and move. Until that study is done, extended caffeine sits in a hopeful but unproven place: a familiar, evidently safe medicine that clearly does what it is meant to do for breathing, with the most important benefit — a healthier future for the child — still waiting to be confirmed. For parents, the most useful takeaway is that this is an active, promising area of research, and that the questions being asked are exactly the right ones.

References

  1. Eichenwald E, Corwin M, McEntire B, et al. Intermittent hypoxia and caffeine in infants born preterm: the ICAF Randomized Clinical Trial. Arch Dis Child Fetal Neonatal Ed. 2025; Epub ahead of print. doi:10.1136/archdischild-2025-329230 ↩
  2. Schmidt B, Roberts RS, Davis P, et al. Caffeine therapy for apnea of prematurity. N Engl J Med. 2006;354(20):2112–2121. doi:10.1056/NEJMoa054065 ↩
  3. Schmidt B, Anderson PJ, Doyle LW, et al. Survival without disability to age 5 years after neonatal caffeine therapy for apnea of prematurity. JAMA. 2012;307(3):275–282. doi:10.1001/jama.2011.2024 ↩
  4. Schmidt B, Roberts RS, Anderson PJ, et al. Academic performance, motor function, and behavior 11 years after neonatal caffeine citrate therapy for apnea of prematurity. JAMA Pediatr. 2017;171(6):564–572. doi:10.1001/jamapediatrics.2017.0238 ↩
  5. Hunt CE, Corwin MJ, Weese-Mayer DE, et al. Longitudinal assessment of hemoglobin oxygen saturation in preterm and term infants in the first six months of life. J Pediatr. 2011;159(3):377–383. doi:10.1016/j.jpeds.2011.02.011 ↩
  6. Poets CF, Roberts RS, Schmidt B, et al. Association between intermittent hypoxemia or bradycardia and late death or disability in extremely preterm infants. JAMA. 2015;314(6):595–603. doi:10.1001/jama.2015.8841 ↩
  7. Eichenwald EC; AAP Committee on Fetus and Newborn. Apnea of prematurity. Pediatrics. 2016;137(1):e20153757. doi:10.1542/peds.2015-3757 ↩
  8. Rhein LM, Dobson NR, Darnall RA, et al. Effects of caffeine on intermittent hypoxia in infants born prematurely: a randomized clinical trial. JAMA Pediatr. 2014;168(3):250–257. doi:10.1001/jamapediatrics.2013.4371 ↩
  9. Dobson NR, Rhein LM, Darnall RA, et al. Caffeine decreases intermittent hypoxia in preterm infants nearing term-equivalent age. J Perinatol. 2017;37(10):1135–1140. doi:10.1038/jp.2017.82 ↩
  10. Abu Jawdeh EG, Huang H, Westgate PM, et al. Intermittent hypoxemia in preterm infants: a potential proinflammatory process. Am J Perinatol. 2021;38(13):1313–1319. doi:10.1055/s-0040-1712951 ↩
  11. Darnall RA, Chen X, Nemani KV, et al. Early postnatal exposure to intermittent hypoxia in rodents is proinflammatory, impairs white matter integrity, and alters brain metabolism. Pediatr Res. 2017;82(1):164–172. doi:10.1038/pr.2017.102 ↩
  12. Oliphant EA, Hanning SM, McKinlay CJD, et al. Caffeine for apnea and prevention of neurodevelopmental impairment in preterm infants: systematic review and meta-analysis. J Perinatol. 2024;44(6):785–801. doi:10.1038/s41372-024-01939-x ↩
  13. Carlo WA, Eichenwald EC, Carper BA, et al. Extended caffeine for apnea in moderately preterm infants: the MoCHA Randomized Clinical Trial. JAMA. 2025;333(24):2154–2163. doi:10.1001/jama.2025.5791 ↩