Milrinone exposure in extremely preterm infants carried a measurable hypotension signal
Milrinone is being used more often in preterm neonates for acute pulmonary hypertension and to blunt hemodynamic instability after patent ductus arteriosus closure, but its safety profile in the smallest infants remains thin. In a retrospective cohort from a level IV neonatal intensive care unit at the University of Iowa, 34 extremely low gestational age infants exposed to milrinone in the first four postnatal weeks were examined for early hemodynamic effects. The cohort was small, but the signal was not subtle: 8 infants, or 23.5%, developed hypotension or required escalation of vasoactive therapy within 6 hours of infusion start.
In bedside terms, the drug was given into a physiology already under strain. These infants were born at a mean gestational age of 24.5 weeks and had a mean birth weight of 676 grams, a population in which vascular tone, myocardial reserve, and pulmonary adaptation are all fragile. A phosphodiesterase-3 inhibitor such as milrinone can improve contractility and relax vascular smooth muscle, but the same mechanism can lower systemic pressure when cardiac output cannot compensate. In the transitional circulation of an extremely preterm newborn, even a modest fall in vascular resistance can expose limited reserve quickly.
Smaller, younger infants were the ones who destabilized
The infants who became hypotensive were younger and smaller at the time milrinone began. Their postmenstrual age was 27.3 weeks versus 28.8 weeks in those who remained stable, and their weight was 796 grams versus 1104 grams. Respiratory status also separated the groups early: the hypotensive infants had a higher Respiratory Severity Score during the first four hours after milrinone initiation, 7.6 compared with 4.4 in the non-hypotensive group.
Those patterns matter because they point to a phenotype rather than a random adverse event. Lower weight and younger postmenstrual age usually track with less mature autonomic control, weaker myocardial performance, and greater susceptibility to vasodilation. A higher respiratory severity score suggests greater lung disease burden, which often sits alongside higher pulmonary vascular resistance and a more precarious cardiopulmonary balance. In that setting, milrinone may improve one side of the circulation while tipping the other into instability.
The study’s primary outcome captured either sustained low blood pressure for at least an hour within six hours of starting the infusion, or the need to begin or increase vasoactive medication. Hypotension was defined rigorously as diastolic or both diastolic and mean arterial pressure below the third percentile for gestational age at exposure, or hemodynamic deterioration requiring pressor support. That endpoint keeps the focus on what clinicians actually confront at the bedside: the decision to support blood pressure, not simply a numeric dip on the monitor.
A drug chosen for physiology can still destabilize it
Milrinone has a logical appeal in neonatal intensive care. By increasing cyclic adenosine monophosphate signaling, it can enhance myocardial performance and promote vasodilation, which is attractive when pulmonary hypertension, low systemic blood flow, or post-closure instability are part of the picture. Animal data cited in the article suggest that the immature myocardium may respond differently in the early transitional period, and that concern aligns with the human data here. When the myocardium and vasculature are still adapting after birth, a drug that depends on preserved reserve may behave less predictably.
The investigators did not show that milrinone alone caused every hypotensive episode. The sickest infants are the most likely to receive the drug, and those infants often carry multiple simultaneous drivers of instability, including respiratory failure and shifting shunt physiology. Even so, the timing of the events is hard to ignore. A hypotensive response appearing soon after infusion start raises the possibility that the medication itself, the underlying disease state, or both contributed to the collapse in perfusion pressure.
One sentence deserves emphasis on its own: 23.5% of exposed infants developed hypotension or needed more vasoactive support.
What this retrospective cohort can and cannot answer
The study was conducted between 2018 and 2020 at a single center and included only infants born at 22 0/7 to 28 6/7 weeks’ gestation who received milrinone during the first four postnatal weeks. Infants with systemic hypotension at the time of initiation, known or suspected genetic syndromes, or major congenital malformations were excluded. That design strengthened internal consistency, but it also narrowed the population to a highly selected group of very preterm neonates.
A small sample size limits what can be claimed about causality, dose-response relationships, and the contribution of polypharmacy. The analysis could not statistically disentangle background vasoactive exposure from the effects of milrinone itself. It also cannot tell clinicians which exact infusion strategy, timing, or co-treatment pattern would minimize risk. Those unanswered points are not minor; they are the practical variables that determine whether a drug helps or destabilizes an infant in real time.
Still, the direction of the findings is clinically useful. The infants most likely to destabilize were those with the least physiologic reserve: smaller size, younger postmenstrual age, and higher respiratory severity. That cluster supports a cautious, physiology-based approach rather than reflexive expansion of use. In a population this fragile, hemodynamic monitoring needs to begin before the infusion starts and continue through the first hours of exposure, when pressure loss appears most likely to emerge.
Bedside implications for neonatal hemodynamic support
For neonatologists, the message is not to abandon milrinone, but to respect its pharmacology. In a preterm infant with pulmonary hypertension or post-ductus instability, the drug may still be appropriate when the competing problem is excessive afterload or poor ventricular performance. Yet this report argues against treating it as a benign adjunct. Blood pressure, vasoactive requirements, respiratory burden, and gestational maturity should all shape the threshold for starting the infusion and the intensity of monitoring afterward.
The broader implication is straightforward: the smallest infants may pay the highest hemodynamic price for a therapy intended to stabilize them. Further studies are needed to define safer dosing, better selection criteria, and the interaction between milrinone and other vasoactive agents in extremely preterm newborns, especially those with the narrowest physiologic margin.