Preterm Heart Guidelines Reveal Deep Divides in Neonatal Intensive Care

In the fragile first days of life, a small blood vessel that once bypassed the lungs in the womb is supposed to seal itself shut. In many premature infants, it does not. The resulting condition, patent ductus arteriosus, or PDA, sits at the center of one of the most contested debates in modern neonatology: when does a persistent opening in the heart become dangerous enough to treat, and how aggressively should clinicians intervene? A new survey of leading neonatal intensive care units across North America has now laid bare just how differently the most experienced centers answer those questions, offering an unusually candid portrait of a field still searching for consensus.

The study, published in the Journal of Perinatology by a team led by Sharada H. Gowda of Indiana University and Riley Children’s Hospital, gathered written PDA management guidelines from hospitals belonging to the Children’s Hospitals Neonatal Consortium, a network of 44 quaternary-level neonatal intensive care units in the United States and Canada. Between November 2023 and February 2024, the researchers emailed designated representatives at every member center, asking whether their institution maintained a formal written guideline for managing the condition in premature infants and requesting a copy if one existed. Thirty-eight of the 44 centers responded, an 86 percent response rate, and 25 of those, roughly two-thirds, provided written protocols for analysis.

What the researchers found was a field united by a broad philosophy but divided on nearly every practical detail. Most guidelines described a stepwise approach that begins with expectant, or watchful, management, escalates to drug therapy when the ductus becomes hemodynamically significant, and reserves procedural closure for infants who fail medical treatment. Yet beneath that shared skeleton, the specifics varied dramatically. Screening strategies, diagnostic thresholds, drug choices, dosing schedules, feeding policies, and the timing of interventions all differed from center to center, sometimes in striking ways.

The diagnosis itself is the first fault line. Hemodynamically significant PDA, abbreviated hsPDA, refers to a ductal shunt substantial enough to disturb the infant’s circulation, potentially stealing blood from the gut and lungs and contributing to serious complications such as bronchopulmonary dysplasia, necrotizing enterocolitis, intraventricular hemorrhage, and prolonged intensive care stays. Because clinical signs such as a heart murmur, bounding pulses, or worsening respiratory status are notoriously nonspecific, objective echocardiographic assessment has become the cornerstone of diagnosis. In the survey, 88 percent of guidelines specified echocardiographic criteria, most commonly ductal diameter, the left atrium-to-aorta ratio, shortened isovolumic relaxation time, and flow reversal in the descending aorta. Less frequently cited markers included flow reversal in the celiac trunk or middle cerebral artery, parameters that reflect more severe systemic hypoperfusion.

Clinical criteria appeared in 80 percent of the guidelines, and respiratory support requirements dominated that category. Among centers that referenced breathing support, 42 percent cited escalating respiratory needs, 37 percent required at least noninvasive ventilation or nasal continuous positive airway pressure, and a small number demanded intubation or used a quantitative respiratory severity score. Other clinical triggers included pulmonary edema, cardiomegaly, pulmonary hemorrhage, widened pulse pressure, and a hyperdynamic precordium, the pounding chest sensation produced by a large shunt. Only 12 percent of centers, however, employed a formal clinical scoring system, a strikingly low figure given growing evidence that composite scores may predict outcomes better than any single parameter.

Screening practices revealed similar heterogeneity. Ninety-two percent of guidelines described risk-based or symptom-based screening for the condition, but the timing of the first echocardiogram ranged from within 36 hours of birth for the most immature infants at some centers to delayed evaluation in the second week of life at others. Some centers screened purely on gestational age, some purely on symptoms, and about a third used a combination of the two. This scatter mirrors the unsettled evidence base: randomized trials and meta-analyses have repeatedly failed to show a clear benefit of routine early treatment compared with expectant management, leaving each institution to chart its own course through the ambiguity.

When treatment is warranted, ibuprofen has emerged as the workhorse drug, recommended as first-line pharmacotherapy by more centers than any alternative. Indomethacin, the traditional cyclooxygenase inhibitor that dominated PDA therapy for decades, has been displaced, while intravenous acetaminophen, increasingly popular at the bedside, remains underrepresented in formal protocols, a discrepancy the authors interpret as evidence that written guidelines lag behind evolving clinician preferences. A quarter of centers recommended prophylactic therapy for the highest-risk infants, such as those born before 25 to 26 weeks without antenatal steroid exposure, with some using indomethacin and others administering acetaminophen at 15 milligrams per kilogram every six hours for twelve doses. The number of treatment courses allowed also varied widely, from a single course at one center to as many as four at others.

Feeding practices during therapy, long a source of anxiety because of theoretical concerns about gut injury, showed a reassuring convergence. Not a single guideline recommended holding feeds entirely during pharmacotherapy, and most centers either allowed full feeding, restricted volume modestly, or permitted trophic feeds. This aligns with accumulating evidence that withholding feeds prolongs the time needed to reach full nutrition without reducing gastrointestinal complications, and that feed interruption may even explain some of the poor outcomes previously attributed to the drugs themselves. On the expectant side, centers that described conservative management uniformly recommended fluid restriction of 120 to 140 milliliters per kilogram per day, with most also endorsing higher breathing pressures, diuretics, and higher hemoglobin targets, while explicitly encouraging rather than restricting nutrition.

Perhaps the most consequential trend captured by the survey is the rise of transcatheter closure. Ninety-two percent of guidelines addressed device-based closure, and 83 percent of those recommended a course of pharmacotherapy before resorting to the catheter procedure. Minimum weight thresholds for device closure ranged from 650 to 1,800 grams, and several centers specified waiting until 14 or 21 days of life. Surgical ligation, once the default definitive treatment, appeared in only 48 percent of guidelines and was frequently reserved for infants too small for the catheter approach. The shift reflects the maturation of minimally invasive occluder devices and a growing recognition that surgical ligation carries risks of its own, including vocal cord paralysis and the controversial post-ligation syndrome, for which, remarkably, only one center in the entire consortium offered written guidance.

The authors are careful to note the study’s limits: written guidelines may not match what actually happens at the bedside, the survey captured no patient outcomes, and the findings come exclusively from large referral centers that may not represent community hospitals. The timing also matters, since the data collection preceded several landmark trials, including BeNeDuctus, BabyOSCAR, and the Neonatal Research Network PDA Randomized Clinical Trial, each of which concluded that conservative management performs as well as or better than drug treatment, though all were constrained by early termination, treatment crossover, and the exclusion of the sickest infants. Still, the picture that emerges is of a specialty in transition, holding onto a common stepwise framework while awaiting the evidence needed to settle its deepest disagreements. For the smallest patients, whose lives often hinge on decisions made within hours of birth, the path from a leaking vessel in the heart to a healed one still depends, to a remarkable degree, on which hospital door they enter first.

Subject of Research: Variation in diagnosis and management of hemodynamically significant patent ductus arteriosus across neonatal intensive care units

Article Title: Variations in diagnosis and management of hemodynamically significant patent ductus arteriosus across neonatal intensive care units in the Children’s Hospitals Neonatal Consortium

Article References: Gowda, S. H., Levy, P. T., Hamrick, S., Keene, S., Sen, S., Rajgarhia, A., Rao, R., Ball, M. K., Weems, M. F., & on behalf of CHNC Cardiac Focus Group (2026). Variations in diagnosis and management of hemodynamically significant patent ductus arteriosus across neonatal intensive care units in the Children’s Hospitals Neonatal Consortium. Journal of Perinatology. https://doi.org/10.1038/s41372-026-02910-8

Image Credits: AI Generated

DOI: 10.1038/s41372-026-02910-8

Keywords: patent ductus arteriosus, preterm infants, neonatal intensive care, echocardiography, ibuprofen, acetaminophen, transcatheter closure, surgical ligation, clinical guidelines, Children's Hospitals Neonatal Consortium, hsPDA, neonatology