Abstract
IMPORTANCE
The management of patent ductus arteriosus (PDA) in preterm infants is controversial.
OBJECTIVE
To determine whether expectant management compared with active treatment of a protocol-defined PDA in preterm infants decreases the incidence of death or bronchopulmonary dysplasia (BPD).
DESIGN, SETTING, AND PARTICIPANTS
A randomized clinical trial including infants born at 22 to 28 weeks’ gestation and diagnosed with a protocol-defined PDA between the age of 48 hours and 21 days at screening. The trial was conducted from December 2018 to December 2024 at 33 hospitals within the National Institute of Child Health and Human Development Neonatal Research Network. The final date of follow-up was June 2025.
INTERVENTIONS
Infants with PDA were randomized to expectant management (n = 242) or active treatment (n = 240; acetaminophen, ibuprofen, or indomethacin) to close the PDA.
MAIN OUTCOMES AND MEASURES
The primary outcome was death or BPD at 36 weeks’ postmenstrual age. The secondary outcomes included the components of the primary outcome and other morbidities of prematurity.
RESULTS
A total of 482 infants were randomized (median gestational age, 25 weeks [IQR, 24 to 27 weeks]; median birth weight, 760 g [IQR, 620 to 935 g]). The trial was stopped for futility and safety after the 50% interim analysis for the primary outcome due to higher survival in the expectant management group. The incidence of death or BPD was 80.9% (195/241) of infants in the expectant management group vs 79.6% (191/240) of infants in the active treatment group (adjusted risk difference, 1.2% [95% CI, −5.7% to 8.1%]; P = .73). The incidence of death before 36 weeks’ postmenstrual age was 4.1% (10/241) of infants in the expectant management group vs 9.6% (23/240) of infants in the active treatment group (adjusted risk difference, −5.6% [95% CI, −10.1% to −1.2%]; P = .01). Infections resulting in death occurred in 0.8% (2/241) of infants in the expectant management group vs 3.8% (9/240) of infants in the active treatment group.
CONCLUSIONS AND RELEVANCE
In extremely preterm infants with a protocol-defined PDA, death or BPD did not differ between the expectant management group and the active treatment group. Survival was substantially higher with expectant management.
TRIAL REGISTRATION
ClinicalTrials.gov Identifier: NCT03456336
Graphical Abstract
Patent ductus arteriosus (PDA) is commonly diagnosed in preterm infants. During fetal life, the ductus arteriosus connects the pulmonary artery to the aorta, allowing right ventricular blood to bypass the lungs and enter systemic circulation. In most full-term infants (born between 39 weeks 0 days and 40 weeks 6 days’ gestation), the ductus arteriosus closes soon after birth.1 However, in preterm infants, closure is often delayed2 and may not occur until after hospital discharge.3 In extremely preterm infants, PDA is associated with several complications of prematurity, including bronchopulmonary dysplasia (BPD).4
A Cochrane Review5 evaluating active treatment vs expectant management before 7 postnatal days reported that early treatment of PDA “probably results in little to no difference in mortality (moderate-certainty evidence).” However, a meta-analysis,6 which was limited to trials of active treatment begun in the first 2 postnatal weeks (and were published between 2010 and 2024), reported that active treatment was associated with increased mortality; no individual trial demonstrated a statistically significant difference.
Despite these recent studies, there continues to be wide variation in clinical practice.7,8 Approximately one-quarter of infants born at 22 to 28 weeks’ gestation in the US receive pharmacological treatment for PDA closure.9 Although pharmacological and surgical interventions are generally effective in closing the PDA (70%−80% closure after treatment with acetaminophen, ibuprofen, or indomethacin), these interventions have not been shown to reduce mortality or key morbidities of prematurity, including BPD.10
Prior to publication of the recent meta-analysis,6 we initiated a randomized clinical trial including preterm infants with a protocol-defined PDA to evaluate the risks and benefits of expectant management vs active treatment (pharmacological treatment with acetaminophen, ibuprofen, or indomethacin). We hypothesized that expectant management would reduce the incidence of death or BPD at 36 weeks’ postmenstrual age compared with active treatment.
Methods
Trial Design
This registry-embedded, comparative-effectiveness, randomized clinical trial was conducted at hospitals participating in the Eunice Kennedy Shriver National Institute of Child Health and Human Development Neonatal Research Network (NRN). All NRN hospitals contributed prospectively reported data to the NRN generic database, which is a registry of infants born before 29 weeks’ gestation. Infants were screened and enrolled from December 2018 to December 2024 (Figure 1) following guidelines defined in the trial protocol (Supplement 1). The final date of hospital follow-up was June 2025. There is ongoing neurodevelopmental follow-up of the infants to 2 years. The outcomes, as well as medical and surgical PDA interventions, were recorded through 120 days’ postnatal age or until hospital discharge or transfer (if either the discharge or transfer occurred earlier).
Figure 1. Flow Diagram of Preterm Infants With Patent Ductus Arteriosus (PDA).
aMay have met more than 1 criterion.
bRequired steroid treatment or had kidney insufficiency, low platelet count, active bleeding, or necrotizing enterocolitis.
cCommonly specified reasons include infant already treated, physician preference to treat, language or communication barrier with parent, parent not willing or interested in participating in any research study, and infections or other clinical conditions.
dOne infant was later found to be ineligible due to cardiopulmonary compromise at time of screening. This infant is included in all analyses.
eStratified by Neonatal Research Network center and gestational age strata (22–25 weeks’ and 26–28 weeks’ gestation).
fInfants in the expectant management group received treatment only after cardiopulmonary compromise (received acetaminophen, ibuprofen, or indomethacin; underwent cardiac catheterization or surgical ligation; or continued expected management). Infants in the active treatment group received pharmacological treatment within 48 hours of diagnosis (received acetaminophen, ibuprofen, or indomethacin).
gThe data for this infant were excluded from all analyses.
hDid not receive treatment. The data for this infant were included in all analyses.
iIncludes all randomized infants with parental consent. In the expectant management group, 8 infants were still in the hospital at 120 days; 1 infant died in the hospital and 5 infants were alive and still in the hospital at 1 year; and 2 infants had not reached 1-year follow-up. In the active treatment group, 10 infants were still in the hospital at 120 days; 6 infants were alive and still in the hospital at 1 year; 2 infants had been discharged or transferred to another facility; and 2 infants had not reached 1-year follow-up.
The trial was approved by the institutional review board at each participating center. Written informed consent was obtained from a parent or guardian before enrollment. The trial was granted an exemption to investigational new drug regulations by the US Food and Drug Administration. The trial protocol appears in Supplement 1 and the statistical analysis plan appears in Supplement 2. The clinical investigators were masked to the cumulative outcome data during the trial. The Consolidated Standards of Reporting Trials (CONSORT) reporting guideline was used for this study.
Patients
Eligible infants (1) were born from 22 weeks 0 days to 28 weeks 6 days’ gestation, (2) were between the age of 48 hours and 21 days at the time of screening, and (3) had a hemodynamically significant PDA (defined per the trial protocol using echocardiographic and clinical criteria). This postnatal age range is when most neonatologists in the US typically perform an echocardiographic evaluation for a PDA.11 Presence of a PDA was determined using the modified criteria from McNamara and Sehgal.12 Infants were categorized as having (1) no or asymptomatic PDA, (2) symptomatic PDA (met echocardiographic criteria of mild, moderate, or severe with a small- or medium-sized PDA or echocardiographic criteria of mild or moderate with a large-sized PDA), or (3) cardiopulmonary compromise (met echocardiographic criteria of severe with a large-sized PDA). Only the infants with symptomatic PDA were eligible for enrollment (eTable 1 in Supplement 3).
Infants were excluded if they had cardiopulmonary compromise; had a known congenital heart defect other than an atrial or ventricular septal defect; had a pulmonary malformation (eg, congenital lobar emphysema or congenital pulmonary airway malformation); had previously received pharmacological or procedural treatment for PDA; or had any condition (in the opinion of the investigator) that would preclude enrollment (eTable 2 in Supplement 3) such as receiving corticosteroid treatment or having kidney insufficiency, thrombocytopenia, active bleeding, or necrotizing enterocolitis/spontaneous intestinal perforation (each of these conditions is a relative contraindication to treatment with PDA therapies). The rationale for excluding infants with cardiopulmonary compromise was that most neonatologists involved in the conduct of this trial did not have equipoise to assign these infants to expectant management. To increase generalizability, infants who had received prophylactic indomethacin for the indication of interventricular hemorrhage prevention were eligible. Nine of the NRN centers used prophylactic indomethacin during the trial period.
Randomization
The participants were randomized 1:1 by the NRN data coordinating center at RTI International to expectant management or active treatment, stratified by NRN center and gestational age strata (22–25 weeks’ and 26–28 weeks’ gestation) using permuted block sizes of 4 or 6. Siblings from multiple births were randomized separately. The participants were required to be randomized within 48 hours of meeting eligibility criteria (Table 1). The postnatal day of echocardiographic screening for PDA varied by site and was not standardized. At sites where echocardiography was not routinely performed, consent would be obtained before the echocardiographic screening was conducted.
Table 1. Baseline Maternal and Infant Characteristics.
| Expectant management (n = 241)a | Active treatment (n = 240)b | |
|---|---|---|
| Maternal characteristics | ||
| Age, median (IQR), y | 29 (25–33) | 28 (24–32) |
| Race and ethnicity, No. (%)c | ||
| American Indian or Alaska Native | 1 (0.4) | 1 (0.4) |
| Asian | 9 (3.7) | 7 (2.9) |
| Black or African American | 93 (38.6) | 82 (34.2) |
| Hispanic or Latino | 55 (22.9) | 58 (24.4) |
| More than 1 race | 7 (2.9) | 5 (2.1) |
| Native Hawaiian or Other Pacific Islander | 1 (0.4) | 1 (0.4) |
| White | 112 (46.5) | 120 (50.0) |
| Unknown or not reported | 18 (7.5) | 24 (10.0) |
| Maternal education, No. (%) | ||
| <High school | 21 (8.7) | 28 (11.7) |
| High school degree | 61 (25.3) | 60 (25.0) |
| Partial college, trade, or technical | 52 (21.6) | 57 (23.8) |
| College degree or higher | 37 (15.4) | 33 (13.8) |
| Unknown | 70 (29.0) | 62 (25.8) |
| Maternal health insurance, No. (%) | ||
| Private | 141 (58.5) | 126 (52.5) |
| Public | 87 (36.1) | 93 (38.8) |
| Self-pay, uninsured, or other | 9 (3.7) | 18 (7.5) |
| Unknown | 4 (1.7) | 3 (1.3) |
| Maternal medications, No./total (%) | ||
| Antenatal steroids | 214/240 (89.2) | 213/239 (89.1) |
| Magnesium sulfate | 205/234 (87.6) | 191/231 (82.7) |
| Antibiotics within 72 h, No./total (%) | 172/236 (72.9) | 176/233 (75.5) |
| Chorioamnionitis, No./total (%) | ||
| Histological | 81/208 (38.9) | 84/206 (40.8) |
| Clinical | 25/239 (10.5) | 27/234 (11.5) |
|
| ||
| Infant characteristics | ||
| Assigned sex at birth, No. (%) | ||
| Female | 120 (49.8) | 117 (48.8) |
| Male | 121 (50.2) | 123 (51.3) |
| Gestational age | ||
| Median (IQR), wk | 25.6 (24.6–27.0) | 25.6 (24.3–27.3) |
| Group, No. (%) | ||
| 22 wk 0 d to 25 wk 6 d | 137 (56.8) | 136 (56.7) |
| 26 wk 0 d to 28 wk 6 d | 104 (43.2) | 104 (43.3) |
| Postnatal days at randomization, median (IQR) | 10 (7–14) | 10 (7–14) |
| Birth weight, median (IQR), g | 760 (620–940) | 760 (610–923) |
| Small for gestational age, No. (%)d | 19 (7.9) | 20 (8.3) |
| Multiple birth, No. (%) | 65 (27.0) | 56 (23.3) |
| Delivery mode, No. (%) | ||
| Cesarean | 183 (75.9) | 182 (75.8) |
| Vaginal vertex | 52 (21.6) | 46 (19.2) |
| Vaginal breech | 5 (2.1) | 10 (4.2) |
| Unknown | 1 (0.4) | 2 (0.8) |
| Mode of respiratory support at trial entry, No. (%) | ||
| Conventional ventilator | 102 (42.3) | 99 (41.3) |
| NCPAP or NIPPV | 87 (36.1) | 80 (33.3) |
| High-frequency ventilator | 48 (19.9) | 58 (24.2) |
| Apgar score at 5 min, median (IQR) | 7 (5–8) [n = 238] | 7 (4–8) [n = 235] |
| Fraction of inspired oxygen (FIO2) at screening, median (IQR) | 0.28 (0.23–0.33) | 0.28 (0.23–0.35) |
| Size of PDA shunt on echocardiographic screening, No. (%) | ||
| Smalle | 28 (11.6) | 28 (11.7) |
| Moderatef | 72 (29.9) | 74 (30.8) |
| Largeg | 141 (58.5) | 138 (57.5) |
| Clinical PDA criteria, No. (%) | ||
| Mildh | 127 (52.7) | 126 (52.5) |
| Moderatei | 107 (44.4) | 107 (44.6) |
| Severej | 7 (2.9) | 7 (2.9) |
Abbreviations: NCPAP, nasal continuous positive airway pressure; NIPPV, nasal intermittent positive pressure ventilation; PDA, patent ductus arteriosus.
Treatment only after cardiopulmonary compromise (received acetaminophen, ibuprofen, or indomethacin; underwent cardiac catheterization or surgical ligation; or continued expected management).
Pharmacological treatment within 48 hours of diagnosis (received acetaminophen, ibuprofen, or indomethacin).
Selected from provided categories and reported in the maternal electronic health record.
Weight at birth below the 10th percentile as determined by the Alexander growth curve.
Classified as 0.1- to 1.5-mm diameter, ductal velocity greater than 2.5 m/s, and ratio of left atrium to aorta less than 1.5:1.
Classified as having at least 1 of the following: 1.5- to 3-mm diameter, ductal velocity of 1.5 to 2.5 m/s, or ratio of left atrium to aorta ratio of 1.5–2:1.
Classified as having at least 1 of the following: greater than 3-mm diameter, ductal velocity less than 1.5 m/s, ratio of left atrium to aorta greater than 2:1, or diastolic flow reversal in the abdominal aorta.
Defined as FIO2 less than 0.30 and requiring positive airway pressure (NCPAP, NIPPV, mechanical ventilation, high-frequency ventilation, oxygen flow rate of 1 L/min via nasal cannula).
Defined as having hypotension requiring single vasopressor medication or both FIO2 of 0.30 to 0.50 and positive airway pressure.
Defined as having hypotension requiring more than 1 vasopressor medication or both FIO2 greater than 0.50 and requiring mechanical ventilation (conventional or high frequency).
Intervention
In the expectant management group, treatments to close the PDA were not to be administered unless the participant developed cardiopulmonary compromise or reached 36 weeks’ postmenstrual age (eTable 1 in Supplement 3). In the active treatment group, participants received pharmacological treatment (acetaminophen, which was added in a protocol amendment after 14% of infants had been randomized due to increasing use in clinical practice,8 ibuprofen, or indomethacin). The choice of agent, dose, and route (intravenous or enteral) was determined by the clinical team. Initial treatment was to begin within 48 hours of the trial-defined PDA diagnosis. Subsequent echocardiographic surveillance and pharmacological treatment decisions were made at the discretion of the clinical care team. Surgical intervention (PDA ligation or cardiac catheterization with device placement) in the active treatment group was left to the clinical team’s discretion if the use of the pharmacological treatment did not successfully close the PDA.
In both groups, if participants developed cardiopulmonary compromise, the clinical team made decisions to prescribe medications, intervene surgically (PDA ligation or cardiac catheterization with device placement), or continue observation. In both groups, if a participant received additional courses of pharmacological treatment using a different medication (eg, acetaminophen followed by ibuprofen), it was considered a protocol deviation. Participants could be withdrawn at the request of a parent/guardian or a physician.
Outcomes
The primary outcome was death or BPD (assessed using a physiological definition13) at 36 weeks’ postmenstrual age. The secondary outcomes at 36 weeks’ postmenstrual age included components of the primary outcome (death or BPD), moderate or severe BPD (ie, grade II or III) using the definition of Jensen et al,14 necrotizing enterocolitis diagnosed via the modified Bell staging criteria or autopsy, and growth (weight, length, and head circumference z scores). Other secondary outcomes were retinopathy of prematurity (stage ≥3 that required an intervention or treatment) and PDA ligation or cardiac catheterization to close the PDA at hospital discharge, transfer, or 120 postnatal days, whichever occurred sooner.
The prespecified safety outcomes included the following adverse events: necrotizing enterocolitis (Bell stage ≥II), intestinal perforation, kidney insufficiency (stage 3 acute kidney injury using the neonatal modified KDIGO [Kidney Disease: Improving Global Outcomes] criteria15), or any event resulting in death or classified as a serious adverse event between randomization and 36 weeks’ postmenstrual age.
Statistical Analysis
We hypothesized that expectant management would reduce the incidence of death or BPD from 50% to 40% (an absolute risk reduction of 10%) compared with active treatment. We planned to enroll 836 patients, which would provide 80% power to identify the hypothesized difference in the primary outcome with an α (type I error) of .05, assuming a lost to follow-up rate of 5% and accounting for the 3 interim efficacy analyses.
The interim analyses were planned when 25%, 50%, and 75% of participants had outcome data at 36 weeks’ postmenstrual age. The independent NRN data and safety monitoring board could recommend stopping for futility if conditional power16 for the primary outcome was less than 15% at the 50% or 75% interim analysis. Efficacy of the primary outcome was evaluated at each interim analysis using a Lan-DeMets α-spending function17 with an O’Brien-Fleming stopping boundary,18 such that the final analysis was evaluated relative to P < .04. The safety outcomes of mortality, necrotizing enterocolitis, or a combination of these outcomes were evaluated at each interim analysis and at the final analysis using a Pocock stopping boundary19 of P < .02.
Per the intention-to-treat principle, outcomes were analyzed for all randomized participants. A prespecified per-protocol analysis excluded (1) participants with any protocol violation and (2) participants in the expectant management group who were discontinued from the trial and then treated prior to cardiopulmonary compromise. The protocol violations were confirmed by 3 investigators masked to group assignment when feasible. Participants were compared using robust Poisson regression to estimate relative risk for the outcomes, adjusting for trial center (with small centers pooled by geographic region) and gestational age stratum (<26 weeks’ vs ≥26 weeks’ gestation). The prespecified subgroups for the primary outcome were assessed by adding a subgroup × treatment group interaction term with a P < .10 prespecified cut point for evaluating within-subgroup comparisons.
Anthropometric z scores were compared using linear regression with the same adjustment. The efficacy outcomes and the adverse events were descriptively evaluated by the type of medication received and by timing of the administration of medication in the expectant management group (never, after cardiopulmonary compromise, or before cardiopulmonary compromise, which was protocol violation). No adjustments were made for the multiple comparisons in the secondary outcome analyses; however, adjustments were made for the interim analysis of the incidence of death and necrotizing enterocolitis. We also compared enrolled infants vs nonenrolled infants in the underlying registry to assess treatment practices outside the trial. All trial comparisons include unadjusted 95% CIs that should not be used for hypothesis testing.
The analyses were based on a statistical analysis plan (Supplement 2) approved by masked trial investigators prior to database lock and were performed using SAS software, version 9.4 (SAS Institute Inc).
Results
Participants
From December 11, 2018, to December 20, 2024, 5079 infants born before 29 weeks’ gestation were assessed for eligibility (Figure 1). Written informed consent was obtained for 487 eligible infants; of these, 482 were randomly assigned to either expectant management (n = 242) or active treatment (n = 240). The baseline characteristics were similar between groups (Table 1). The parents of an infant randomized to the expectant management group withdrew their consent for use of any data and those data were excluded from the analyses, leaving 241 infants in that group.
Treatments for PDA
Of the 241 infants in the expectant management group, 60 (24.9%) received pharmacological treatment to close the PDA before 36 weeks’ postmenstrual age; 21 (8.7%) received acetaminophen, 14 (5.8%) received ibuprofen, 16 (6.6%) received indomethacin, and 9 (3.7%) received more than 1 of these medications; and 24 (10%) underwent PDA ligation or cardiac catheterization. Of the 60 infants who received pharmacological treatment, 44 were treated before cardiopulmonary compromise and before 36 weeks’ postmenstrual age (a protocol violation) and 4 were treated after cardiopulmonary compromise and after 36 weeks’ postmenstrual age (not a protocol violation).
Of the 240 infants in the active treatment group, 238 (99.2%) received pharmacological treatment; 62 (25.8%) received acetaminophen, 85 (35.4%) received ibuprofen, 69 (28.8%) received indomethacin, and 22 (9.2%) received more than 1 of these medications; and 46 (19.2%) underwent a PDA ligation or cardiac catheterization (eTable 3 in Supplement 3).
Cardiopulmonary compromise occurred in 14.1% (34/241) of the infants in the expectant management group vs 9.2% (22/240) in the active treatment group. In regard to protocol violations, they occurred in 19.9% (48/241) of infants in the expectant management group vs 2.9% (7/240) in the active treatment group (eTable 4 in Supplement 3).
Primary Outcome
The trial was halted on December 23, 2024, by the Eunice Kennedy Shriver National Institute of Child Health and Human Development (study sponsor) for both futility in observing a difference of the primary outcome after the 50% interim analysis and for a safety concern of substantially increased mortality in 1 of the groups. At the 50% interim analysis, conditional power for the primary outcome was 2%. The primary outcome of death or BPD by 36 weeks’ postmenstrual age occurred in 80.9% (195/241) of infants in the expectant management group vs 79.6% (191/240) of infants in the active treatment group (adjusted risk difference, 1.2% [95% CI, −5.7% to 8.1%]; P = .73) (Table 2). The primary outcome results were consistent across the prespecified subgroups of gestational age, sex, PDA clinical criteria, and PDA size (eTable 5A in Supplement 3) and in the per-protocol population (eTable 6 in Supplement 3).
Table 2.
Primary and Secondary Outcomes
| Expectant management (n = 241)a | Active treatment (n = 240)b | Adjusted between-group difference (95% CI)c | Adjusted relativerisk (95% CI)d | P value | |
|---|---|---|---|---|---|
| Primary outcome e | |||||
| Death or BPD at 36 weeks’ postmenstrual age, No. (%)f | 195 (80.9) | 191 (79.6) | RD, 1.20 (−5.73 to 8.12) | 1.02 (0.93 to 1.11) | .73 |
|
| |||||
| Secondary outcomes (components of primary outcome) at 36 weeks’ postmenstrual age | |||||
| Death, No. (%) | 10 (4.1) | 23 (9.6) | RD, −5.63 (−10.05 to −1.21) | 0.41 (0.20 to 0.83) | .01 |
| BPD, No./total (%)f | 185/231 (80.1) | 168/217 (77.4) | RD, 2.20 (−5.21 to 9.61) | 1.03 (0.94 to 1.13) | .56 |
|
| |||||
| Other secondary outcomes at 36 weeks’ postmenstrual age | |||||
| Severity of BPD using Jensen et al14 definition, No. | 222 | 211 | |||
| Category, No. (%)g | |||||
| Grade I (mild) | 68 (30.6) | 57 (27.0) | |||
| Grade II (moderate) | 107 (48.2) | 104 (49.3) | |||
| Grade III (severe) | 28 (12.6) | 25 (11.8) | |||
| Moderate or severe BPD using Jensen et al14 definition, No. (%) | 135 (60.8) | 129 (61.1) | RD, 0.32 (−8.29 to 8.93) | 1.01 (0.87 to 1.16) | |
| Proven necrotizing enterocolitis, No. (%)h | 24 (10.0) | 27 (11.3) | RD, −1.30 (−6.75 to 4.15) | 0.88 (0.53 to 1.48) | |
| Medical (without surgery)i | 19 (7.9) | 13 (5.4) | |||
| Surgicalj | 5 (2.1) | 14 (5.8) | |||
| Necrotizing enterocolitis or death, No. (%) | 30 (12.4) | 44 (18.3) | RD, −6.02 (−12.30 to 0.27) | 0.67 (0.44 to 1.02) | |
| Length, No. | 221 | 205 | |||
| z Score, mean (SD) | −2.0 (0.8) | −2.0 (0.9) | MD, 0.04 (−0.12 to 0.20) | ||
| z Score, median (IQR) | −1.9 (−2.5 to −1.4) | −1.9 (−2.5 to −1.4) | |||
| Weight, No. | 222 | 206 | |||
| z Score, mean (SD) | −1.1 (0.8) | −1.1 (0.7) | MD, 0.01 (−0.13 to 0.15) | ||
| z Score, median (IQR) | −1.1 (−1.7 to −0.6) | −1.2 (−1.6 to −0.6) | |||
| Head circumference, No. | 221 | 202 | |||
| z Score, mean (SD) | −1.5 (1.0) | −1.6 (1.0) | MD, 0.05 (−0.14 to 0.23) | ||
| z Score, median (IQR) | −1.3 (−2.2 to −1.0) | −1.6 (−2.1 to −1.0) | |||
|
| |||||
| Secondary outcomes at 120 d k | |||||
| Died, No. (%) | 14 (5.8) | 27 (11.3) | RD, −5.67 (−10.53 to −0.81) | 0.49 (0.27 to 0.91) | |
| Severe retinopathy of prematurity, No./total (%)l | 49/229 (21.4) | 56/215 (26.0) | RD, −5.10 (−12.31 to 2.12) | 0.80 (0.59 to 1.09) | |
| Death or severe retinopathy of prematurity, No./total (%) | 64/237 (27.0) | 82/234 (35.0) | RD, −8.38 (−16.01 to −0.74) | 0.76 (0.59 to 0.98) | |
| Surgical closure, No. (%) | 40 (16.6) | 55 (22.9) | RD, −6.01 (−12.91 to 0.90) | 0.74 (0.52 to 1.05) | |
| Surgical ligation, No. (%) | 8 (3.3) | 13 (5.4) | |||
| Cardiac catheterization, No. (%) | 34 (14.1) | 44 (18.3) | |||
Abbreviations: BPD, bronchopulmonary dysplasia; MD, mean difference; RD, risk difference.
Treatment only after cardiopulmonary compromise (received acetaminophen, ibuprofen, or indomethacin; underwent cardiac catheterization or surgical ligation; or continued expected management).
Pharmacological treatment within 48 hours of diagnosis (received acetaminophen, ibuprofen, or indomethacin).
Estimated using robust generalized linear models (normal distribution with identity links). All models were adjusted for pooled trial center and gestational age strata (22–25 or 26–28 weeks).
Estimated using robust Poisson regression. All models were adjusted for pooled trial center and gestational age strata (22–25 or 26–28 weeks).
If the trial had not been stopped early for futility, the statistical significance criteria for the primary outcome would have been P < .04 using a Lan-DeMets α-spending function with an O’Brien-Fleming stopping boundary based on 3 planned interim assessments (the interim analyses were planned when 25%, 50%, and 75% of participants had outcome data at 36 weeks’ postmenstrual age).
To treat BPD, infants received support via a ventilator or continuous positive airway pressure or received low levels of supplemental oxygen (<30%). Also underwent a standardized assessment during a timed stepwise reduction of administered oxygen to room air with clearly defined oxygen saturation criteria from Walsh et al13 for “passing” or “failing.”
Categorized according to the mode of respiratory support administered regardless of the prior duration or current level of oxygen therapy.
Based on modified Bell staging criteria or via an autopsy.
Comprised stage IIA, stage IIB, or stage IIIA.
Comprised stage IIIB.
The median assessment time was at 117 (IQR, 91.5–120) days of life. A total of 421 infants (96%) were assessed through discharge, 18 (4%) through 120 days, and 1 was transferred to another hospital at 111 days. Severe retinopathy of prematurity, surgical ligation, and cardiac catheterization were assessed between randomization and the earlier of death, discharge, transfer, or 120 days of age. Death was analyzed at 36 weeks and 120 days of age.
Defined as stage 3 or worse (retinal detachment). Treatments included photocoagulation (laser), intraocular medication (bevacizumab injection), or an anti–vascular endothelial growth factor drug.
Secondary Outcomes
Before 36 weeks’ postmenstrual age, 4.1% (10/241) of infants in the expectant management group died vs 9.6% (23/240) in the active treatment group (adjusted risk difference, −5.6% [95% CI, −10.0% to −1.2%]; P = .01). This corresponds to a number needed to treat of 18, meaning that for every 18 infants treated with expectant management instead of active treatment, 1 more infant survived. Before discharge, transfer, or 120 days, 5.8% (14/241) of infants in the expectant management group died vs 11.3% (27/240) of infants in the active treatment group (adjusted risk difference, −5.7% [95% CI, −10.5% to −0.8%]). The results were similar in the per-protocol population (eTable 6 in Supplement 3). Mortality was similar among infants treated with each of the pharmacological treatments (eTable 7A and eTable 7B in Supplement 3) and across treatment status for the expectant management group (received no treatment vs received treatment before or after cardiopulmonary compromise; eTables 8–9 in Supplement 3).
In the post hoc analyses, a Kaplan-Meier plot (Figure 2A) showed divergence in mortality within 2 weeks of enrollment. A post hoc analysis of treatment heterogeneity for death (eTable 5B in Supplement 3) identified a treatment group × PDA clinical criteria interaction (protocol definitions appear in eTable 1 in Supplement 3). For mild PDA in the expectant management group vs in the active treatment group, the adjusted risk difference was −1.4% (95% CI, −8.2% to 5.3%), whereas the adjusted risk difference was −9.4% (95% CI, −15.0% to −3.9%) for moderate or severe PDA.
Figure 2. Secondary Outcomes of Death by 36 Weeks’ Postmenstrual Age and Severity Grades for Bronchopulmonary Dysplasia (BPD).
A, The median observation time was 60 days (IQR, 50–70 days). The P value was calculated using an unadjusted Kaplan-Meier log-rank test.
B, The severity of BPD was based on the Jensen et al14 definition.
Among infants who underwent screening and were deemed eligible but were not enrolled in the trial (eTable 10 in Supplement 3), the data from the NRN in-hospital registry20 of all infants showed that 66% (875/1320) received pharmacological treatments for PDA by discharge, transfer, or 120 days and 6.8% (92/1349) died by 36 weeks’ postmenstrual age (eTable 11 in Supplement 3).
The other differences for the secondary outcomes were seen only when mortality was included as part of the outcome (Table 2). At 36 weeks’ postmenstrual age, BPD occurred in 80.1% (185/231) of infants in the expectant management group vs 77.4% (168/217) in the active treatment group (adjusted risk difference, 2.2% [95% CI, −5.2% to 9.6%]). The data comparisons for the BPD severity grades appear in Figure 2B.
The occurrence of adverse events was mostly similar between groups (Table 3 and eTables 12–17 in Supplement 3). However, the incidence of infections resulting in death was 0.8% (2/241) in the expectant management group vs 3.8% (9/240) in the active treatment group (P = .04). The median time from randomization to onset of infections leading to death was 32 days (IQR, 18–45 days) in the expectant management group vs 9 days (IQR, 5–14 days) in the active treatment group (eTable 18 in Supplement 3). From randomization to 36 weeks’ postmenstrual age, the central line duration was similar in the expectant management group (14 days [IQR, 6–28 days]) vs in the active treatment group (15 days [IQR, 9–27 days]) (eTable 19 in Supplement 3).
Table 3.
Adverse Events (AEs) and Deaths
| No. (%)a | ||
|---|---|---|
| Expectant management (n = 241)b | Active treatment (n = 240)c | |
| By 36 weeks’ postmenstrual age d | ||
| Infants with ≥1 reported AE | 40 (16.6) | 45 (18.8) |
| Infants with ≥1 AE that resulted in death | 10 (4.1) | 23 (9.6) |
|
| ||
| AEs that resulted in death by MedDRA system organ class and preferred term | ||
| Infections and infestatione | 2 (0.8) | 9 (3.8) |
| Sepsis neonatal | 1 (0.4) | 5 (2.1) |
| Septic shock | 0 | 3 (1.3) |
| Group B streptococcus neonatal sepsis | 1 (0.4) | 0 |
| Pneumonia Escherichia | 0 | 1 (0.4) |
| Gastrointestinal disorder (neonatal necrotizing enterocolitis) | 3 (1.2) | 6 (2.5) |
| Respiratory, thoracic, and mediastinal disorder | 5 (2.1) | 4 (1.7) |
| Neonatal respiratory failure | 3 (1.2) | 0 |
| Neonatal respiratory distress syndrome | 1 (0.4) | 1 (0.4) |
| Acute respiratory failure | 1 (0.4) | 0 |
| Respiratory failure | 0 | 1 (0.4) |
| Pulmonary hemorrhage neonatal | 0 | 1 (0.4) |
| Neonatal pulmonary hypertension | 0 | 1 (0.4) |
| Cardiac disorder | 0 | 2 (0.8) |
| Cardiac arrest neonatal | 0 | 1 (0.4) |
| Cardiopulmonary failure | 0 | 1 (0.4) |
| Blood and lymphatic system disorder (splenic infarction) | 0 | 1 (0.4) |
| Hepatobiliary disorder (hepatic infarction) | 0 | 1 (0.4) |
| Nervous system disorder (posthemorrhagic hydrocephalus) | 0 | 1 (0.4) |
|
| ||
| Infant died before 36 weeks’ postmenstrual age, median (IQR) | ||
| Postmenstrual age, wk | 30.4 (25.7–50.7) | 27.9 (26.1–32.3) |
| Lifespan, wk | 5.8 (3.4–25.9) | 4.1 (2.3–7.3) |
|
| ||
| Infant died after 36 weeks’ postmenstrual age, No. (%) | ||
| Died before hospital discharge, facility transfer, or reaching 120 d | 4 (1.7) | 4 (1.7) |
| Had BPD only | 2 (0.8) | 2 (0.8) |
| Had BPD and pulmonary hypertension | 2 (0.8) | 0 |
| Had BPD and severe central nervous system insult | 0 | 2 (0.8) |
Abbreviations: BPD, bronchopulmonary dysplasia; MedDRA, Medical Dictionary for Regulatory Activities.
For each row, an infant is counted only once. Some of the data are expressed as median (IQR) as indicated below.
Treatment only after cardiopulmonary compromise (received acetaminophen, ibuprofen, or indomethacin; underwent cardiac catheterization or surgical ligation; or continued expected management).
Pharmacological treatment within 48 hours of diagnosis (received acetaminophen, ibuprofen, or indomethacin).
Reportable AEs included necrotizing enterocolitis, intestinal perforation, kidney insufficiency, or any other event that could be classified as a serious AE or related to trial procedures. Serious AEs were events that resulted in death, were life-threatening, required prolongation of hospitalization, resulted in persistent or significant disability or incapacity, or any other serious important medical events. An infant could have had multiple AEs, including multiple AEs that resulted in death. Additional AE information appears in eTables 12–17 in Supplement 3.
P = .04 for between-group comparison (calculated from the Fisher exact test).
Discussion
In this randomized clinical trial, expectant management did not reduce the incidence of the combined outcome of death or BPD at 36 weeks’ postmenstrual age compared with active treatment in preterm infants with a protocol-defined PDA at postnatal days’ 2 to 21 without cardiopulmonary compromise. However, mortality more than doubled in the active treatment group. Importantly, this finding is consistent with point estimates suggesting increased mortality in the BeNeDuctus trial,21 in the Baby-OSCAR trial,22 and in the PDA-TOLERATE trial,23 which individually did not reach statistical significance, as well as with a recent meta-analysis,6 which included these trials along with other smaller trials that did reach statistical significance.
There were important differences between the current trial and the Baby-OSCAR trial22 and the BeNeDuctus trial.21 In those trials,21,22 infants were screened and randomized within the first 72 hours after birth, whereas in the current trial the median age at randomization was 10 days. The Baby-OSCAR trial22 and the BeNeDuctus trial21 only used ibuprofen, whereas the current trial allowed any of the 3 common pharmacological treatments based on physician preference (acetaminophen, ibuprofen, or indomethacin). Ibuprofen exposure in the Baby-OSCAR trial22 and in the BeNeDuctus trial21 was linked to higher rates of moderate to severe BPD, raising concern about cyclooxygenase and peroxidase inhibitors in lung development; in contrast, the current trial did not find a difference in BPD after ibuprofen exposure. In the current trial, approximately one-fifth of infants in the expectant management group were treated before 36 weeks’ postmenstrual age and before meeting cardiopulmonary compromise criteria vs less than 3% in the Baby-OSCAR trial22 and less than 1% in BeNeDuctus trial.21 Such crossover may have biased the results toward the null. Taken together, the current trial more closely reflects clinical practice among clinicians who defer PDA screening until symptoms develop, whereas the Baby-OSCAR trial22 and the BeNeDuctus trial21 are more applicable to clinicians who screen and treat infants earlier. Nevertheless, despite the differences in trial design, timing of randomization, and choice of pharmacological treatment, the pharmacological closure of the PDA did not confer a clinical benefit in the Baby-OSCAR trial,22 in the BeNeDuctus trial,21 or in the current trial and was associated with harm.
In the current trial, the increased risk of mortality observed in the active treatment group may be partly attributable to a higher incidence of infection. One possible mechanism is that pharmacological treatment (whether with acetaminophen, ibuprofen, or indomethacin) is often accompanied by the withholding or delaying of enteral feedings or increasing the use of parenteral nutrition, both of which may increase the risk of sepsis.24 Alternatively, the medications themselves may alter the immune system25 or contribute directly to infection risk through reduced mesenteric blood flow26 or gastrointestinal mucosal injury,27 potentially facilitating bacterial translocation and sepsis.
Because of lack of evidence showing clinical benefit, the American Academy of Pediatrics Committee on the Fetus and Newborn recommended in 2025 against prophylactic or routine treatment of PDA in preterm infants during the first 2 postnatal weeks.28 An international survey of neonatologists demonstrated a lack of national guidelines, heterogenous methods of screening and management of PDA, and widely varied responses to statements regarding clinical equipoise, even among those who had been involved in prior PDA research.29 It is extraordinarily challenging to conduct a trial studying a therapy about which there are such variable but strongly held opinions. At the clinical centers participating in this trial, two-thirds of eligible but nonenrolled infants received pharmacological treatment.
The strengths of this trial include its multicenter design, large sample size, relevance to current neonatal approaches to PDA management, and use of clinically meaningful primary and secondary outcomes beyond PDA closure. This trial also used a consistent and objective definition of PDA based on common and accessible echocardiographic and clinical criteria, although the definition of symptomatic or clinically relevant PDA differs between studies.30 Neurodevelopmental follow-up at 2 years is ongoing.
Limitations
This study has limitations. First, the trial did not reach the full sample size due to meeting prespecified stopping criteria during the 50% interim analysis and a lower than expected enrollment rate. Second, by design, randomized treatment was not masked to allow clinicians to manage potential medication adverse effects such as oliguria. Third, echocardiographic timing was not standardized to a specific postnatal day, reflecting typical clinical practice. Fourth, the trial excluded the sickest infants—those with cardiopulmonary compromise, which is characterized by a large PDA and severe clinical signs.
Conclusions
In extremely preterm infants with a protocol-defined PDA, death or BPD did not differ between the expectant management group and the active treatment group. Survival was substantially higher with expectant management.
Supplementary Material
Key Points.
Question
Among preterm infants born at 22 to 28 weeks’ gestation with a protocol-defined patent ductus arteriosus, does expectant management compared with active treatment (pharmacological treatment with acetaminophen, ibuprofen, or indomethacin) decrease the incidence of death or bronchopulmonary dysplasia (BPD)?
Findings
In this multicenter, randomized clinical trial, the incidence of death or BPD at 36 weeks’ postmenstrual age was not different between groups (80.9% in expectant management group vs 79.6% in active treatment group). However, the incidence of death by 36 weeks’ postmenstrual age was lower in the expectant management group than in the active treatment group (4.1% vs 9.6%, respectively).
Meaning
The incidence of death or BPD was not different between groups, but more infants survived in the expectant management group.
Role of the Funder/Sponsor:
The NIH, the NICHD, and NCATS had a role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Funding/Support:
This study was supported by grants UG1 HD872293, UG1 HD112093, UG1 HD87226, UG1 HD68244, UG1 HD40689, UG1 HD40492, UG1 HD34216, UG1 HD68244, UG1 HD21364, UG1 HD112100, UG1 HD68278, UG1 HD68263, UG1 HD27851, UG1 HD27853, UG1 HD27880, UG1 HD112097, UG1 HD27853, UG1 HD40492, UG1 HD27904, UG1 HD68244, UG1 HD53089, UG1 HD27851, UG1 HD40492, UG1 HD53109, UG1 HD68263, UG1 HD40689, UG1 HD53109, UG1 HD112079, UG1 HD40492, and U24 HD95254 from the National Institutes of Health (NIH) and the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD) and by grants UL1 TR1067, UL1 TR2489, UL1 TR454, UL1 TR1425, UL1 TR1085, UL1 TR1425, UL1 TR1449, UL1 TR454, UL1 TR442, and UL1 TR1117 from the National Center for Advancing Translational Sciences (NCATS). The grants were provided through cooperative agreements with the Neonatal Research Network (NRN) to support researchers who conduct rigorous multicenter clinical trials aiming to improve health care outcomes of neonates. The NRN centers and the data coordinating centers are chosen competitively every 5 to 7 years through open competition as per usual NIH peer review processes.
Footnotes
Conflict of Interest Disclosures: Dr Ambalavanan reported owning stock in Alveolus Bio, serving as a medical adviser to Resbiotic, and serving on the data and safety monitoring board for Oak Hill Bio. Dr Davis reported serving as a consultant to Lansinoh. Dr Slaughter reported receiving grants from Abbott. Dr Cotten reported receiving personal fees from ReAlta Life Sciences and having a patent at Cryo-Cell International and receiving royalties. Dr Sánchez reported receiving personal fees from Merck Sharp and Dohme. No other disclosures were reported.
Group Information: The members of the Eunice Kennedy Shriver National Institute of Child Health and Human Development Neonatal Research Network appear in Supplement 4.
Disclaimer: The comments and views of the authors do not necessarily represent the views of the NIH, the NICHD, NCATS, the US Department of Health and Human Services, or the US government.
Meeting Presentation: Presented in part at Hot Topics in Neonatology; December 9, 2025; Washington, DC.
Data Sharing Statement: See Supplement 5.
Contributor Information
Matthew M. Laughon, Department of Pediatrics, University of North Carolina at Chapel Hill.
Sonia M. Thomas, Biostatistics and Epidemiology Division, RTI International, Research Triangle Park, North Carolina.
Kristi L. Watterberg, Department of Pediatrics, University of New Mexico Health Sciences Center, Albuquerque.
Kathleen A. Kennedy, McGovern Medical School, University of Texas Health Science Center, Houston.
Martin Keszler, Department of Pediatrics, Warren Alpert Medical School, Brown University, Providence, Rhode Island.
Namisavayam Ambalavanan, Division of Neonatology, University of Alabama, Birmingham.
Alexis S. Davis, Lucile Packard Children’s Hospital Stanford, Stanford University, Palo Alto, California.
Jonathan L. Slaughter, Nationwide Children’s Hospital, College of Medicine, Ohio State University, Columbus.
Ronnie Guillet, Division of Neonatology, School of Medicine and Dentistry, University of Rochester, Rochester, New York.
Tarah T. Colaizy, Department of Pediatrics, University of Iowa, Iowa City.
C. Michael Cotten, Department of Pediatrics, Duke University, Durham, North Carolina.
Megan A. Dhawan, Children’s Hospital of Philadelphia, University of Pennsylvania, Philadelphia.
Carl L. Bose, Department of Pediatrics, University of North Carolina at Chapel Hill.
Jennifer Talbert, Department of Pediatrics, University of North Carolina at Chapel Hill, Biostatistics and Epidemiology Division, RTI International, Research Triangle Park, North Carolina.
Sarah Smucny, Rainbow Babies and Children’s Hospital, Case Western Reserve University, Cleveland, Ohio.
William E. Benitz, Lucile Packard Children’s Hospital Stanford, Stanford University, Palo Alto, California.
Matthew A. Rysavy, McGovern Medical School, University of Texas Health Science Center, Houston.
Robin K. Ohls, University of Utah Health Sciences Center, Salt Lake City.
Mariana C. Baserga, University of Utah Health Sciences Center, Salt Lake City.
Sara B. DeMauro, Hospital of the University of Pennsylvania, Philadelphia.
Mambarambath Jaleel, Parkland Memorial Hospital, University of Texas Southwestern Medical Center, Dallas.
Wesley M. Jackson, Department of Pediatrics, University of North Carolina at Chapel Hill.
Waldemar A. Carlo, Division of Neonatology, University of Alabama, Birmingham.
Karen M. Puopolo, Pennsylvania Hospital, Philadelphia.
Anna Maria Hibbs, Rainbow Babies and Children’s Hospital, Case Western Reserve University, Cleveland, Ohio.
Anup Katheria, Sharp Mary Birch Hospital for Women and Newborns, San Diego, California.
Pablo J. Sánchez, Nationwide Children’s Hospital, College of Medicine, Ohio State University, Columbus.
Carl T. D’Angio, Division of Neonatology, School of Medicine and Dentistry, University of Rochester, Rochester, New York.
Ravi M. Patel, Department of Pediatrics, Emory University, Atlanta, Georgia.
Beth Ann Johnson, Good Samaritan Hospital, Cincinnati, Ohio.
Valerie Y. Chock, Lucile Packard Children’s Hospital Stanford, Stanford University, Palo Alto, California.
Abhay J. Bhatt, Department of Pediatrics, University of Mississippi Medical Center, Jackson.
Stephanie L. Merhar, University of Cincinnati Hospital, Cincinnati, Ohio.
Ryan Moore, Maynard Children’s Hospital, East Carolina University Health, Greenville, North Carolina.
Abbot R. Laptook, Department of Pediatrics, Warren Alpert Medical School, Brown University, Providence, Rhode Island.
Sarvin Ghavam, Virtua Voorhees Hospital, Voorhees, New Jersey.
Janell Fuller, Department of Pediatrics, University of New Mexico Health Sciences Center, Albuquerque.
Shilpa Vyas-Read, Grady Memorial Hospital, Emory University, Atlanta, Georgia.
Stephen D. Kicklighter, WakeMed Health and Hospitals, Raleigh, North Carolina.
Baiba Steinbrekera, Sanford Health, Sioux Falls, South Dakota.
Kathryn Anderson, Sanford Health, Sioux Falls, South Dakota.
Anne Marie Reynolds, University at Buffalo, Buffalo, New York.
Myra H. Wyckoff, William P. Clements Jr University Hospital, Dallas, Texas.
Cassandra Montoya, Prentice Women’s Hospital, Northwestern University, Chicago, Illinois.
Abhik Das, Social, Statistical, and Environmental Sciences Unit, RTI International, Rockville, Maryland.
Barbara Do, Biostatistics and Epidemiology Division, RTI International, Research Triangle Park, North Carolina.
Samantha Chang, Biostatistics and Epidemiology Division, RTI International, Research Triangle Park, North Carolina.
Rosemary D. Higgins, Eunice Kennedy Shriver National Institute of Child Health and Human Development, Bethesda, Maryland, Research and Sponsored Programs, Florida Gulf Coast University, Fort Myers.
Michele C. Walsh, Eunice Kennedy Shriver National Institute of Child Health and Human Development, Bethesda, Maryland.
REFERENCES
- 1.Gentile R, Stevenson G, Dooley T, Franklin D, Kawabori I, Pearlman A. Pulsed Doppler echocardiographic determination of time of ductal closure in normal newborn infants. J Pediatr 1981;98(3):443–448. doi: 10.1016/S0022-3476(81)80719-6 [DOI] [PubMed] [Google Scholar]
- 2.Semberova J, Sirc J, Miletin J, et al. Spontaneous closure of patent ductus arteriosus in infants ≤1500 g. Pediatrics. 2017;140(2):e20164258. doi: 10.1542/peds.2016-4258 [DOI] [PubMed] [Google Scholar]
- 3.Herrman K, Bose C, Lewis K, Laughon M. Spontaneous closure of the patent ductus arteriosus in very low birth weight infants following discharge from the neonatal unit. Arch Dis Child Fetal Neonatal Ed 2009;94(1):F48–F50. doi: 10.1136/adc.2007.129270 [DOI] [PubMed] [Google Scholar]
- 4.Sankar MN, Chock V, Myers F, et al. ; Eunice Kennedy Shriver National Institute of Child Health and Human Development Neonatal Research Network. Patent ductus arteriosus, hydrocortisone, and outcome among infants born extremely preterm: secondary analysis of the hydrocortisone trial. J Pediatr 2025;281:114535. doi: 10.1016/j.jpeds.2025.114535 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Mitra S, Scrivens A, Fiander M, Disher T, Weisz DE . Early treatment versus expectant management of hemodynamically significant patent ductus arteriosus for preterm infants. Cochrane Database Syst Rev 2025;6(6):CD013278. doi: 10.1002/14651858.CD013278.pub3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Buvaneswarran S, Wong YL, Liang S, Quek SC, Lee J. Active treatment vs expectant management of patent ductus arteriosus in preterm infants: a meta-analysis. JAMA Pediatr 2025;179(8):877–885. doi: 10.1001/jamapediatrics.2025.1025 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Bixler GM, Powers GC, Clark RH, Walker MW, Tolia VN. Changes in the diagnosis and management of patent ductus arteriosus from 2006 to 2015 in United States neonatal intensive care units. J Pediatr 2017;189:105–112. doi: 10.1016/j.jpeds.2017.05.024 [DOI] [PubMed] [Google Scholar]
- 8.Kaluarachchi DC, Rysavy MA, Carper BA, et al. Secular trends in patent ductus arteriosus management in infants born preterm in the National Institute of Child Health and Human Development Neonatal Research Network. J Pediatr 2024;266:113877. doi: 10.1016/j.jpeds.2023.113877 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Bell EF, Hintz SR, Hansen NI, et al. ; Eunice Kennedy Shriver National Institute of Child Health and Human Development Neonatal Research Network. Mortality, in-hospital morbidity, care practices, and 2-year outcomes for extremely preterm infants in the US, 2013–2018. JAMA. 2022; 327(3):248–263. doi: 10.1001/jama.2021.23580 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Mitra S, de Boode WP, Weisz DE, Shah PS. Interventions for patent ductus arteriosus (PDA) in preterm infants: an overview of Cochrane Systematic Reviews. Cochrane Database Syst Rev 2023;4(4):CD013588. doi: 10.1002/14651858.CD013588.pub2 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Lee JH, Greenberg RG, Quek BH, et al. Association between early echocardiography, therapy for patent ductus arteriosus, and outcomes in very low birth weight infants. Cardiol Young. 2017;27(9):1732–1739. doi: 10.1017/S1047951117001081 [DOI] [PubMed] [Google Scholar]
- 12.McNamara PJ, Sehgal A. Towards rational management of the patent ductus arteriosus: the need for disease staging. Arch Dis Child Fetal Neonatal Ed 2007;92(6):F424–F427. doi: 10.1136/adc.2007.118117 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Walsh MC, Yao Q, Gettner P, et al. ; National Institute of Child Health and Human Development Neonatal Research Network. Impact of a physiologic definition on bronchopulmonary dysplasia rates. Pediatrics. 2004;114(5):1305–1311. doi: 10.1542/peds.2004-0204 [DOI] [PubMed] [Google Scholar]
- 14.Jensen EA, Dysart K, Gantz MG, et al. The diagnosis of bronchopulmonary dysplasia in very preterm infants: an evidence-based approach. Am J Respir Crit Care Med 2019;200(6):751–759. doi: 10.1164/rccm.201812-2348OC [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Selewski DT, Charlton JR, Jetton JG, et al. Neonatal acute kidney injury. Pediatrics. 2015;136 (2):e463–e473. doi: 10.1542/peds.2014-3819 [DOI] [PubMed] [Google Scholar]
- 16.Lachin JM. A review of methods for futility stopping based on conditional power. Stat Med 2005;24(18):2747–2764. doi: 10.1002/sim.2151 [DOI] [PubMed] [Google Scholar]
- 17.Gordon Lan KK, DeMets D. Discrete sequential boundaries for clinical trials. Biometrika. 1983;70 (3):659–663. doi: 10.1093/biomet/70.3.659 [DOI] [Google Scholar]
- 18.O’Brien PC, Fleming TR. A multiple testing procedure for clinical trials. Biometrics. 1979;35(3): 549–556. doi: 10.2307/2530245 [DOI] [PubMed] [Google Scholar]
- 19.Pocock SJ. Group sequential methods in the design and analysis of clinical trial. Biometrika. 1977; 64:191–199. doi: 10.1093/biomet/64.2.191 [DOI] [Google Scholar]
- 20.Bell EF, Stoll BJ, Hansen NI, et al. Contributions of the NICHD Neonatal Research Network’s generic database to documenting and advancing the outcomes of extremely preterm infants. Semin Perinatol 2022;46(7):151635. doi: 10.1016/j.semperi.2022.151635 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Hundscheid T, Onland W, Kooi EMW, et al. ; BeNeDuctus Trial Investigators. Expectant management or early ibuprofen for patent ductus arteriosus. N Engl J Med 2023;388(11):980–990. doi: 10.1056/NEJMoa2207418 [DOI] [PubMed] [Google Scholar]
- 22.Gupta S, Subhedar NV, Bell JL, et al. ; Baby-OSCAR Collaborative Group. Trial of selective early treatment of patent ductus arteriosus with ibuprofen. N Engl J Med 2024;390(4):314–325. doi: 10.1056/NEJMoa2305582 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Clyman RI, Liebowitz M, Kaempf J, et al. ; PDA-TOLERATE (PDA: TO LEave it alone or Respond And Treat Early) Trial Investigators. PDA-TOLERATE trial: an exploratory randomized controlled trial of treatment of moderate-to-large patent ductus arteriosus at 1 week of age. J Pediatr 2019;205:41–48.e6. doi: 10.1016/j.jpeds.2018.09.012 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Stoll BJ, Hansen N, Fanaroff AA, et al. Late-onset sepsis in very low birth weight neonates: the experience of the NICHD Neonatal Research Network. Pediatrics. 2002;110(2 pt 1): 285–291. doi: 10.1542/peds.110.2.285 [DOI] [PubMed] [Google Scholar]
- 25.Bancos S, Bernard MP, Topham DJ, Phipps RP. Ibuprofen and other widely used non-steroidal anti-inflammatory drugs inhibit antibody production in human cells. Cell Immunol 2009;258 (1):18–28. doi: 10.1016/j.cellimm.2009.03.007 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Coombs RC, Morgan ME, Durbin GM, Booth IW, McNeish AS. Gut blood flow velocities in the newborn: effects of patent ductus arteriosus and parenteral indomethacin. Arch Dis Child 1990;65(10 Spec No):1067–1071. doi: 10.1136/adc.65.10_Spec_No.1067 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Bjarnason I, Scarpignato C, Holmgren E, Olszewski M, Rainsford KD, Lanas A. Mechanisms of damage to the gastrointestinal tract from nonsteroidal anti-inflammatory drugs. Gastroenterology. 2018;154(3):500–514. doi: 10.1053/j.gastro.2017.10.049 [DOI] [PubMed] [Google Scholar]
- 28.Ambalavanan N, Aucott SW, Salavitabar A, Levy VY; Committee on Fetus and Newborn; Section on Cardiology and Cardiac Surgery. Patent ductus arteriosus in preterm infants. Pediatrics. 2025;155(5):e2025071425. doi: 10.1542/peds.2025-071425 [DOI] [PubMed] [Google Scholar]
- 29.Hundscheid T, El-Khuffash A, McNamara PJ, de Boode WP. Survey highlighting the lack of consensus on diagnosis and treatment of patent ductus arteriosus in prematurity. Eur J Pediatr 2022;181(6):2459–2468. doi: 10.1007/s00431-022-04441-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Shepherd JL, Noori S. What is a hemodynamically significant PDA in preterm infants? Congenit Heart Dis 2019;14(1):21–26. doi: 10.1111/chd.12727 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.