ABSTRACT
Pulmonary hypertension (PHTN) in infants with developmental lung disease, such as bronchopulmonary dysplasia (BPD), chronic lung disease of infancy (CLD), or congenital diaphragmatic hernia (CDH), can be exacerbated by atrial septal shunts secondary to atrial septal defects (ASD). While transcatheter ASD closure may reduce pulmonary overcirculation, data on post‐closure hemodynamic and pharmacologic outcomes remain limited. This single‐center retrospective study aimed to characterize changes in PHTN severity, respiratory support, and medication use 1 year after early transcatheter ASD closure (defined as closure at ≤ 1 year of age). Eligible patients were infants with BPD, CLD, or CDH who underwent early transcatheter ASD closure between 2021 and 2024 and had preprocedural PHTN medication use and respiratory support. Sixteen infants met the inclusion criteria. At 1 year, excluding the 3 who died, 10 of 13 infants (76.9%) showed improved PHTN severity, including 6 (60%) with complete resolution. Of the 13 infants, 6 (46.2%) weaned off all respiratory support. Average diuretic dosage (mg/kg/day) decreased by 92.9%, and vasodilator dosage declined by 47.0%. Infants with ASDs ≥ 5 mm and gestational age (GA) < 32 weeks required significantly longer diuretic therapy than those with smaller ASDs (< 5 mm) and GA ≥ 32 weeks. No similar associations were found with vasodilator weaning. These findings suggest early transcatheter ASD closure may offer therapeutic benefit in select high‐risk infants, resulting in improved hemodynamics and reduced medication dependence. Although limited by small sample size and retrospective design, this study supports the potential for individualized weaning strategies and the need for prospective multicenter investigations.
Keywords: chronic lung disease of infancy, eccentricity index, respiratory support, transcatheter intervention, vasodilator and diuretic weaning
1. Introduction
Pulmonary hypertension (PHTN) due to lung diseases or hypoxia has been included in Group 3 according to the 2018 World Symposium on PHTN, and the 2019 updated consensus recommendations of the European Pediatric Pulmonary Vascular Disease Network classification [1]. Group 3.5 was characterized by developmental lung disorders, which included: bronchopulmonary dysplasia (BPD), congenital diaphragmatic hernia (CDH), and other congenital or genetic lung disorders [2].
Although BPD, chronic lung disease (CLD), and CDH are distinct in etiology—BPD and CLD arising from postnatal lung injury and arrested alveolar development, and CDH from pulmonary hypoplasia and vascular maldevelopment due to diaphragmatic malformation—all three conditions can culminate in PHTN through pulmonary vascular remodeling, elevated pulmonary vascular resistance, and right ventricular strain [3, 4]. In CDH, the presence of left‐to‐right atrial‐level shunting due to an atrial septal defect (ASD) may lead to pulmonary overcirculation and impair hemodynamic adaptation in the neonatal period [5]. While these conditions predispose to prolonged and severe PHTN, CDH patients may inherently have a higher risk of early, fixed pulmonary vascular resistance elevation, whereas BPD and CLD patients more commonly exhibit progressive or evolving PHTN influenced by postnatal lung development [6, 7].
Superimposed cardiac lesions, such as ASDs, can further exacerbate pulmonary overcirculation and PHTN [8]. Vyas‐Read et al. reported that preterm infants with an ASD were significantly more likely to develop PHTN within the first 250 days of life compared to those without an ASD [9]. Chronic left‐to‐right shunting across an ASD is believed to contribute to long‐term respiratory morbidity in infants with BPD or CDH [8, 10]. While isolated ASDs are typically benign in healthy infants, early closure may be critical in infants with BPD or CDH to prevent progression of pulmonary vascular disease. Emerging data suggest that early transcatheter ASD closure (≤ 1 year of age) in infants with BPD can reduce medication burden and respiratory support needs; however, limited information exists regarding the trajectory of PHTN resolution or medication weaning following closure. Additionally, it remains unclear how baseline factors such as ASD size, gestational age (GA), and preprocedural PHTN severity are associated with vasodilator and diuretic weaning times.
Therefore, this study aims to characterize changes in PHTN severity and vasodilator and diuretic use at 1 year following early transcatheter ASD closure in infants with BPD, CLD, or CDH, and to identify clinical characteristics associated with medication weaning times.
2. Methods
2.1. Study Population and Design
We conducted a retrospective, single‐center cohort study at Johns Hopkins All Children's Hospital. Infants who had transcatheter ASD closure at ≤ 1 year of age from January 1, 2021 to October 15, 2024 were included in the study if they also had a diagnosis of BPD, CLD, or CDH and were prescribed at least one PHTN medication and required an additional form of respiratory support beyond room air at the time of transcatheter ASD closure. Infants were excluded from the study if they had PHTN‐associated genetic disorders (i.e., Trisomies/polyploidy, chromosomal deletions, duplications, or translocations) or significant cyanotic congenital heart disease (e.g., Tetralogy of Fallot, Truncus Arteriosus, Transposition of the Great Arteries). We utilized the 2019 Jensen criteria, which focuses on respiratory support needs at 36 weeks post‐menstrual age [11, 12, 13]. Infants born < 32 weeks GA who required respiratory support at 36 weeks post‐menstrual age had the diagnosis and severity of BPD classified according to Jensen criteria [11, 12, 13]. Infants born ≥ 32 weeks GA and required respiratory support at 36 weeks post‐menstrual age were classified as having “chronic lung disease of infancy,” which is thought to have a different mechanism of lung injury when compared with the typical arrest of alveolar development occurring in more immature infants born at earlier gestational ages [14]. CDH was confirmed through both fetal ultrasound and postnatal chest x‐ray. All ASD closures were performed by the same pediatric interventional cardiologist. Data were collected via manual chart review at specific time points. For PHTN severity, time points included 1‐month pre‐procedure, 24‐h postprocedure, and 1‐, 3‐, 6‐, and 12‐months postprocedure. For all other outcomes, data were collected at 24‐h pre‐procedure and 1‐, 3‐, 6‐, and 12‐months postprocedure.
2.2. Primary Outcome
The primary study outcome was PHTN severity, which was assessed using echocardiographic measurements interpreted by a board‐certified pediatric cardiologist. Severity was determined by either the ratio of right ventricular systolic pressure (RVSP) to systemic blood pressure (SBP) or by eccentricity index (EI), when tricuspid regurgitation velocity was unavailable.
2.3. Secondary Outcomes
PHTN‐specific medication dosage was defined as the average daily dosage (mg/kg/day) calculated separately for each medication: vasodilators (sildenafil, bosentan) and diuretics (furosemide, bumetanide, acetazolamide, spironolactone, chlorothiazide). Level of respiratory support was defined by the following groups: invasive (mechanical ventilation), noninvasive positive pressure ventilation, continuous positive airway pressure, heated high‐flow nasal cannula (≥ 2 liters per minute), low‐flow nasal cannula (< 2 liters per minute), and unassisted room air. Medication weaning time was defined as the median number of days per infant to discontinuation within each medication class (vasodilators vs. diuretics), based on the average weaning duration per patient.
2.4. Statistical Analysis
Demographic and clinical characteristics were summarized for the total cohort and by type of lung disease, with medians and interquartile ranges for continuous variables and counts with percentages for categorical variables. Study outcomes were analyzed, and comparisons were made between preprocedural and each post‐procedural time point. For multi‐categorical variables, including PHTN severity and level of respiratory support, McNemar‐Bowker tests for correlated proportions were performed to evaluate differences between pre‐ and post‐procedural time points. For continuous variables, including PHTN‐specific medication dosage, Wilcoxon Signed‐Rank tests were used to determine differences across time. Wilcoxon Rank Sum tests were used to evaluate intergroup differences in medication weaning time in days according to clinical characteristics.
RVSP was estimated using the modified Bernoulli equation (RVSP ≈ 4 v² + RAP), where v represents the tricuspid regurgitation jet velocity obtained via continuous‐wave Doppler, and right atrial pressure (RAP) was estimated or measured. If tricuspid regurgitation jet velocity was unavailable, PHTN severity was determined by either cardiologist interpretation or calculation of the end‐systolic EI, defined as the antero‐posterior diameter of the left ventricle parallel to the interventricular septum (D1) divided by the septo‐lateral diameter perpendicular to the interventricular septum (D2). PHTN severity was categorized as follows: none/normal (RVSP/SBP ≤¼ or EI ≤ 1.0), mild (RVSP/SBP >¼ to <½ or EI > 1.0 to < 1.3), and moderate‐to‐severe (RVSP/SBP ≥½ or EI ≥ 1.3). To evaluate interrater reliability, two board‐certified cardiologists independently reviewed one‐third of all echocardiograms. If there were disagreements in the interpretation of PHTN severity, a third cardiologist was consulted to provide a final determination of the severity. Prior literature has demonstrated that spot‐checking a representative subset is an established approach for assessing interrater reliability in echocardiographic studies [15].
Although EI is not yet a universally accepted metric for grading PHTN severity, prior studies support its clinical utility. Burkett et al. demonstrated strong correlations between EI and invasive hemodynamic measures, identifying end‐systolic EI values of 1.2–1.4 as corresponding to right ventricular pressures ≥ 50% systemic [16]. Based on these data, we defined moderate‐to‐severe PHTN as EI ≥ 1.3, a midpoint supported by both Burkett and Abraham et al., who found that EI ≥ 1.3 was associated with RVSP > 50% systemic pressure and higher risk for PHTN‐related complications [16, 17]. An EI of 1.0 was used to define the threshold for PHTN, reflecting the accepted normative cutoff in pediatric cardiology. Although Burkett et al. did not explicitly define “mild” PHTN, the progressive increase in EI and its correlation with invasive pressure suggest that intermediate RVSP/SBP values, such as those between ¼ and ½, may reflect early or mild elevations in pulmonary arterial pressure.
The average daily medication dosage (mg/kg/day) was calculated for each drug. Intravenous dosages were converted to their respective oral equivalents for consistency across time points. Only medications initiated before ASD closure were included in dosage and weaning analysis; postoperative medications were excluded. For infants who discontinued a medication, a dosage of zero (0) was assigned at that time point. Infants who had been on a specific medication before death were included on average daily dosage calculations up until the time of death.
To calculate the median of the average medication weaning time in days per infant, the average days to wean per infant (from initiation to discontinuation) was first computed across all drugs they received within each class (vasodilator vs. diuretic). The median of these per‐infant averages within the vasodilator and diuretic groups was then calculated to represent the overall central tendency of weaning duration for diuretics and vasodilators separately. Given power considerations, infants were stratified by clinical characteristics in a dichotomized fashion to compare weaning times. ASD size was categorized as large (septal defect ≥ 5 mm) or small (septal defect < 5 mm). GA at birth was categorized as extremely‐to‐very‐preterm (< 32 weeks) or moderate‐preterm‐to‐term (≥ 32 weeks), according to the World Health Organization classification of prematurity. PHTN severity was categorized as low (RVSP/SBP < ½ or D1/D2 < 1.3) or high (RVSP/SBP ≥ ½ or D1/D2 ≥ 1.3). If an infant had not discontinued a medication by the 12‐month follow‐up, a duration of 365 days was assigned as the maximum observation period. Infants who died before discontinuing a given medication were excluded from the weaning time analysis, as complete data on medication duration could not be obtained.
To control for Type 1 error due to multiple preprocedural and post‐procedural time point comparisons, a Bonferroni correction was applied by dividing the standard alpha level (0.05) by the number of tests performed. All statistical analysis was conducted using Stata/SE Version 17.1.
2.5. Ethical Approval
This study was approved by the Institutional Review Board (IRB) of Johns Hopkins All Children's Hospital (IRB #00433403), with a waiver of informed consent granted due to the retrospective nature of this study and use of deidentified data.
3. Results
3.1. Baseline Characteristics
As seen in Table 1, the cohort consisted of 16 infants with BPD (n = 6, 37.5%), CLD (n = 6, 37.5%), or CDH (n = 4, 25%) with 6 infants (37.5%) born less than 32 weeks gestation. Seven infants (43.8%) were classified as having a small ASD defect while the rest (n = 9, 56.2%) had large defects. The majority (75%) of patients had moderate PHTN at baseline, and 15 infants (94%) had at least one additional major comorbidity. The most common comorbidities were gastroesophageal reflux disease (93.8%), patent ductus arteriosus (87.5%), and sepsis (81.3%).
Table 1.
Demographic and clinical characteristics of the cohort.
| Characteristic | Total (n = 16) | Bronchopulmonary dysplasia (n = 6) | Chronic lung disease of infancy (n = 6) | Congenital diaphragmatic hernia (n = 4) |
|---|---|---|---|---|
| Age at time of procedure in days, median (interquartile range [IQR]) | 117.0 (90.8–192.0) | 137.0 (126.0–179.5) | 152.0 (94.8–218.3) | 85.5 (80.3–92.0) |
| Birthweight in kg, median (IQR) | 1.6 (0.6–2.5) | 0.6 (0.5–0.6) | 1.9 (1.6–2.5) | 2.6 (2.4–2.7) |
| Gestational age in weeks, median (IQR) | 33.6 (26.1–34.9) | 25.6 (23.0–26.6) | 34.3 (33.3–34.9) | 35.5 (34.6–36.6) |
| Size of atrial septal defect, median (IQR) | 6 (4.0–6.8) | 6 (4.5–7.3) | 6.2 (4.0–9.0) | 4 (3.5–5.0) |
| Procedural Weight in kg, median (IQR) | 4.0 (3.5–4.9) | 4.2 (2.5–5.2) | 4.4 (3.9–7.1) | 3.7 (3.5–3.9) |
| Female, n (%) | 7 (43.8) | 3 (50.0) | 2 (33.3) | 2 (50.0) |
| Bronchopulmonary dysplasia severity, n (%) | ||||
| Grade 1 (mild) | 0 (0.0) | 0 (0.0) | 0 (0.0) | 0 (0.0) |
| Grade 2 (moderate) | 1 (6.3) | 1 (16.7) | 0 (0.0) | 0 (0.0) |
| Grade 3 (severe) | 5 (31.3) | 5 (83.3) | 0 (0.0) | 0 (0.0) |
| Pulmonary hypertension severity (pre‐op), n (%) | ||||
| Mild | 4 (25.0) | 1 (16.7) | 2 (33.3) | 1 (25.0) |
| Moderate | 12 (75.0) | 5 (83.3) | 4 (66.7) | 3 (75.0) |
| Intrauterine growth restriction, n (%) | 7 (43.8) | 3 (50.0) | 4 (66.7) | 0 (0.0) |
| Survival to date, n (%) | 13 (81.3) | 5 (83.3) | 4 (66.7) | 4 (100.0) |
| Gastroesophageal reflux disease, n (%) | 15 (93.8) | 6 (100.0) | 6 (100.0) | 3 (75.0) |
| Aspiration pneumonia, n (%) | 9 (56.3) | 3 (50.0) | 5 (83.3) | 1 (25.0) |
| Tracheomalacia, n (%) | 4 (25.0) | 2 (33.3) | 2 (33.3) | 0 (0.0) |
| Bronchomalacia, n (%) | 4 (25.0) | 3 (50.0) | 1 (16.7) | 0 (0.0) |
| Thyroid disease, n (%) | 3 (18.9) | 1 (16.7) | 1 (16.7) | 1 (25.0) |
| Adrenal insufficiency, n (%) | 3 (18.8) | 3 (50.0) | 0 (0.0) | 0 (0.0) |
| Tracheostomy dependence, n (%) | 7 (43.8) | 2 (33.3) | 4 (66.7) | 1 (25.0) |
| Patent ductus arteriosus, n (%) | 14 (87.5) | 6 (100.0) | 4 (66.7) | 4 (100.0) |
| Pulmonary hypoplasia, n (%) | 8 (50.0) | 1 (16.7) | 3 (50.0) | 4 (100.0) |
| Sepsis, n (%) | 13 (81.3) | 6 (100.0) | 5 (83.3) | 2 (50.0) |
| Other comorbidities, n (%)* | 16 (100.0) | 6 (100.0) | 6 (100.0) | 4 (100.0) |
| Number of comorbidities, median (IQR) | 8.5 (6–10) | 9.5 (9–11) | 8.5 (7–10) | 6.0 (5–7) |
Other comorbidities include: omphalocele, necrotizing enterocolitis, intraventricular hemorrhage, retinopathy of prematurity, oligohydramnios, ventricular septal defect, urinary tract infection, or G‐tube dependence.
Before the procedure, furosemide was the most frequently used diuretic (n = 13, 81.3%), followed by chlorothiazide (n = 11, 68.8%), acetazolamide (n = 4, 25%), spironolactone (n = 3, 18.8%), and bumetanide (n = 2, 12.5%). Vasodilator therapy included sildenafil (n = 10, 62.5%) and bosentan (n = 1, 6.3%) (Table 2). Among infants on vasodilators at baseline, the average number of vasodilators per patient was 1.1. Among those on diuretics at baseline, the average number of diuretics per patient was 2.1.
Table 2.
Medication dosage (mg/kg/day) over time.
| Pre‐procedure | n | 1 month | n | p | 3 months | n | p | 6 months | n | p | 12 months | n | p | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Diuretics | ||||||||||||||
| Spironolactone | ||||||||||||||
| Mean (standard deviation [SD]) | 13.2 (18.9) | 3 | 3.4 (2.4) | 3 | 0.2 | 2.4 (2.5) | 3 | 0.1 | 2.3 (2.4) | 3 | 0.1 | 0.0 | 3 | — |
| Median (interquartile range [IQR]) | 3 (1.5–35) | 3 (1.3–6) | 1.3 (0.6–5.3) | 1.3 (0.5–5.0) | 0.0 | |||||||||
| Bumetanide | ||||||||||||||
| Mean | 0.2 | 2 | 0.2 | 2 | — | 0.07 | 2 | — | 0.0 | 2 | — | 0.0 | 2 | — |
| Median | 0.2 | 0.2 | 0.07 | 0.0 | 0.0 | |||||||||
| Acetazolamide | ||||||||||||||
| Mean (SD) | 7.5 (2.9) | 4 | 3.8 (4.9) | 4 | 0.3 | 2.5 (4.9) | 4 | 0.1 | 2.2 (4.4) | 4 | 0.1 | 1.9 (3.9) | 4 | 0.1 |
| Median (IQR) | 7.5 (5–10) | 2.5 (0–7.6) | 0.0 (0–4.9) | 0.0 (0–4.4) | 0.0 (0–3.9) | |||||||||
| Chlorothiazide | ||||||||||||||
| Mean (SD) | 28.8 (13.6) | 11 | 26.2 (16.8) | 11 | 0.3 | 18.6 (17.2) | 11 | 0.04 | 13.2 (14.8) | 10 | 0.02 | 2.4 (4.5) | 8 | 0.02 |
| Median (IQR) | 40 (15–40) | 39.1 (10–40) | 9.7 (0–40) | 9.4 (0–22.1) | 0.0 (0–4.8) | |||||||||
| Furosemide | ||||||||||||||
| Mean (SD) | 5 (2.6) | 13 | 3.6 (1.9) | 13 | 0.3 | 2.4 (1.8) | 13 | 0.002 | 1.1 (1.1) | 12 | 0.002 | 0.1 (0.5) | 10 | 0.002 |
| Median (IQR) | 4.7 (4–6) | 3 (2–4.1) | 2 (1.6–2.1) | 0.75 (0–2) | 0.0 (0–0) | |||||||||
| Vasodilators | ||||||||||||||
| Bosentan | ||||||||||||||
| Mean | 5.3 | 1 | 4.5 | 1 | — | 4.6 | 1 | — | 4.3 | 1 | — | 3.9 | — | |
| Sildenafil | ||||||||||||||
| Mean (SD) | 3.7 (2.2) | 10 | 4.4 (2.4) | 10 | 0.5 | 4.5 (2.3) | 10 | 0.2 | 4.0 (2.3) | 10 | 0.8 | 1.2 (1.1) | 10 | 0.1 |
| Median (IQR) | 3 (2.4–6) | 3.6 (2.6–6) | 4.1 (2.9–6.2) | 3.0 (2.6–5.5) | 1.5 (0–1.7) | |||||||||
Note: All p‐values are calculated from pre‐procedure using Wilcoxon signed rank tests.
Given the number of tests (4), alpha level was lowered to p = 0.013
Three infants died during the 12‐month follow‐up period; none of the deaths were attributed to ASD closure. Two infants experienced cardiac arrest approximately 6 months post‐closure, and a third died from cardiac arrest around 9 months following the procedure.
3.2. Clinical Outcomes Following ASD Closure
As seen in Figure 1, excluding the 3 infants who died, 10 of the remaining 13 infants (76.9%) demonstrated an improvement in PHTN severity by the 1‐year postoperative time point. Of these 10 infants, 8 of them (80.0%) showed improvement by 3‐months postprocedure and 6 of them (60%) experienced complete resolution of PHTN by 12‐months postprocedure. Significant improvements were seen at the following postoperative time points: 3 months (p = 0.005), 6 months (p = 0.001), and 12 months (p = 0.004). Despite overall improvement, 3 infants (30%) had persistent moderate PHTN at 12 months. These infants had notable comorbidities including tracheostomy dependence and pulmonary hypoplasia.
Figure 1.
Pulmonary Hypertension (PHTN) severity level of infants by time point. Asterisks represent a significant difference in the proportion of PHTN severity levels from pre‐procedure. One infant was missing data at 3 months; 1 infant died at 6 months, and 2 additional infants died at 12 months postprocedure.
While not statistically significant, patients exhibited clinically meaningful improvements in their baseline supplemental respiratory support at the 1‐year postoperative time point (Figure 2). Of the 7 patients receiving invasive respiratory support at the time of ASD closure, 4 infants (57.1%) had reduced requirements at 1‐year postprocedure. Of those requiring invasive support at baseline, 2 infants (28.6%) died, 2 infants (28.6%) remained on invasive support, 1 infant (14.3%) transitioned to noninvasive, 1 infant (14.3%) transitioned to LFNC, and 1 infant (14.3%) was able to wean to room air by 1 year following ASD closure. Similarly, among the 6 infants requiring noninvasive respiratory support, 5 infants (83.3%) had a decrease in respiratory support requirements at the 1‐year postoperative time point. Of those requiring noninvasive support at baseline, 1 infant (16.7%) died, 2 infants (33.3%) transitioned to LFNC, and 3 infants (50%) transitioned to room air by 1 year following ASD closure. In all, 6 of the remaining 13 surviving infants (46.2%) transitioned to room air by the 1‐year postoperative time point.
Figure 2.
Level of respiratory support of infants by time point. One infant died at 6 months, and 2 additional infants died at 12 months postprocedure.
As seen in Table 2 and Figure 3, median diuretic dosage declined by 100% (p = 0.002), with the mean decreasing by 92.9% by 12 months postprocedure. Median and mean vasodilator dosage declined by approximately 50%, though these decreases were not statistically significant. All 16 infants were on diuretics. Excluding the 3 infants who died, 1 infant (7.7%) was weaned off diuretics at 3‐months postprocedure, with an additional 4 (30.8%) and 6 infants (46.2%) weaned off at 6 and 12 months, respectively, totaling 11 patients (84.6%) by the end of the study. Ten infants were on vasodilators. Excluding the 2 infants on vasodilators who died, only 3 infants had weaned off vasodilators by 12 months (37.5%).
Figure 3.
Change in median dosage (mg/kg/day) of vasodilators and diuretics over time. The individual squares in the graphs represent the dosage (mg/kg/day) of study infants at each time point. p‐values reported indicate significant differences in dosages from pre‐procedure according to the corrected α.
Infants with large ASDs demonstrated longer diuretic weaning times than those with small ASDs (278.7 vs. 86.8, respectively, p = < 0.001). Extremely‐to‐very‐preterm infants also demonstrated longer diuretic weaning times compared to moderate‐preterm‐to‐term infants (309.7 vs. 134.5, respectively, p = 0.02). There were no associations between ASD size or GA and vasodilator weaning time (Figures 4, 5, 6). One infant was excluded from the weaning time analysis due to death before discontinuation of vasodilator and diuretic medications, which precluded accurate determination of medication duration.
Figure 4.
Median of the average weaning days per infant by atrial septal defect size for vasodilators and diuretics.
Figure 5.
Median of the average weaning days per infant by gestational age for vasodilators and diuretics.
Figure 6.
Median of the average weaning days per infant by baseline pulmonary hypertension severity for vasodilators and diuretics.
4. Discussion
In this retrospective study of infants with BPD, CLD, or CDH undergoing early transcatheter ASD closure, we observed clinically meaningful improvements in cardiorespiratory status at the 1‐year postoperative time point. Specifically, we found significant reductions in PHTN severity and the need for medication and a downward trend in respiratory support requirements over the 12‐month follow‐up period. These findings highlight the therapeutic impact of ASD closure in promoting hemodynamic stabilization and clinical recovery in this high‐risk population.
Our findings are consistent with prior studies demonstrating the benefit of ASD closure in preterm infants with CLD. Wood et al. and Yung et al. previously reported improvement in respiratory outcomes and reduced diuretic use following ASD closure in similar cohorts [18, 19]. Webb et al. developed a novel “BPD‐ASD” score to assess clinical status after ASD closure in BPD patients who weighed ≤ 10 kg, noting significant improvement in BPD‐ASD score, diuretic use, and respiratory support that was outside the natural course of BPD [20]. Our study reflected similar trends, showing significant improvements in PHTN severity as early as 3 months postprocedure, that persisted through 12 months postprocedure. Those with moderate PHTN continued to improve between 6 and 12 months, emphasizing the potential for delayed remodeling and a variable timeline of hemodynamic recovery. While prior studies have shown improvement in respiratory outcomes and reduced diuretic use, our study further quantifies these effects with detailed trajectory data on PHTN severity and medication weaning, which was stratified by baseline clinical characteristics.
In addition, we explored the influence of ASD size, GA, and baseline PHTN severity on the time required to wean off medications, offering new insights into how individual patient factors may guide clinical decision‐making. Additionally, our cohort had a notably high comorbidity burden, reinforcing the relevance of our findings in a real‐world, medically complex population.
We further explored specific associations between baseline characteristics and weaning outcomes. Diuretic weaning was significantly delayed in infants with larger ASDs and earlier GA, supporting the concept that both the magnitude of left‐to‐right shunting and pulmonary immaturity exacerbate volume overload and delay cardiopulmonary recovery. These findings suggest that although ASD closure facilitates improvement, the severity of baseline characteristics can still modulate the pace of pharmacologic offloading. Vasodilator weaning times did not vary significantly by these baseline factors, suggesting a more cautious and uniform approach, possibly reflecting clinician preference to delay therapy reduction until stable hemodynamics and durable vascular remodeling are achieved. In practice, clinicians often rely on serial echocardiographic assessments, clinical appearance, and oxygen requirements when determining readiness for medication wean. The observed reduction in both diuretic and vasodilator use likely reflects improvements in both respiratory and cardiovascular status. Medication weaning, while not a direct surrogate for disease resolution, can serve as a pragmatic marker of improving pulmonary hemodynamics and gas exchange. However, it is important to note that weaning strategies remain variable and without standardized guidelines.
We recognize the limitations of our retrospective and single‐center study design and its small sample size. Our inability to analyze BPD, CLD, and CDH subgroups separately limits conclusions regarding differential response between these conditions, making generalizable conclusions challenging. The lack of a control group—infants with developmental lung disease who did not undergo transcatheter ASD closure—limited direct comparisons and the ability to definitively determine the benefit of ASD closure in this high‐risk population. Additionally, we were unable to control for potentially confounding comorbidities such as recurrent infections, aspiration, or variations in supportive care. One comorbidity in particular, patent ductus arteriosus, was present in 87.5% of infants, and may have influenced clinical outcomes independently of ASD closure; however, in most subjects, the patent ductus arteriosus had either closed spontaneously or was treated via transcatheter intervention, typically occurring approximately 3 to 4 months before ASD closure. Only one subject had a persistent patent ductus arteriosus following ASD closure, which may have impacted this infant's clinical trajectory (i.e., weaning of respiratory support and PHTN‐specific medications). Given these limitations, we wish to emphasize that the findings from this study are not intended as prescriptive treatment recommendations, but rather as a consideration for other institutions evaluating the potential role of early transcatheter ASD closure in infants with developmental lung disease. Our intent is to share observational outcomes that may help inform future studies and clinical decision‐making, rather than to advocate for routine closure in all such patients.
Early closure may be considered as part of a comprehensive management strategy in high‐risk infants with developmental lung disease. Our findings suggest that baseline factors such as larger ASD size and earlier GA may be useful clinical indicators for anticipating longer durations of diuretic therapy. Future prospective multicenter studies with larger cohorts and standardized follow‐up protocols are essential to validate our findings and to better define the optimal timing and strategies for post‐closure management in this complex patient population. Additionally, further research is needed to identify predictors of successful outcomes and to refine post‐procedural care, including individualized approaches to weaning pulmonary medications in vulnerable infants with developmental lung disease.
Author Contributions
Grace A. Freire, MD, and Bruce F. Landeck, MD, contributed to the interpretation of echocardiographic data. Joana S. Machry, MD, Amy L. Kiskaddon, PharmD, MBA, Grace Freire, MD, and Marisol Betensky, MD, MPH, assisted in reviewing and editing this manuscript. James A. Thompson, MD, performed all transcatheter atrial septal defect closures and provided preliminary demographic information for the study cohort. Statistical analysis was conducted by Johns Hopkins All Children's Hospital biostatisticians Jamie L. Fierstein, PhD, and Dina Ashour. All authors contributed to the review of the article and approved the final version for submission.
Conflicts of Interest
The authors declare no conflicts of interest.
Acknowledgments
The authors would like to acknowledge Karly Anderson, Senior Clinical Data and Analytic Specialist for the Institute of Clinical and Translational Research at Johns Hopkins All Children's Hospital, for her assistance in building the REDCap database used for the completion of the data analysis for this study. We also thank Maher Abadeer, MD, pediatric cardiologist at Johns Hopkins All Children's Hospital, for contributing to the echocardiographic review process as the third cardiologist to support interrater reliability for our study. This study received the support of the Johns Hopkins All Children's Foundation Institutional Research Award Grant through the Clinical and Translational Research Track and the funding agency had no role in the design of the study; in the collection, analysis, or interpretation of data; in the writing of the article; or in the decision to publish the results.
Wiegand J. L., Thompson J. A., Landeck B. F., et al., “Pulmonary Hypertension Outcomes After Closure of Atrial Septal Defect in Infants With Developmental Lung Disease,” Pulmonary Circulation 15 (2025): 1‐11, 10.1002/pul2.70164.
References
- 1. Hansmann G., Koestenberger M., Alastalo T. P., et al., “2019 Updated Consensus Statement on the Diagnosis and Treatment of Pediatric Pulmonary Hypertension: The European Pediatric Pulmonary Vascular Disease Network (EPPVDN), Endorsed by AEPC, ESPR and ISHLT,” Journal of Heart and Lung Transplantation 38, no. 9 (2019): 879–901, 10.1016/j.healun.2019.06.022. [DOI] [PubMed] [Google Scholar]
- 2. Rosenzweig E. B., Abman S. H., Adatia I., et al., “Paediatric Pulmonary Arterial Hypertension: Updates on Definition, Classification, Diagnostics and Management,” European Respiratory Journal 53, no. 1 (2019): 1801916, 10.1183/13993003.01916-2018. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Kim D. H., Kim H. S., Choi C. W., Kim E. K., Kim B. I., and Choi J. H., “Risk Factors for Pulmonary Artery Hypertension in Preterm Infants With Moderate or Severe Bronchopulmonary Dysplasia,” Neonatology 101, no. 1 (2012): 40–46, 10.1159/000327891. [DOI] [PubMed] [Google Scholar]
- 4. Gupta V. S. and Harting M. T., “Congenital Diaphragmatic Hernia‐Associated Pulmonary Hypertension,” Seminars in Perinatology 44, no. 1 (2020): 151167, 10.1053/j.semperi.2019.07.006. [DOI] [PubMed] [Google Scholar]
- 5. Engel C., Leyens J., Bo B., et al., “Arterial Hypertension in Infants With Congenital Diaphragmatic Hernia Following Surgical Repair,” European Journal of Pediatrics 183, no. 7 (2024): 2831–2842, 10.1007/s00431-024-05509-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Hansmann G., Sallmon H., Roehr C. C., Kourembanas S., Austin E. D., and Koestenberger M., “Pulmonary Hypertension in Bronchopulmonary Dysplasia,” Pediatric Research 89, no. 3 (2021): 446–455, 10.1038/s41390-020-0993-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Charisopoulou D., Bini R. M., Riley G., Janagarajan K., Moledina S., and Marek J., “Repair of Isolated Atrial Septal Defect in Infants Less Than 12 Months Improves Symptoms of Chronic Lung Disease or Shunt‐Related Pulmonary Hypertension,” Cardiology in the Young 30, no. 4 (2020): 511–520, 10.1017/S1047951120000463. [DOI] [PubMed] [Google Scholar]
- 8. Thomas V. C., Vincent R., Raviele A., Diehl H., Qian H., and Kim D., “Transcatheter Closure of Secundum Atrial Septal Defect in Infants Less Than 12 Months of Age Improves Symptoms of Chronic Lung Disease,” Congenital Heart Disease 7, no. 3 (2012): 204–211, 10.1111/j.1747-0803.2010.00442.x. [DOI] [PubMed] [Google Scholar]
- 9. Vyas‐Read S., Guglani L., Shankar P., Travers C., and Kanaan U., “Atrial Septal Defects Accelerate Pulmonary Hypertension Diagnoses in Premature Infants,” Frontiers in Pediatrics 6 (2018): 342, 10.3389/fped.2018.00342. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Haregu F., McCulloch M., Vergales B., Garrod A., Conaway M., and Hainstock M., “Transcatheter Occlusion of Left‐To‐Right Shunts in Premature Infants With Bronchopulmonary Dysplasia,” Neonatology 120, no. 1 (2023): 57–62, 10.1159/000527267. [DOI] [PubMed] [Google Scholar]
- 11. Jensen E. A., Dysart K., Gantz M. G., et al., “The Diagnosis of Bronchopulmonary Dysplasia in Very Preterm Infants. An Evidence‐Based Approach,” American Journal of Respiratory and Critical Care Medicine 200, no. 6 (2019): 751–759, 10.1164/rccm.201812-2348OC. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Vyas‐Read S., Logan J. W., Cuna A. C., et al., “A Comparison of Newer Classifications of Bronchopulmonary Dysplasia: Findings From the Children's Hospitals Neonatal Consortium Severe BPD Group,” Journal of Perinatology 42, no. 1 (2022): 58–64, 10.1038/s41372-021-01178-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Higgins R. D., Jobe A. H., Koso‐Thomas M., et al., “Bronchopulmonary Dysplasia: Executive Summary of a Workshop,” Journal of Pediatrics 197 (2018): 300–308, 10.1016/j.jpeds.2018.01.043. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Colin A. A., McEvoy C., and Castile R. G., “Respiratory Morbidity and Lung Function in Preterm Infants of 32 to 36 Weeks' Gestational Age,” Pediatrics 126, no. 1 (2010): 115–128, 10.1542/peds.2009-1381. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. McCrary A. W., Barker P. C. A., Torok R. D., et al., “Agreement of an Echocardiogram‐Based Diagnosis of Pulmonary Hypertension in Infants at Risk for Bronchopulmonary Dysplasia Among Masked Reviewers,” Journal of Perinatology 39, no. 2 (2019): 248–255, 10.1038/s41372-018-0277-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Burkett D. A., Patel S. S., Mertens L., Friedberg M. K., and Ivy D. D., “Relationship Between Left Ventricular Geometry and Invasive Hemodynamics in Pediatric Pulmonary Hypertension,” Circulation: Cardiovascular Imaging 13, no. 5 (2020): e009825, 10.1161/CIRCIMAGING.119.009825. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Abraham S. and Weismann C. G., “Left Ventricular End‐Systolic Eccentricity Index for Assessment of Pulmonary Hypertension in Infants,” Echocardiography 33, no. 6 (2016): 910–915, 10.1111/echo.13171. [DOI] [PubMed] [Google Scholar]
- 18. Wood A. M., Holzer R. J., Texter K. M., et al., “Transcatheter Elimination of Left‐To‐Right Shunts in Infants With Bronchopulmonary Dysplasia Is Feasible and Safe,” Congenital Heart Disease 6, no. 4 (2011): 330–337, 10.1111/j.1747-0803.2011.00540.x. [DOI] [PubMed] [Google Scholar]
- 19. Yung D., Jackson E. O., Blumenfeld A., et al., “A Multidisciplinary Approach to Severe Bronchopulmonary Dysplasia Is Associated With Resolution of Pulmonary Hypertension,” Frontiers in Pediatrics 11 (2023): 1077422, 10.3389/fped.2023.1077422. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Webb M. K., Cuevas Guaman M., Sexson Tejtel S. K., et al., “Atrial Septal Defect Closure Is Associated With Improved Clinical Status in Patients ≤10 Kg With Bronchopulmonary Dysplasia,” Pulmonary Circulation 13, no. 4 (2023): e12299, 10.1002/pul2.12299. [DOI] [PMC free article] [PubMed] [Google Scholar]