Impact of Surgical Diaphragmatic Repair on Central Airway Shape in Neonatal Congenital Diaphragmatic Hernia

ABSTRACT

Rationale

Neonates with congenital diaphragmatic hernia (CDH) frequently experience central airway abnormalities, including tracheobronchomalacia, which persist post‐surgical repair. However, these complications often remain underdiagnosed due to reliance on symptomatic evaluation and limited use of bronchoscopy. Ultrashort echo time (UTE) magnetic resonance imaging (MRI) offers a non‐invasive, three‐dimensional method to assess airway dynamics and evaluate tracheobronchomalacia.

Objectives

To quantify changes in central airway morphology and dynamics before and after surgical diaphragmatic repair in neonates with CDH using UTE MRI.

Methods

This study examined neonates with CDH admitted between 2015 and 2020 who underwent both pre‐ and post‐surgical UTE MRI. Airway dynamics were assessed using respiratory‐gated radial three‐dimensional UTE images at four breathing phases. Three‐dimensional airway surfaces were generated from the cricoid to the main bronchi. Changes from pre‐ to post‐surgery in cross‐sectional areas, dynamic collapse, and eccentricity index were analyzed.

Results

Six neonates with left‐sided CDH were included. Following surgery, tracheal minimum eccentricity index at end expiration decreased significantly (0.68 ± 0.06 to 0.53 ± 0.15, p = 0.023), indicating increased collapse. The ipsilateral bronchus demonstrated similar changes, with eccentricity index decreasing significantly from 0.65 ± 0.04 to 0.55 ± 0.11 (p = 0.048), while the contralateral bronchus remained stable. Mean tracheal cross‐sectional area decreased from 22.7 ± 6.1 mm² to 19.1 ± 3.1 mm² (p = 0.115), while dynamic motion increased from 21 ± 7% to 40 ± 15% (p = 0.055).

Conclusions

Neonates with CDH have significant central airway abnormalities pre‐surgery, which persists and may worsen post‐repair, particularly in the trachea and ipsilateral bronchus. These findings suggest tracheobronchomalacia is an underrecognized component of CDH. Integrating airway assessment into surgical planning could improve post‐operative outcomes.

1. Introduction

Congenital diaphragmatic hernia (CDH) occurs in approximately 1 in 2500–4000 live births and represents a significant challenge in neonatal care [1, 2]. The condition is characterized by an abnormal opening in the diaphragm that allows abdominal organs to enter the thoracic cavity, restricting lung development. This anatomical defect requires immediate intervention upon birth, with surgical repair of the diaphragm being the primary treatment. Despite successful surgical correction, these infants often experience compromised respiratory function requiring mechanical ventilation support during their first 3 years of life [3].

The displacement of abdominal organs into the thoracic cavity in CDH creates complex anatomical changes that extend beyond the diaphragm and lungs. The bronchus connected to the affected lung often develops an abnormal curvature due to spatial constraints within the thoracic cavity [4, 5, 6]. While diaphragmatic repair restores normal anatomical boundaries, the impact of this surgical intervention on airway shape and function remains poorly understood. Of particular concern is the potential development of tracheobronchomalacia, a condition characterized by excessive dynamic collapse of the trachea and/or bronchi during breathing.

Current evidence suggests that tracheobronchomalacia may be underrecognized in patients with CDH. A previous study identified tracheomalacia in 21 out of 323 CDH subjects, but this diagnosis relied primarily on symptomatic presentation rather than direct airway assessment [7]. While bronchoscopy remains the gold standard for diagnosing tracheobronchomalacia, it is not routinely performed in all CDH cases and lacks widespread availability in assessing infant patients. Consequently, diagnosis often depends on clinical symptoms such as barking cough or fixed wheezing, potentially underestimating the true prevalence of this condition [7].

Recent advances in imaging technology offer new opportunities for non‐invasive and non‐ionizing airway assessment. Ultrashort echo time (UTE) magnetic resonance imaging (MRI) has emerged as a valuable tool for evaluating tracheobronchomalacia, with previous studies demonstrating strong correlation between three‐dimensional imaging‐based assessments and bronchoscopy findings [8, 9, 10]. This technique is particularly suitable for neonatal patients and serial assessment, as it uses non‐ionizing radiation and can capture airway dynamics during natural breathing without requiring breath‐holds, sedation, or invasive devices which may obscure the underlying dynamics [11, 12, 13]. Gated UTE MRI enables visualization of the airway at distinct phases of respiration, providing detailed information about dynamic airway motion throughout the breathing cycle.

The present study aims to quantitatively assess airway structure and function in patients with CDH using UTE MRI, both before and after diaphragmatic repair surgery, in order to better understand the relationship between surgical repair and subsequent changes in airway abnormalities. This knowledge could inform strategies to optimize surgical approaches and post‐operative care, potentially improving respiratory outcomes in this vulnerable patient population.

2. Methods

2.1. Study Population

This research included neonates with CDH admitted to the neonatal intensive care unit (NICU) at Cincinnati Children's Hospital Medical Center between 2015 and 2020. Inclusion criteria required research MRI scans both before and after diaphragmatic repair surgery. During imaging sessions, participants received their clinically prescribed respiratory support as needed.

All imaging was performed following the approval from the Institutional Review Board of Cincinnati Children's Hospital Medical Center (protocol No.:2014–3621). Written consent was obtained from the families of neonatal participants prior to inclusion in the study.

2.2. Image Acquisition

Airway imaging was performed using a 1.5 T neonatal MRI scanner with a radial UTE sequence [11, 13, 14]. Each 16 min scan acquired approximately 200,000 projections at 0.7 mm isotropic resolution [11, 13]. Gated image reconstruction utilized the initial point of the free induction decay signal, which inherently encodes respiratory motion. Following identification and removal of bulk motion artifacts, the remaining respiratory motion data were categorized into four distinct phases of breathing: end expiration, peak inspiration, end inspiration, and peak expiration [11, 13, 14]. More detailed information on this MRI acquisition method has been published previously [12, 15, 16, 17].

2.3. Airway Segmentation and Surface Generation

Three‐dimensional virtual airway surfaces were generated from the MR images through segmentation of the airway lumen using active contour segmentation in ITK‐SNAP software (version 3.8, Penn Image Computing and Science Laboratory) [18]. The initial threshold intensity was set at the mean value between airway lumen and airway tissue intensities [13]. For each subject, the complete airway surface from the cricoid to the end of the main bronchi was segmented.

2.4. Quantitative Airway Analysis

2.4.1. Cross‐Sectional Area Measurements

Changes in airway dimensions before and after surgery were quantified through calculation of average cross‐sectional areas throughout the respiratory cycle for both pre‐ and post‐surgical scans. For tracheal analysis, centerlines were generated extending from the cricoid to the carina, with perpendicular cross‐sectional planes created at 1 mm intervals [19, 20, 21, 22, 23]. For bronchial analysis, separate centerlines were generated from the carina to the terminal point of each main bronchus.

2.4.2. Airway Dynamics Assessment

Airway dynamics were evaluated through measurement of the percentage change in cross‐sectional area when comparing airway surfaces from inspiration and expiration phases [9, 22, 24]. This calculation utilized the maximum area change relative to the inspiration area along the length of the airway. This metric provided a quantitative measure of dynamic airway collapse during breathing.

2.4.3. Eccentricity Index Calculation

Airway shape and potential malacia were assessed through calculation of the eccentricity index for each cross‐sectional plane. For each plane, the major diameter (longest distance between two points on the perimeter) and minor diameter (diameter measured perpendicular to the midpoint of the major diameter) were determined [9, 22]. The eccentricity index was calculated as the ratio of minor to major diameter, with lower values indicating more elliptical or collapsed airways. The minimum eccentricity index was identified for each airway region (trachea and main bronchi) to characterize the most severe shape abnormalities.

2.5. Statistical Analysis

Paired t‐tests with unequal variance were employed to evaluate changes in airway measurements before and after surgery. For comparisons with historical data from previous studies, unpaired t‐tests with unequal variance were utilized [22]. Statistical significance was defined as p < 0.05. Primary outcome measures included changes in cross‐sectional area, airway dynamics (percent change during breathing), and minimum eccentricity index.

3. Results

3.1. Study Population Characteristics

Among 22 neonates with CDH admitted to the NICU who underwent at least one airway research MRI, six participants met the criteria for pre‐ and post‐surgical imaging analysis. All participants had left‐sided hernias. The mean gestational age of the six subjects was 34.6 ± 2.6 weeks. The pre‐surgery MRI scan was typically performed about 3 days before the surgery, and the interval between the pre‐ and post‐surgical MRI scans was 29 ± 12 days. Body weight increased by a mean of 0.5 ± 0.6 kg between imaging timepoints. Detailed demographic information, including respiratory support requirements and individual weight changes, is presented in Table 1.

Table 1.

Subject information pre‐ and post‐surgery.

Variable	Subject 1	Subject 2	Subject 3	Subject 4	Subject 5	Subject 6
Sex	M	F	F	F	F	F
Gestational age (weeks)	34	32	38.6	32	33.6	37.6
FETO procedure	Yes	No	No	Yes	Yes	No
CDH severity PPLV (%)	21.4	29.6	N/A	25	N/A (LHR = 0.71)	17.7
Pre‐surgery weight at MRI (kg)	2.64	2.19	3.07	2.65	2.93	3.6
Post‐surgery weight at MRI (kg)	2.92	3.2	3.78	3	3.2	4
Pre‐surgery respiratory support and flow settings	Ventilated (BiVent), PIP = 17, PEEP = 5	Ventilated (BiVent), PIP = 15, PEEP = 5	Ventilated (BiVent), PIP = 15, PEEP = 5	Ventilated (BiVent), PIP = 22, PEEP = 5	Ventilated (PC SIMV + PS), PIP = 22, PEEP = 6	Ventilated (BiVent), PIP = 22, PEEP = 5
Post‐surgery respiratory support	CPAP (7 cmH₂O)	Room air	Room air	Nasal cannula (0.5 Lpm)	Nasal cannula (0.5 Lpm)	Room air
Number of days between MRI scans	13	40	28	49	23	20

Abbreviations: CPAP, continuous positive airway pressure; FETO, fetoscopic endotracheal occlusion; LHR, lung area to head circumference ratio; Lpm, liters per minute; PEEP, positive end‐expiratory pressure; PIP, positive inspiratory pressure; PC SIMV + PS, Pressure Controlled Synchronized Intermittent Mandatory Ventilation with Pressure Support; PPLV, percent predicted lung volume.

3.2. Tracheal Morphology

Tracheal cross‐sectional area measurements revealed notable post‐surgical changes. Figure 1 illustrates these changes in a representative participant, where cross‐sectional area variations are plotted for four respiratory phases: end expiration (blue), peak inspiration (red), end inspiration (orange), and peak expiration (purple). A marked reduction in tracheal cross‐sectional area was observed at the location 26 mm from the cricoid following surgery, with area values decreasing by approximately 50% across all respiratory phases.

Figure 1.

The cross‐sectional area of the trachea pre‐surgery (A) and post‐surgery (B) of an example subject with congenital diaphragmatic hernia. The cross‐sectional areas of the end expiration, peak inspiration, end inspiration, and peak expiration airways are shown in blue, red, orange, and purple, respectively. Numbers 1 and 2 demonstrate the position of the cricoid and carina of the airway. The airway cross‐sectional area at a location measured 26 mm from the cricoid was reduced from about 40 to 20 mm² following surgery, as indicated by the green dashed arrows. [Color figure can be viewed at wileyonlinelibrary.com]

Across all participants, mean tracheal cross‐sectional area decreased from 22.7 ± 6.1 mm² to 19.1 ± 3.1 mm² following surgery (mean difference: −3.6, 95% CI: −8.6 to 1.3; p = 0.115), as shown in Figure 2A. This reduction was observed in five of the six participants. Post‐surgical changes in airway dynamics were quantified through analysis of cross‐sectional area variations during the respiratory cycle. Figure 2B demonstrates that the maximum percentage change in tracheal cross‐sectional area increased from 21% ± 7% pre‐surgery to 40% ± 15% post‐surgery (mean difference: 19.3, 95% CI: −0.6 to 39.1; p = 0.055). Enhanced airway dynamics were observed in four of the six participants following surgery, indicating increased tracheal motion during breathing.

Figure 2.

The average tracheal cross‐sectional area (A) and the percentage of maximum change in the tracheal cross‐sectional area (B) of the six subjects both pre‐ and post‐surgery. On average, the average of the tracheal cross‐sectional area was reduced from 22.7 ± 6.1 to 19.1 ± 3.1 mm² following the surgery (mean difference: –3.6 mm², 95% CI: –8.6 to 1.3 mm²; p = 0.115). The average percentage of maximum change in tracheal cross‐sectional areas in pre‐ and post‐surgery was 21% ± 7% and 40% ± 15%, respectively (mean difference: 19.3, 95% CI: −0.6 to 39.1; p = 0.055). Plot elements: average = red cross; median = black line; interquartile range (IQR) = box; data within 1.5 times the IQR less than 25% or more than 75% = whiskers; squares with black lines = subjects 1–6. [Color figure can be viewed at wileyonlinelibrary.com]

Tracheal shape changes were characterized using the minimum eccentricity index. Throughout the respiratory cycle, the tracheal minimum eccentricity index decreased from 0.60 ± 0.02 pre‐surgery to 0.47 ± 0.12 post‐surgery (mean difference: −0.123, 95% CI: −0.262 to 0.015; p = 0.071). Figure 3 presents the analysis of specific respiratory phases, which revealed significant changes in airway shape. At end expiration, the mean minimum eccentricity index of the trachea decreased from 0.68 ± 0.06 to 0.53 ± 0.15 (mean difference: −0.15, 95% CI: −0.26 to 0.03; p = 0.023). Similarly, at end inspiration, the mean minimum eccentricity index decreased from 0.66 ± 0.06 to 0.52 ± 0.14 (mean difference: −0.15, 95% CI: −0.25 to −0.04; p = 0.016). These reductions in eccentricity index were observed in five of the six participants at both respiratory phases, indicating increased airway collapse.

Figure 3.

The minimum of the eccentricity index of the trachea at end expiration (A), and end inspiration (B) for pre‐ and post‐surgery. The average of the minimum of the eccentricity index decreased from 0.68 ± 0.06 to 0.53 ± 0.15 after surgery for the end expiration airway (mean difference: −0.15, 95% CI: −0.26 to 0.03; p = 0.023). On average, the minimum of the eccentricity index decreased from 0.66 ± 0.06 to 0.52 ± 0.14 after surgery for end inspiration airway (mean difference: ‐0.15, 95% CI: −0.25 to −0.04; p = 0.016). Plot elements: average = red cross; median = black line; interquartile range (IQR) = box; data within 1.5 times the IQR less than 25% or more than 75% = whiskers; squares with black lines = subjects 1–6. [Color figure can be viewed at wileyonlinelibrary.com]

3.3. Bronchial Morphology

Analysis of the bronchial airways revealed distinct patterns between ipsilateral and contralateral sides. The mean cross‐sectional area of the ipsilateral (left) bronchus increased from 7.62 ± 2.17 to 8.78 ± 2.47 mm² following surgery, though this change was not statistically significant (mean difference: 1.15, 95% CI: −0.69 to 3.00; p = 0.170). In contrast, the contralateral (right) bronchial cross‐sectional area remained more stable (pre‐surgery: 12.28 ± 4.12 mm², post‐surgery: 12.01 ± 1.86 mm², mean difference: −0.20, 95% CI: −4.12 to 3.73; p = 0.903).

The minimum eccentricity index of the ipsilateral bronchus decreased significantly from 0.65 ± 0.04 to 0.55 ± 0.11 following surgery (mean difference: −0.105, 95% CI: −0.207 to −0.002; p = 0.048) as shown in Figure 4. The contralateral bronchus showed a non‐significant decrease from 0.59 ± 0.08 to 0.47 ± 0.19 (mean difference: −0.118, 95% CI: −0.410 to 0.175; p = 0.348). Increased bronchial collapse was observed in five of six participants for the ipsilateral bronchus, while changes in the contralateral bronchus were more variable, with increased collapse noted in four participants.

Figure 4.

The minimum of the eccentricity index of the ipsilateral (A) and contralateral (B) bronchus throughout the respiratory cycle pre‐ and post‐surgery. On average, the minimum eccentricity index of the ipsilateral bronchus decreased from 0.65 ± 0.04 to 0.55 ± 0.11 after surgery (mean difference: −0.105, 95% CI: −0.207 to −0.002; p = 0.048). In the contralateral bronchus, the average minimum eccentricity index decreased from 0.59 ± 0.08 to 0.47 ± 0.19 (mean difference: −0.118, 95% CI: −0.410 to 0.175; p = 0.348). Plot elements: average = red cross; median = black line; interquartile range (IQR) = box; data within 1.5 times the IQR less than 25% or more than 75% = whiskers; squares with black lines = subjects 1–6. [Color figure can be viewed at wileyonlinelibrary.com]

3.4. Comparison With Historical Data

Figure 5 presents a comparison of tracheal minimum eccentricity indices between age‐matched participants with CDH and historical data from a previous study [22]. This comparison revealed that participants with CDH exhibited lower eccentricity indices both pre‐ and post‐surgery compared to individuals with diagnosed tracheobronchomalacia, suggesting more severe airway collapse in the CDH population.

Figure 5.

The minimum of the eccentricity index of the trachea in subjects with congenital diaphragmatic hernia (CDH) pre‐ and post‐surgery in the current study (left) and in subjects with and without tracheomalacia in a previous study (right). The average of the minimum of the eccentricity index decreased from 0.60 ± 0.02 to 0.47 ± 0.11 after surgery (mean difference: −0.123, 95% CI: −0.262 to 0.015; p = 0.071). Six subjects in the previous study were respiratory controls, and the remaining 18 subjects were diagnosed with bronchopulmonary dysplasia or tracheoesophageal defects. The average of the minimum of the eccentricity index of the trachea in subjects with and without tracheomalacia were higher compared to the subjects with CDH both pre‐ and post‐surgery. Plot elements: average = cross; median = black line; interquartile range (IQR) = box; data within 1.5 times the IQR less than 25% or more than 75% = whiskers; 95% confidence intervals = blue lines [22]. [Color figure can be viewed at wileyonlinelibrary.com]

4. Discussion

This study revealed significant airway abnormalities in patients with CDH at baseline, which were further increased following surgical repair. The key finding was the presence of tracheomalacia and ipsilateral bronchomalacia before and after diaphragmatic repair, evidenced by airway dynamics and more eccentric airway lumens. Post‐surgical airways showed reduced cross‐sectional areas compared to pre‐surgery despite the natural growth that would be expected over the intervening period (mean 29 days), suggesting that surgical intervention and/or associated with changes in respiratory support at the NICU may impact airway development.

The observed changes in airway shape were substantial. The minimum eccentricity index at end‐expiration decreased significantly from 0.68 ± 0.06 to 0.53 ± 0.15 (mean difference: −0.15, 95% CI: −0.26 to 0.03; p = 0.023), indicating a more collapsed airway state. More notably, tracheal dynamics nearly doubled from 21% ± 7% to 40% ± 15% following surgery, approaching statistical significance (mean difference: 19.3%, 95% CI: −0.6% to 39.1%; p = 0.055). These changes occurred despite weight gain averaging 0.5 kg across the cohort, suggesting that normal airway growth was disrupted due to the surgical intervention and/or associated with changes in respiratory support.

Several prior studies have used UTE MRI to evaluate neonatal airways with tracheomalacia [8, 9, 22]. Hysinger et al. [8] validated bronchoscopic findings against UTE‐MRI–derived metrics—most notably the maximum percent change in tracheal cross‐sectional area—and identified 40% as a diagnostic threshold for tracheomalacia with 48% sensitivity and 93% specificity. Receiver‐operating curves showed good ability of UTE MRI to identify tracheomalacia, with an area under the curve (AUC) of 0.78, indicating that the technique had solid overall diagnostic performance compared with bronchoscopy. In the current CDH study, the post‐surgery average maximum change was 40% ± 15%, indicating the mean change in area was right at the threshold of tracheomalacia diagnosis. In a separate study conducted by our group on neonates with tracheomalacia evaluated using UTE MRI, we identified an alternative diagnostic parameter—the tracheal minimum eccentricity index—with a threshold value of 0.7, which reliably corresponded with bronchoscopy findings. In the current study, the average minimum eccentricity index after surgery was 0.53 ± 0.15 at end expiration and 0.52 ± 0.14 at end inspiration—below the threshold for tracheomalacia at both time points.

When compared to previously studied neonatal populations (Figure 5), our CDH cohort showed more severe airway shape abnormalities. The tracheal eccentricity indices in our subjects were significantly lower than those reported in neonates with diagnosed tracheomalacia (p = 0.005 pre‐surgery, p = 0.013 post‐surgery) [22]. This suggests that CDH may be associated with more severe airway malacia than typically observed in other conditions such as bronchopulmonary dysplasia or tracheoesophageal defects. The finding that both pre‐ and post‐surgical measurements showed greater abnormality than the historical cohort raises important questions about the underlying developmental relationship between CDH and airway malacia.

These findings are important given previous studies which have shown that tracheomalacia affects airway mechanics and respiratory outcomes in neonates. The increased work of breathing associated with tracheomalacia can impact overall growth and development, as these infants must expend additional energy for respiration [13, 22, 25]. The presence of airway malacia may explain why some patients with CDH require prolonged respiratory support and experience delayed recovery despite successful diaphragmatic repair.

These results may have important clinical implications if they can be replicated in a larger cohort. The high prevalence and severity of airway malacia in our cohort suggest that routine airway assessment may be considered in the management of CDH. Early identification of tracheobronchomalacia could allow for simultaneous treatment during the initial diaphragmatic repair, potentially avoiding the need for additional surgical interventions. However, a larger and more comprehensive study of subjects with CDH is needed to provide clinicians with more detailed guidance.

Between 2015 and 2020, our institution treated 118 patients diagnosed with CDH. The average percent predicted lung volume (PPLV) in this study cohort (available in 4 out of the 6 subjects) was 23.4%. The average PPLV of all patients treated between 2015 and 2020 was 26.3%. In a previous multicenter study of 447 patients, only 89 (20%) underwent fetoscopic endotracheal occlusion (FETO), whereas in our study, the FETO group accounted for 50% of cases, indicating a higher proportion of severe cases in our cohort [26]. Several limitations should be considered when interpreting these results. The small sample size (n = 6) limits statistical power and generalizability. However, of the 16 subjects excluded from this study, eight underwent MRI scans before surgery, all of whom were ventilated during imaging. The other eight were scanned after surgery; among them, two were ventilated, five were breathing room air, and one was using a nasal cannula during the post‐surgery MRI. The respiratory data indicate that the six selected subjects can serve as a representative subset of the broader population, as most subjects exhibit similar respiratory support settings. Additionally, MRI resolution constraints restricted our analysis to the trachea and main bronchi, potentially missing important changes in smaller airways.

One consideration is that all these patients had an endotracheal tube in place before the repair, which may have helped keep the airway open, and only one remained on a ventilator afterward. Additionally, most, if not all, of these infants experienced a reduction in respiratory support (e.g., decreased positive end‐expiratory pressure (PEEP)). This raises the question of whether the reduced airway area and increased collapse are due to the surgery itself or the reduction in PEEP. However, the bronchial changes suggest a surgical cause, as they appear on only one side, whereas both sides are likely to equally affected by the changes in respiratory support. Future studies using higher‐resolution imaging techniques, such as Fermat looped, orthogonally encoded trajectories (FLORET) spiral UTE imaging or cine MRI, could provide more comprehensive assessment of the airway tree [27, 28]. Furthermore, longer‐term follow‐up would be valuable to understand whether these airway changes persist or resolve over time.

5. Conclusion

In conclusion, this study demonstrates that CDH and surgical repair of the herniation are associated with significant abnormalities in airway shape and dynamics, particularly in the trachea and ipsilateral bronchus. The severity of these abnormalities suggests that airway malacia should be considered an important aspect of post‐surgical CDH management. Future research should focus on developing protocols for early identification and management of airway malacia in patients with CDH, potentially leading to improved respiratory outcomes in this vulnerable population.

Author Contributions

Chamindu C Gunatilaka: conceptualization. methodology. formal analysis. visualization. writing – original draft; investigation. Sahr Alisher: formal analysis. writing – original draft. visualization. investigation. Margo Waters: formal analysis. writing – original draft. investigation. Qiwei Xiao: formal analysis. writing – original draft. Xavier Hoyos Cordon: methodology. writing – original draft. Nara S Higano: methodology. writing – original draft. Jason C Woods: writing – original draft. supervision. funding acquisition. Paul Kingma: conceptualization. writing – original draft. supervision. Alister J Bates: conceptualization. supervision. funding acquisition. writing – original draft.

Conflicts of Interest

The authors declare no conflicts of interest.

Gunatilaka C. C., Alisher S., Waters M., et al., “Impact of Surgical Diaphragmatic Repair on Central Airway Shape in Neonatal Congenital Diaphragmatic Hernia,” Pediatric Pulmonology 61 (2026): e71493. 10.1002/ppul.71493.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.