Neonates With Tracheomalacia Generate Auto-Positive End-Expiratory Pressure via Glottis Closure

Abstract

Background: In pediatrics, tracheomalacia is an airway condition that causes tracheal lumen collapse during breathing and may lead to the patient requiring respiratory support. Adult patients can narrow their glottis to self-generate positive end-expiratory pressure (PEEP) to raise the pressure in the trachea and prevent collapse. However, auto-PEEP has not been studied in newborns with tracheomalacia. The objective of this study was to measure the glottis cross-sectional area throughout the breathing cycle and to quantify total pressure difference through the glottis in patients with and without tracheomalacia.

Research question: Do neonates with tracheomalacia narrow their glottises? How does the glottis narrowing affect the total pressure along the airway?

Study design and methods: Ultrashort echo time MRI was performed in 21 neonatal ICU patients (11 with tracheomalacia, 10 without tracheomalacia). MRI scans were reconstructed at four different phases of breathing. All patients were breathing room air or using noninvasive respiratory support at the time of MRI. Computational fluid dynamics simulations were performed on patient-specific virtual airway models with airway anatomic features and motion derived via MRI to quantify the total pressure difference through the glottis and trachea.

Results: The mean glottis cross-sectional area at peak expiration in the patients with tracheomalacia was less than half that in patients without tracheomalacia (4.0 ± 1.1 mm² vs 10.3 ± 4.4 mm²; P = .002). The mean total pressure difference through the glottis at peak expiration was more than 10 times higher in patients with tracheomalacia compared with patients without tracheomalacia (2.88 ± 2.29 cm H₂O vs 0.26 ± 0.16 cm H₂O; P = .005).

Interpretation: Neonates with tracheomalacia narrow their glottises, which raises pressure in the trachea during expiration, thereby acting as auto-PEEP.

Keywords: BPD; auto-PEEP; computational fluid dynamics; glottis; tracheomalacia.

Graphical abstract

Figure 1

A, Diagram showing the segmented airway surface and a cross-section of the glottis of a patient without tracheomalacia. B, Graph showing the variation of glottis cross-sectional area during the breathing cycle in one patient with tracheomalacia(red) and one patient without tracheomalacia (blue). Each data point on the curve represents a different breathing phase: end exp, peak insp, end insp, and peak exp. The glottis cross-sectional area of the patient with tracheomalacia at peak exp was approximately half of that patient’s glottis cross-sectional area at end exp. exp = expiration; insp = inspiration.

Figure 2

A, B, Box-and-whisker plots showing glottis cross-sectional area at peak expiration of the patients with tracheomalacia (red) and without tracheomalacia (blue) based on airway condition (A) and the clinical diagnosis (B). The average glottis cross-sectional area at peak expiration was larger for patients without tracheomalacia compared with patients with tracheomalacia. Plot elements: median = solid line; mean = cross; interquartile range (IQR) = box; data within 1.5 times the IQR less than 25% or more than 75% = whiskers. BEF = bronchoesophageal fistula; BPD = bronchopulmonary dysplasia; TEF/EA = tracheoesophageal fistula/esophageal atresia; TM = tracheomalacia.

Figure 3

Change in total pressure along the airway at peak expiration in one patient with tracheomalacia (red) and one without tracheomalacia (blue). Labels 1, 2, 3, and 4 indicate the positions of the nasopharynx, larynx, glottis, and carina, respectively. The two purple arrows on the top of each airway show the direction of airflow. The total pressure difference through the glottis for the patients with tracheomalacia and without tracheomalacia are indicated in red and blue dashed arrows, respectively.

Figure 4

A-C, Box-and-whisker plots showing total pressure difference through the glottis (A), total pressure difference from carina to larynx (B), and the ratio between (A) and (B) expressed as a percentage (C) at peak expiration. Plot elements: mean = cross; median = solid line; interquartile range (IQR) = box; data within 1.5 times the IQR less than 25% or more than 75% = whiskers; outlier = circle.

Figure 5

A, B, Box-and-whisker plots showing average glottis cross-sectional area during breathing (A) and total pressure difference from the larynx to carina at peak inspiration (B). Plot elements: mean = cross; median = solid line; interquartile range (IQR) = box; data within 1.5 times the IQR less than 25% or more than 75% = whiskers.