Abstract Rationale Bronchopulmonary dysplasia (BPD) is characterized by arrested development and injury of the lung parenchyma and airways. Tracheal enlargement may impair an infant’s ability to wean off respiratory support due to its contribution to anatomical dead space. Ultra-short echo time (UTE) MRI enables non-invasive assessment of the large airways; we quantified tracheal volume in infants with BPD and examined its relationship to disease severity and ventilator history. Methods 176 infants (93 female) with Jensen BPD grade 1 (n = 16), grade 2 (n = 73), and grade 3 (n = 87), as well as respiratory-control term infants (n = 5) underwent UTE MRI on a 1.5T scanner at 40±3 weeks post-menstrual age. 71 patients were invasively ventilated at MRI, 47 received positive pressure invasive ventilation (2 liters per minute of oxygen), while 58 were ventilated without positive pressure or breathed room air. Tracheal volumes were quantified from 3D images reconstructed at end-expiration and end-inspiration 1, averaged between respiration states, and normalized to body-surface-area. Modified Ochiai lung structure scores were also determined. Correlation to total intubation days prior to MRI, as well as ventilator settings at MRI were performed using Spearman rank coefficient and inter-group differences using a Kruskal-Wallis test with post-hoc Dunn’s test. Results Normalized tracheal volume correlated with days on invasive ventilation before MRI in ventilated (r = 0.33, p = 0.005), and non-ventilated infants without positive pressure (r = 0.41, p = 0.001) but not in those receiving positive pressure (r=-0.06, p = 0.689). (Figure 1a). Among those non-ventilated without positive pressure, tracheal volumes were significantly elevated in grade 2 (median IQR = 3.1 2.7,3.6 mL/m2) and grade 3 patients (3.0 2.5,3.7 mL/m2) compared with controls (2.0 1.7,2.2 mL/m2) (p = 0.004 and p = 0.025, respectively) (Figure 1b). Tracheal volumes correlated with Ochiai score in the ventilated patients (r = 0.30, p = 0.013), but not the non-ventilated patients (both p 0.394). In ventilated patients, tracheal volume correlated with FiO2 (r = 0.38, p = 0.001) and tidal volume per bodyweight (r = 0.37, p 0.0001), but not mean airway pressure (r = 0.14, p = 0.262) at MRI. Conclusions Tracheal size was significantly correlated with total intubation time considering the entire cohort. In ventilated patients, tracheal volumes were associated with parenchymal disease and oxygen requirements, likely reflecting overall disease severity and the respiratory support needed to maintain adequate oxygenation. These findings indicate combined iatrogenic effects on airway development from intubation and mechanical ventilation, along with the contribution of disease severity. Overall, MRI-based tracheal measurements may help assess increased dead space and guide ventilator management to mitigate tracheal enlargement. References 1Bates AJ et al. JMRI. 2019. This abstract is funded by: NIH HL146689
Munidasa et al. (Fri,) studied this question.