Abstract
BACKGROUND: While positive pressure ventilation has been considered an important contributing factor associated with pulmonary air leaks, studies examining the association between specific ventilatory settings during acute-phase high-frequency oscillatory ventilation (HFOV) and pulmonary air leaks among extremely preterm infants are limited. METHODS: This was a single-center retrospective cohort study conducted at an institution that primarily used HFOV after intubation in extremely preterm infants. We analyzed data from extremely preterm infants born between 2010 and 2021. The primary outcome was pulmonary air leakage during the first 7 days of life. The exposure variable was the maximum mean airway pressure (MAP) on HFOV during the first 7 days of life or before the onset of pulmonary air leaks. Maximum MAP was categorized into three groups: low (7-10 cmH(2)O), moderate (11-12 cmH(2)O), and high (13-15 cmH(2)O) MAP categories. We conducted robust Poisson regression analyses after adjustment for perinatal confounders, using the low MAP category as the reference. RESULTS: The cohort included 171 infants (low MAP, 123; moderate MAP, 27; and high MAP, 21). The median (interquartile range) gestational age and birth weight were 25.7 (24.3-26.7), 25.7 (24.9-26.9), and 25.3 (24.3-26.6) weeks and 760 (612-878), 756 (648-962), and 734 (578-922) g for infants in the low, moderate, and high MAP categories, respectively. Compared to infants in the low MAP category, those in the high MAP category had a higher incidence of pulmonary air leaks (4.1% vs. 33.3%; adjusted risk ratio, 5.4; 95% confidence interval, 1.6-18.5). In contrast, there was no clear difference in the risk of pulmonary air leaks between the moderate and low MAP categories (3.7% vs. 4.1%; adjusted risk ratio, 0.9; 95% confidence interval, 0.1-6.1). CONCLUSION: Extremely preterm infants requiring high MAP (≥13 cmH(2)O) in acute-phase HFOV had a higher risk of pulmonary air leak during the first 7 days of life.