Risk Factors for Pneumothorax and Its Association with Ventilation in Neonates

The mode of ventilation that is implicated in pneumothorax is the one at the time of its diagnosis. Although there is evidence that air leak starts many hours before it is clinically evident, there are no prior studies that have investigated the association of pneumothorax with the mode of ventilation few hours before rather than at the time of its diagnosis.

A retrospective case-control study was conducted in the neonatal intensive care unit (NICU) between 2006 and 2016 where cases of neonates with pneumothorax were compared with gestational age-matched control neonates without pneumothorax. Respiratory support associated with pneumothorax was classified as the mode of ventilation 6 hours before the clinical diagnosis of pneumothorax. We investigated the factors that were different between cases and controls, and between cases of pneumothorax on bubble continuous positive airway pressure (bCPAP) and invasive mechanical ventilation (IMV).

Of the 8,029 neonates admitted in the NICU during the study period, 223 (2.8%) developed pneumothorax. Among these, 127 occurred among 2,980 (4.3%) neonates on bCPAP, 38 among 809 (4.7%) neonates on IMV, and the remaining 58 among 4,240 (1.3%) neonates on room air. Those with pneumothorax were more likely to be male, have higher body weight, require respiratory support and surfactant administration, and have bronchopulmonary dysplasia (BPD). Among those who developed pneumothorax, there were differences in the gestational age, gender, and use of antenatal steroids between those who were on bCPAP as compared to those on IMV. IMV was associated with increased odds of pneumothorax as compared to those on bCPAP in a multivariable regression analysis. Cases on IMV had higher incidence of intraventricular hemorrhage, retinopathy of prematurity, BPD, and necrotizing enterocolitis, as well as longer length of stay as compared to those on bCPAP.

Neonates who require any respiratory support have higher incidence of pneumothorax. Among those on respiratory support, those on IMV had higher odds of pneumothorax and worse clinical outcomes as compared to those on bCPAP.

· The process of air leak leading to pneumothorax in majority of neonates starts much before it is clinically diagnosed.. · It is possible to detect the air leak early in the process by subtle changes in the signs, symptoms and changes in lung function.. · True association of the ventilation associated with pneumothorax is not at the time of diagnosis of pneumothorax but few hours before it is diagnosed.. · There is higher incidence of pneumothorax in neonates on any respiratory support.. · There is significantly higher incidence of pneumothorax among neonates on invasive ventilations as compared to noninvasive ventilation after correction for all other clinical factors..