(679) Noninvasive Ventilation or CPAP in High-Risk Small Infants With Respiratory Distress in the PICU

Introduction: Respiratory distress in infants is a common reason for admission to the pediatric ICU. High-flow nasal cannula (HFNC) is first-line treatment at our institution, but some infants require noninvasive ventilation (NIV) or CPAP. This study aims to report our experience using NIV/CPAP in infants < 10 kg.

Methods: We conducted a review of infants < 10 kg treated with NIV/CPAP July 1, 2017 and May 31, 2021 in our PICU and pediatric cardiac ICU, excluding those who received these modalities for post-extubation support. Demographic, medical history, support type and settings, vital signs, pulse oximetry, gas exchange, and reintubation data were extracted from the electronic health record. We compared successfully treated NIV/CPAP patients with intubated subjects using Chi-squared and Mann-Whitney tests for categorical and continuous data, respectively.

Results: We studied 62 patients with median age 96 (6.5-308) days and weight 4.5 (3.4-6.6) kg. Of these, 22 (35%) required intubation. There were no significant differences in demographics, medical history, cardiac surgery need, primary interface, pre-NIV/CPAP support, device used to deliver NIV/CPAP, aerosolized medications delivered, and blood gases before initiation. We found that 57 (92%) received HFNC before NIV/CPAP. Intubated subjects were less likely to have bronchiolitis (27% vs 60%, P=0.04), less likely to be discharged from the hospital to home (68% vs 93%, P=0.02), had a longer median hospital LOS (26.9 [21.0-50.5] vs. 10.4 [5.6-28.4] days, P=0.002), longer median ICU LOS (14.6 [7.9-25.2] vs. 5.8 [3.8-12.4] days, P=0.004). Initial vital signs and FiO2 were similar, but SpO2 was lower and FiO2 higher at 6 and 12 hours after support initiation for patients who required intubation. Initial NIV settings were similar, but subjects who failed NIC/CPAP had higher maximum and final inspiratory/expiratory pressure.

Conclusions: Most infants who failed initial HFNC support were successfully managed without intubation using NIV or CPAP. Bronchiolitis was associated with a lower rate of intubation, whereas lower SpO2 and higher FiO2 levels were associated with higher rates of intubation.