Evolution of Mechanical Ventilation Practices in Neonatal Cardiac Patients: Single-Center Retrospective Analysis of Three 1-Year Epochs During 2000-2020.

Abstract

Objectives: We sought to better understand how mechanical ventilation (MV) practices have evolved in neonates after cardiac surgery and evaluate whether such changes were associated with outcomes.

Design: Single-center, retrospective study of three different 1 year-long periods: 2001, 2011, and 2020.

Setting: Quaternary institution with a dedicated cardiac PICU in Toronto, ON, Canada.

Patients: Neonates (≤ 28 d) receiving MV after cardiac surgery.

Interventions: None.

Measurements and main results: Settings, modes, duration of MV, noninvasive ventilation (NIV), PICU, and hospital stay were compared across epochs. A competing risk analysis was performed to determine whether changes in MV delivery were associated with MV liberation. The study included 291 patients with 101 (35%), 96 (33%), and 94 (32%) in 2001, 2011, and 2020, respectively. We did not identify differences in baseline characteristics, diagnosis, and severity category across our epochs. We found the following practice changes by epoch. First, a decrease in tidal volume (V t ) from (mean ± sd ) 12.1 ± 3.1 in 2001 to 6-7 mL/kg in 2011 and 2020 ( p < 0.001). This practice was associated with chronological reduction in driving pressure (DP): 15.4 ± 2.5, 14.3 ± 2.2, and 13.0 ± 2.6 cm H 2 O ( p < 0.001). Second, postextubation NIV was used more frequently in 2020. Third, faster weaning was introduced by 2020 and an increase in ventilator-free days, including NIV days, was observed ( p < 0.001). Fourth, after adjusting for demographic characteristics, epoch, severity, MV pre-surgery, higher V t , and DP were associated with lower prevalence of MV liberation, with respective hazard ratios (HRs) 0.818 (95% CI, 0.699-0.957); p = 0.012 and HR 0.865 (95% CI, 0.825-0.907); p < 0.001.

Conclusions: In neonates requiring MV after cardiac surgery, we have found that practices in our center changed over three 1-year epochs, from 2001, to 2011, and to 2020. Reduction in V t and early weaning when combined were associated with more ventilator-free days. Decrease in DP and V t were both associated with sooner MV liberation.