Results:
A total of 410 patients were included in this study. Demographics and clinical characteristics were not different among the groups (Table 1). The use of adjunctive therapies did not vary among the groups (supplement table 2). Intravenous magnesium sulfate use was common in our cohort (86.8%); the vast majority received intermittent dosing with 1 patient receiving a continuous infusion. The number of doses of magnesium received were significantly different between study periods on univariate analysis. On multivariable analysis as magnesium doses increased length of HFNC also increased and thus the magnesium doses likely were not the driver of the decreased HFNC duration seen (supplemental table 3). Aminophylline was used infrequently (1.7%). Terbutaline and heliox were not used at all in this cohort.
For the primary outcome measure, HFNC duration decreased from 26.8 hours to 18.2 hours during PDSA 2 and it remained stable throughout PDSA 3 and PDSA 4 (Figure 2). For the secondary outcome measures, PICU LOS decreased from 41 hours to 31.8 hours during PDSA 2 and did not change during PDSA 3 and PDSA 4 (figure 3). Hospital LOS also decreased from 86.5 hours to 68 hours after PDSA 2 and remained stable throughout PDSA 3 and 4 (Figure 4). Continuous albuterol duration remained stable at 20.5 hours throughout the study (supplement figure 5).
Data on balancing measures is found in Table 2. IMV use declined throughout the study, from 9.2% during the pre-intervention period to 2.6% during PDSA 4. Seven-day PICU and hospital readmission rates were higher during PDSA 1 only (p=0.007 and p=0.028, respectively) and afterward returned to 0-1.2% for the reminder of the study. The use of NIV and ECMO did not change significantly throughout the study.