Initiation of Oral Feeding in Pediatric Patients While on High Flow Nasal Cannula Support: A Retrospective Analysis

Background: High-flow nasal cannula (HFNC) is an efficient, open circuit, noninvasive ventilatory support used to manage acute respiratory failure (ARF) in disease processes like bronchiolitis, status asthmaticus and pneumonia. Enteral feeding can be established in HFNC, with studies validating its safety, but there are no practice-based guidelines for the settings of HFNC at which oral feeding can be initiated in the pediatric population. Our objective is to determine the HFNC settings at which per os (PO) feeding was initiated successfully without adverse events in patients admitted with ARF to our pediatric intensive care unit (PICU).

Methods: Retrospective chart review was done for patients admitted to PICU in ARF aged 1 day to 21 years who received HFNC support from January 2014 to December 2019. Subjects on mechanical ventilation, had tube feedings or were never fed while on HFNC or had nil per os (NPO) orders while on HFNC were excluded. Included subjects were divided into the following age groups: 1 day to ≤ 6 months, > 6 months to ≤ 2 years, > 2 to ≤ 5 years, > 5 to ≤ 12 years and > 12 years to ≤ 21 years. Median HFNC setting at which PO feeding was successfully commenced was determined by univariate analysis. Chart review confirmed that subjects tolerated PO feeds without any adverse events at the corresponding HFNC setting. A Kruskal-Wallis test was used for the assessment of any significant difference for the HFNC settings at which feeding was commenced among the different age groups. Statistical significance was set at p<0.05. All analyses were performed using SAS v9.4 (SAS Institute Inc., Cary, NC).

Results: A total of 621 charts were reviewed. 130 patients met the exclusion criteria. 152 (31%) of 491 included subjects constituted age groups 6 months to ≤ 2 years and 147(30%) constituted >2 to ≤ 5 years (Figure 1). Male: Female ratio was 1.3:1. The median HFNC setting at which PO feedings were successfully initiated for different age groups were 5 L/min for ≤ 6 months (IQR 4-6); 6 L/min for > 6 months to ≤ 2yrs (IQR 5-8); 8 L/min for > 2 years to ≤ 5 years (IQR 6-8) and for >5 years to ≤ 12 years (IQR 6-10); and 10 L/min for >12 years (IQR 8-10) (Table 1). There is a statistically significant difference in the median HFNC settings at which PO feeding was successfully initiated, with older age groups commencing feeds at higher median HFNC settings (p < 0.0001).

Conclusion: This study determined empiric age-based HFNC settings at which PO feeding was successfully initiated without any adverse effects in pediatric age group. These findings can be used to guide standardized practice for the initiation of PO feeding in the pediatric critical care setting.