Predictors of success of high flow nasal cannula (HFNC) therapy in children with respiratory distress in pediatric intensive care unit

• Main Authors: A V Lalitha, T K Anjankumar, S Sumithra
• Format: Article
• Language: English
• Published: Wolters Kluwer Medknow Publications 2018-01-01
• Series: Journal of Pediatric Critical Care
• Subjects: hfnc; respiratory distress
• Online Access: http://www.jpcc.org.in/article.asp?issn=2349-6592;year=2018;volume=5;issue=1;spage=22;epage=28;aulast=Lalitha

Objective: To identify clinical and laboratory variables that predict success of High Flow Nasal Canula (HFNC) therapy in children presenting to Pediatric Intensive Care Unit (PICU) with respiratory distress of varied etiology. Methods: This was a prospective observational study conducted in PICU of St John’s medical college, Bengaluru over a period of one year from May 2016 to April 2017. All children 1 month to 2 years presenting with moderate/severe respiratory distress of varied etiology were included for the study. The standard exclusion criteria of noninvasive ventilation were used. They were started on HFNC and clinical and laboratory variables were monitored at regular intervals. The participants were categorized as responders and non-responders (defined by need for escalation of respiratory support to invasive ventilation while on HFNC therapy). Results: One hundred fourteen childrens participated in the study, 99 of them responded and 15 were non-responders. Fifty nine had pneumonia(51.7%), 14 had WALRI (12.2%),13 had bronchiolitis (11.4%), 12 had post extubation respiratory distress(10.5%), 8 had fluid overload with respiratory distress(7%), 4 had upper air way obstruction (1.7% ), 4 had respiratory distress due to other causes. No differences were noted between these two groups for age, sex, weight, comorbid conditions. Decreasing heart rate, decrease in the mean respiratory rate and improving distress score at the end of 1 hour and 6 hours strongly correlated with response to therapy. The PRISM III scores were significantly higher (p =0.004) in non-responders. Respiratory distress due to pneumonia as underlying etiology was another predictor of failure of HFNC (p=0.009). No differences were noted in arterial blood gas parameters between these two groups, however paO2 among responders was significantly higher compared to non-responders at the initiation of HFNC and at 1 hour. Conclusion: Improving trends of clinical vital signs like heart rate, respiratory rate and RDS while on HFNC therapy are the key to success. Majority of children who failed HFNC had pneumonia as the underlying etiology of respiratory distress.