Cánula nasal de alto flujo en lactantes: Experiencia en una unidad de paciente crítico

Abstract

Introduction: The high flow nasal cannula (HFNC) is a method of respiratory support that is increasingly being used in paediatrics due to its results and safety. Objective: To determine the efficacy of HFNC, as well as to evaluate the factors related to its failure and complications associated with its use in infants. Patients and method: An analysis was performed on the demographic, clinical, blood gas, and radiological data, as well as the complications of patients connected to a HFNC in a critical care unit between June 2012 and September 2014. A comparison was made between the patients who failed and those who responded to HFNC. A failure was considered as the need for further respiratory support during the first 48 hours of connection. The Kolmogorov Smirnov, Mann-Whitney U, chi squared and the Exact Fisher test were used, as well as correlations and a binary logistic regression model for P ≤ .05. Results: The study included 109 patients, with a median age and weight: 1 month (0.2-20 months) and 3.7 kg (2-10 kg); 95 percentile: 3.7 months and 5.7 kg, respectively. The most frequent diagnosis and radiological pattern was bronchiolitis (53.2%) and interstitial infiltration (56%). Around 70.6% responded. There was a significant difference between failure and response in the diagnosis (P = .013), radiography (P = 018), connection context (P < .0001), pCO2 (median 40.7 mmHg [15.4-67 mmHg] versus 47.3 mmHg [28.6-71.3 mmHg], P = .004) and hours on HFNC (median 60.75 hrs [5-621.5 hrs] versus 10.5 hrs [1-29 hrs], P < .0001). The OR of the PCO2 ≥ 55 mmHg for failure was 2.97 (95% CI; 1.08-8.17; P = .035). No patient died and no complications were recorded. Conclusion: The percentage success observed was similar to that published. In this sample, the failure of HFNC was only associated with an initial pCO2 ≥ 55 mmHg. On there being no complications reported as regards it use, it is considered safe, although a randomised, controlled, multicentre study is required to compare and contrast these results.