[Efficacy of high flow nasal oxygen therapy in children with acute respiratory failure]

Objective: To investigate the efficacy of high flow nasal cannula (HFNC) in children with acute respiratory failure. Methods: A prospective study was conducted. A total of 153 patients aged from 1 to 14 years with acute respiratory failure were enrolled, who were admitted to pediatric intensive care unit (PICU) of Shanghai Children's Hospital from January 2018 to December 2019. HFNC success was defined as no need for invasive mechanical ventilation and successfully withdrawn from HFNC, while HFNC failure was defined as need for invasive mechanical ventilation. HFNC at a flow rate of 2 L/(kg·min) (maximum ≤ 60 L/min) with inhaled oxygen concentration (FiO₂) between 0.30 and 1.00 was applied to maintain percutaneous oxygen saturation (SpO₂) of 0.94-0.97. Parameters including arterial partial pressure of oxygen (PaO₂), partial pressure of carbon dioxide in artery (PaCO₂), SpO₂ and PaO₂/FiO₂ were collected before and during the application of HFNC at 1 h, 6 h, 12 h, 24 h and 48 h, as well as over 48 h after HFNC withdrawn. Comparison between the groups was performed by student t test, Mann-Whitney U test or chi-square test. The sensitivity and specificity of the above parameters in predicting HFNC success were evaluated by receiver operating characteristic (ROC) curve. Results: A total of 153 children (70 males and 83 females) were enrolled. Among them, 131 (85.6%) cases were successfully weaned off from HFNC and 22 (14.4%) failed. The duration of HFNC was 57 (38, 95) hours in the successful group, and the PaO₂/FiO₂ before HFNC application and after HFNC was withdrawn were 187 (170, 212) mmHg (1 mmHg=0.133 kPa) and 280 (262, 292) mmHg, respectively. The duration of HFNC in the failure group was 19 (9, 49) hours, and the PaO₂/FiO₂ before HFNC application and after HFNC withdrawn were 176 (171, 189) mmHg and 159 (156, 161) mmHg, respectively. The values of PaO₂/FiO₂ were significantly higher in the successful group than those in the failed group at using HFNC initially 1 h (196 (182, 211) vs. 174 (160, 178) mmHg, Z =-5.105, P<0.01), 6 h (213 (203, 220) vs. 168 (157, 170) mmHg, Z =-6.772, P<0.01), 12 h (226 (180, 261) vs. 165 (161, 170) mmHg, Z =-4.308, P<0.01), 24 h (229 (195, 259) vs. 165 (161, 170) mmHg, Z=-4.609, P<0.01) and 48 h (249 (216, 273) vs. 163 (158, 169) mmHg, Z =-4.628, P<0.01) after the HFNC application, and over 48 h after HFNC was withdrawn (277 (268, 283) vs. 157 (154, 158) mmHg, Z=-3.512, P<0.01). Moreover, the PaO₂ levels were significantly higher in the successful group than those in the failed group using HFNC initially at 1 h (73.7 (71.0, 76.7) vs. 70.0 (66.2, 71.2) mmHg, Z=-4.587, P<0.01) and 6 h (79.0 (75.0, 82.0) vs. 71.0 (62.0, 72.0) mmHg, Z=-5.954, P<0.01) after HFNC application. Also, the SpO₂ levels showed the same differences at 1 h (0.96 (0.95, 0.96) vs. 0.94 (0.92, 0.94), Z =-4.812, P<0.01) and 6 h (0.96 (0.95, 0.97) vs. 0.94(0.91, 0.95), Z=-5.024, P<0.01) after HFNC application. Forty eight hours after HFNC was withdrawn, the PaO₂ (88.0 (81.7, 95.0) vs. 63.7 (63.3, 66.0) mmHg, Z =-3.032, P<0.01) and SpO₂ (0.96 (0.94, 0.98) vs. 0.91 (0.90, 0.92), Z=-3.957, P<0.01) were also significantly higher in the successful group. Regarding the HFNC complications, there was one case with atelectasis and one with pneumothorax in the failure group. HFNC was used as sequential oxygen therapy after extubation in 79 children, successful in all. ROC curve showed that the area under curve of PaO₂/FiO₂ in predicting HFNC success was 0.990, and the optimal cut-off value was 232 mmHg with the 95%CI of 0.970-1.000 (P<0.01). Conclusions: HFNC could be used as a respiratory support strategy for children with mild to moderate respiratory failure and as a sequential oxygen therapy after extubation. The PaO₂/FiO₂ when HFNC withdrow is the optimal index to evaluate the success of HFNC application.