Résumé
Objective:To compare the predictive ability of SpO 2/FiO 2(S/F) and ROX index on the failure of high-flow nasal cannula(HFNC)therapy in children with acute respiratory failure after congenital heart disease surgery, and to identify the best cut-off point. Methods:Through a case-control study, the clinical data of 371 children with acute respiratory failure after congenital heart surgery treated with HFNC admitted to Guangzhou Women and Children′s Medical Center from January 2018 to December 2021 were retrospectively analyzed.The primary outcome was the need for re-intubation within 48 h after extubation of invasive ventilation.The ability of S/F and ROX index to predict HFNC failure was compared, and the optimal cut-off point was determined based on the area under the curve (AUC) of receiver operating characteristic curve.Results:A total of 371 children were included, of whom 27 (7.3%) eventually required mechanical ventilation within 48 h. The S/F prediction accuracy was highest after 6 h of HFNC treatment(AUC=0.712, 95% CI 0.599-0.825, P=0.001), and the best cut-off point for S/F was 178 mmHg(1 mmHg=0.133 kPa)(sensitivity 74.9%, specificity 69.6%). Whereas the prediction accuracy of the ROX index was highest after 12 hours of HFNC treatment, the AUC was 0.737(95% CI 0.623-0.851, P=0.002), and the best cut-off point of the ROX index was 5.865(sensitivity 72.4% specificity 66.7%). The difference in AUC between S/F after 6 h of HFNC treatment and ROX after 12 h was not statistically significant ( P=0.444), with higher sensitivity and specificity, and earlier prediction time(6 hours) in the former. Conclusion:Children with acute respiratory failure after congenital heart surgery have a strong predictive ability of S/F after 6 h of HFNC treatment, and the risk of HFNC treatment failure is higher in children with S/F <178 mmHg.