High-flow nasal cannula (HFNC) therapy is commonly used to prevent reintubation after planned extubation. In clinical practice, there are no appropriate tools to evaluate whether HFNC therapy was successful or failed after planned extubation. In this retrospective observational study, we investigated whether the use of the ROX index was appropriate to differentiate between HFNC success and failure within 72 h after extubation and to develop an integrated model including the ROX index to improve the prediction of HFNC success in patients receiving HFNC therapy after planned extubation. Of 276 patients, 50 patients (18.1%) were reintubated within 72 h of extubation. ROX index values of >8.7 at 2 h, >8.7 at 6 h, and >10.4 at 12 h after HFNC therapy were all meaningful predictors of HFNC success in extubated patients. In addition, the integrated model including the ROX index had a better predictive capability for HFNC success than the ROX index alone. In conclusion, the ROX index at 2, 6, and 12 h could be applied to extubated patients to predict HFNC success after planned extubation. To improve its predictive power, we should also consider an integrated model consisting of the ROX index, sex, body mass index, and the total duration of ventilator care.

Background : In elderly patients admitted to an intensive care unit (ICU), delirium is a common disease and has a poor prognosis. However, early detection of delirium is difficult in elderly patients because most of them have different degrees of senile brain change and most delirium manifests as hypoactivity. To improve the prognosis of the ICU, it is important to identify risk factors for delirium and actively select high-risk patients for delirium. Aims and objectives : To Identify risk factors for delirium and develop modeling using risk factors. Method : This study was a multicenter prospective observation study performed on critically ill patients at four academic tertiary care hospitals in South Korea between June 2021 and December 2022. Results : Of the 220 patients, 70 patients (31.8%) were diagnosed with delirium. The median age of these patients was 76 years and 51(72.9%) were male. Smoking history, ventilator use, psychologic history, and hemodialysis were statistically significant independent predictors of delirium in elderly patients. When developing modeling using risk factors, Model IV including smoking history, ventilator use, psychological history, and hemodialysis was superior to other models. The performance of Model IV was fair (AUC, 0.754; 95% CI, 0.686–0.822; P<0.001). Conclusions : Risk factors for delirium in elderly patients admitted to the ICU were smoking history, ventilator use, psychological history, and hemodialysis. If ventilator and CRRT are applied to elderly smokers with a history of psychosis, active delirium screening is required to detect delirium. Additional studies involving multinational cohort are warranted.