Search Results (2)

Background and Objectives: Extubation success in ICU patients is crucial for reducing ventilator-associated complications, morbidity, and mortality. The Rapid Shallow Breathing Index (RSBI) is a widely used predictor for weaning from mechanical ventilation. This study aims to determine the predictive value of serial RSBI measurements on extubation success in ICU patients on mechanical ventilation. Materials and Methods: This prospective observational study was conducted on 86 ICU patients at Hitit University between February 2024 and July 2024. Patients were divided into successful and unsuccessful extubation groups. RSBI values were compared between these groups. Results: This study included 86 patients (32 females, 54 males) with a mean age of 54.51 ± 12.1 years. Extubation was successful in 53 patients and unsuccessful in 33. There was no significant difference in age and intubation duration between the groups (p = 0.246, p = 0.210). Significant differences were found in RSBI-1a and RSBI-2 values (p = 0.013, p = 0.011). The median RSBI-2a was 80 in the successful group and 92 in the unsuccessful group (p = 0.001). The ΔRSBI was higher in the unsuccessful group (p = 0.022). ROC analysis identified optimal cut-off values: RSBI-2a ≤ 72 (AUC 0.715) and ΔRSBI ≤ −3 (AUC 0.648). RSBI-2a ≤ 72 increased the likelihood of successful extubation by 10.8 times, while ΔRSBI ≤ −3 increased it by 3.4 times. Using both criteria together increased the likelihood by 28.48 times. Conclusions: Serial RSBI measurement can be an effective tool for predicting extubation success in patients on IMV. These findings suggest that serially measured RSBI may serve as a potential indicator for extubation readiness. Full article

The decision to extubate patients on invasive mechanical ventilation is critical; however, clinician performance in identifying patients to liberate from the ventilator is poor. Machine Learning-based predictors using tabular data have been developed; however, these fail to capture the wide spectrum of data available. Here, we develop and validate a deep learning-based model using routinely collected chest X-rays to predict the outcome of attempted extubation. We included 2288 serial patients admitted to the Medical ICU at an urban academic medical center, who underwent invasive mechanical ventilation, with at least one intubated CXR, and a documented extubation attempt. The last CXR before extubation for each patient was taken and split 79/21 for training/testing sets, then transfer learning with k-fold cross-validation was used on a pre-trained ResNet50 deep learning architecture. The top three models were ensembled to form a final classifier. The Grad-CAM technique was used to visualize image regions driving predictions. The model achieved an AUC of 0.66, AUPRC of 0.94, sensitivity of 0.62, and specificity of 0.60. The model performance was improved compared to the Rapid Shallow Breathing Index (AUC 0.61) and the only identified previous study in this domain (AUC 0.55), but significant room for improvement and experimentation remains. Full article