Objective Early identifying sepsis patients who had higher risk of poor prognosis was extremely important. The aim of this study was to develop an artificial neural networks (ANN) model for early predicting clinical outcomes in sepsis. Methods This study was a retrospective design. Sepsis patients from the Medical Information Mart for Intensive Care-III (MIMIC-III) database were enrolled. A predictive model for predicting 30-day morality in sepsis was performed based on the ANN approach. Results A total of 2874 patients with sepsis were included and 30-day mortality was 29.8%. The study population was categorized into the training set (n = 1698) and validation set (n = 1176) based on the ratio of 6:4. 11 variables which showed significant differences between survivor group and nonsurvivor group in training set were selected for constructing the ANN model. In training set, the predictive performance based on the area under the receiver-operating characteristic curve (AUC) were 0.873 for ANN model, 0.720 for logistic regression, 0.629 for APACHEII score and 0.619 for SOFA score. In validation set, the AUCs of ANN, logistic regression, APAHCEII score, and SOFA score were 0.811, 0.752, 0.607, and 0.628, respectively. Conclusion An ANN model for predicting 30-day mortality in sepsis was performed. Our predictive model can be beneficial for early detection of patients with higher risk of poor prognosis.

Sequential invasive-noninvasive ventilation (NIV) improves the outcomes of patients with respiratory failure caused by acute exacerbation of chronic obstructive pulmonary disease (AECOPD); however, there is no clear consensus on the optimal timing of the switch to sequential invasive-NIV in these patients. In the present study, a potential role for the modified Glasgow Coma Scale (GCS) score to guide sequential weaning was investigated. Patients with AECOPD and respiratory failure were prospectively recruited from three study centers (Wenling Hospital Affiliated to Wenzhou Medical University, the First Affiliated Hospital of Wenzhou Medical University and Changsha Central Hospital) between January 1st 2016 and December 31st 2018. Patients were randomly assigned to group A and B, with the switching point for sequential weaning strategy in the two groups being a modified GCS score ≥13 and 10 points, respectively. Each group included 240 patients. Baseline demographic characteristics were comparable in the two groups. The duration of invasive mechanical ventilation (IMV) in group A was significantly shorter than that in group B. However, there were no significant between-group differences with respect to the incidence of re-intubation, ventilator-associated pneumonia, in-hospital mortality or the length of hospital stay. Use of a modified GCS score ≥13 as the switching point for sequential invasive-NIV may help decrease the duration of IMV in patients with AECOPD and respiratory failure.

Background: To compare the efficacy and feasibility of using a modified Glasgow coma scale (GCS) score of 13 or 15 as the criterion for switching chronic obstructive pulmonary disease (COPD) patients with respiratory failure to sequential invasive-noninvasive ventilation. Methods: COPD patients with respiratory failure who had undergone endotracheal intubation and invasive mechanical ventilation (IMV) between June 2017 and June 2020 at 4 different hospitals in China were included. A total of 296 patients were randomly divided into 2 groups. In group A, the patients were extubated and immediately placed on noninvasive ventilation (NIV) when the modified GCS score reached 13. In group B, the same was done when the modified GCS score reached 15. Results: No significant differences in the mean blood pressure, oxygenation index, arterial partial pressure of oxygen, and arterial partial pressure of carbon dioxide were seen between groups A and B before extubation and 3 hours after NIV. The re-intubation times were also similar in the 2 groups. Compared to group B, the length of hospital stay, incidence of ventilator associated pneumonia, and time of invasive ventilation were all significantly lower in group A (P = .041, .001, <.001). Conclusion: Using a modified GCS score of 13 as the criterion for switching from IMV to NIV can significantly reduce the duration of IMV, length of hospital stay, and incidence of ventilator associated pneumonia in COPD patients with respiratory failure.