Establishment and internal-external validation of a 28-day mortality prediction model for septic shock patients with left ventricular systolic dysfunction

Figures

• 

  FIGURE 1. Flowchart of study population screening.

• 

  FIGURE 2. (A) Nomogram for the 28-day mortality risk of septic shock. (B) The online web-based calculator for predicting the 28-day mortality risk of septic shock. (A,B) The nomogram incorporates the following predictors: Atrial arrhythmia, Dopamine, PaO2/FiO2, LVSD, and PH. Each predictor is assigned a score on a specific point scale. To use the nomogram, clinicians first locate the patient’s value for each predictor on the corresponding axis. Then, the points for each predictor are summed to determine the total score, which is subsequently converted into a probability of 28-day mortality of septic shock. P/F, PaO2/FiO2; LVSD, left ventricular systolic dysfunction.

• 

  FIGURE 3. Receiver operating characteristic (ROC) curves. (A) Training cohort. (B) Training cohort with 1,000 bootstrap resamples. (C) Validation cohort. The Y-axis represents sensitivity (true positive rate), and the X-axis represents 1- specificity (false positive rate). The blue stepwise curve shows the ROC curve of the prediction model. The red dashed diagonal line indicates the reference line corresponding to random discrimination (AUC = 0.5). The gray shaded area represents the ROC curves obtained from 1,000 resampling attempts.

• 

  FIGURE 4. Calibration curves for mortality prediction model. (A) Training cohort. (B) Validation cohort. The X-axis represents the nomogram-predicted probability of 28-day mortality, and the Y-axis represents the observed actual probability of 28-day mortality. The dotted line (Apparent) indicates the apparent calibration performance in the cohort, the solid line (Bias-corrected) shows the bias-corrected calibration performance after 1,000 bootstrap resamples.

• 

  FIGURE 5. Decision curve analysis in prediction. (A) Training cohort. (B) Validation cohort. The X-axis represents the threshold probability of 28-day mortality, and the Y-axis represents the net benefit of the model. The solid black line (Predictors) shows the net benefit of using the nomogram to predict 28-day mortality. The gray line (All) indicates the net benefit of assuming all patients will experience 28-day mortality, while the horizontal black line (None) represents the net benefit of assuming no patients will experience 28-day mortality.

• 

  FIGURE 6. ROC comparison of APACHE II score and predictive model. APACHE II score = Acute Physiology and Chronic Health Evaluation II. The Y-axis represents sensitivity (true positive rate), and the X-axis represents 1- specificity (false positive rate). The blue stepwise curve shows the ROC curve of the prediction model. The purple stepwise curve shows the ROC curve of the APACHE II score. The red dashed diagonal line indicates the reference line corresponding to random discrimination (AUC = 0.5).