Table 3.
Main characteristics of ROC curves built for hemodynamic variables of interest
| AUC | Best thresholds (%) | Compared with AUC = 0.5 | Specificity | Sensitivity | PPV | NPV | LR+ | LR− | Youden index | |
|---|---|---|---|---|---|---|---|---|---|---|
| varEtCO2500 | 0.82 [0.67; 0.97] | 5.89 | 0.006 | 1 [1.0; 1.0] | 0.6 [0.33 0.87] | 1.0 | 0.81 | Infinite | 0.4 | 0.6 |
| varEtCO2100 | 0.74 [0.60; 0.89] | 3.0 | 0.034 | 1 [1.0; 1.0] | 0.33 [0.13; 0.6] | 1.0 | 0.68 | Infinite | 0.67 | 0.33 |
| varMAP500 | 0.62 [0.44; 0.83] | 0.31 | ||||||||
| varPP500 | 0.62 [0.43; 0.82] | 0.31 | ||||||||
| varHR500 | 0.65 [0.44; 0.82] | 0.29 | ||||||||
| FtC | 0.49 [0.37; 0.67] | 0.09 |
Mean arterial pressure (MAP), pulse pressure (PP), heart rate (HR), corrected systolic flow time (Ftc), and end-tidal carbon dioxide (EtCO2). Data are expressed as variation from baseline (i.e., ΔEtCO2 for EtCO2 variation) after different volumes of infused fluid: 100 and 500 ml. Area under the curve (AUC), positive predictive value (PPV), negative predictive value (NPV), positive likelihood ratio (LR+), negative likelihood ratio (LR−)