Table 2.
Predictive values of (combinations of) EEG and SSEP parameters.
Predictor | Predicted outcome |
Specificity (95% CI) |
Sensitivity (95% CI) |
PPV (95% CI) |
NPV (95% CI) |
---|---|---|---|---|---|
Continuous EEG pattern at 12 h | Good | 98% (92–100) |
19% (8–37) |
75% (36–96) |
80% (72–86) |
N20 > 3.6 μV* | Good | 96% (90–99) |
31% (17–50) |
71% (42–90) |
82% (74–88) |
Absent SSEP N20* | Poor | 100% (87–100) |
39% (30–49) |
100% (89–100) |
33% (24–44) |
0<N20<0.4 μV * | Poor | 100% (87–100) |
11% (6–20) |
100% (70–100) |
25% (18–34) |
Suppressed or synchronous EEG on suppressed background at 12 h | Poor | 100% (87–100) |
37% (28–47) |
100% (89–100) |
32% (24–43) |
Suppressed or synchronous EEG on suppressed background at 24 h | Poor | 100% (87–100) |
30% (22–40) |
100% (87–100) |
30% (22–40) |
Suppressed or synchronous EEG on suppressed background at 12 h or N20 < 0.4 μV | Poor | 100% (87–100) |
58% (48–67) |
100% (93–100) |
42% (31–54) |
Suppressed or synchronous EEG on suppressed background at 24 h or N20 < 0.4μV | Poor | 100% (87–100) |
53% (43–63) |
100% (92–100) |
39% (29–51) |
CI, confidence interval; SSEP, somatosensory evoked potential.
Timing SSEP generally between 48 and 72 h, never within 24 h after cardiac arrest.