Objectives: The aim of this study was to evaluate the prognostic performance of the peak amplitude of P25/30 cortical somatosensory evoked potentials in predicting nonawakening in targeted temperature management-treated cardiac arrest patients.
Design: Prospective analysis.
Setting: Four academic tertiary care hospitals.
Patients: Eighty-seven cardiac arrest survivors after targeted temperature management.
Interventions: Analysis of the amplitude of P25/30.
Measurements and main results: In all participants, somatosensory evoked potentials were recorded after rewarming, and bilaterally absent pupillary and corneal reflexes were evaluated at 72 hours after the return of spontaneous circulation. We analyzed the amplitudes of the N20 and P25/30 peaks and the N20-P25/30 complex in cortical somatosensory evoked potentials. Upon hospital discharge, 87 patients were dichotomized into the awakening and nonawakening groups. The lowest amplitudes of N20, P25/30, and N20-P25/30 in the awakening patients were 0.17, 0.45, and 0.73 μV, respectively, and these thresholds showed a sensitivity of 70.5% (95% CI, 54.8-83.2%), 86.4% (95% CI, 72.7-94.8%), and 75.0% (95% CI, 59.7-86.8%), respectively, for nonawakening. The area under the curve of the P25/30 amplitude was significantly higher than that of the N20 amplitude (0.955 [95% CI, 0.912-0.998] vs 0.894 [95% CI, 0.819-0.969]; p = 0.036) and was comparable with that of the N20-P25/30 amplitude (0.931 [95% CI, 0.873-0.989]). Additionally, adding resuscitation variables or an absent brainstem reflex to the P25/30 amplitude showed a trend toward improving prognostic performance compared with the use of other somatosensory evoked potential amplitudes (area under the curve, 0.958; 95% CI, 0.917-0.999 and area under the curve, 0.974; 95% CI, 0.914-0.996, respectively).
Conclusions: Our results provide evidence that the absence of the P25/30 peak and a reduction in the P25/30 amplitude may be considered prognostic indicators in these patients.