Abstracts

Rationale: Current guidelines after cardiopulmonary resuscitation (CPR) follow binary systems for classification of the median nerve somatosensory evoked potentials (SSEPs). The goal of this study to investigate other quantifiable markers.  Methods: Retrospective analysis of SSEPs recording at Yale following cardiopulmonary arrest.  Evoked potentials were recorded on day 3 or after the arrest or cooling whichever is later per ACNS guidelines using a Dantec™KEYPOINT® 6 channel system with a sampling frequency at 48 KHz per channel.  Cerebral performance category (CPC) was determined upon intensive care unit discharge. CPC 1-3 was defined as good, CPC 4-5 as poor outcome. We evaluated the performance of amplitudes, amplitudes ratio, and other morphological features as classifiers for outcomes. Referral for evoked potentials was typically sought in cases with equivocal findings by other modalities or to help guide discussions with families.   Results: Total number included N = 84. 86% of patients achieved poor outcomes, of whom 14% achieved CPC 4. The N20 amplitude and N20/N18 amplitude ratio in survivors were>0.66 and >1.2 respectively. In all, late responses were present, lowest amplitude > 0.4. All 55 patients (22 in which ratio calculation possible) with lower amplitude values achieved poor outcome. The highest amplitude, and ratio values in patients with sustained CPC 4 outcomes were 2.4 and 2.6 respectively. In 3 of 4 patients of this group with N20 > 0.66 - N20/N18 amplitude ratio was =< 1.2. 13 patients had detectable N20 > 0.66 and CPC outcome 5, in 8 no evidence of severe injury and care withdrawn factoring other variables. In the 5 with evidence of injury N20/N18 ratio was < 1.2 in 4 (80%). Of 13 patients with poor outcome and late repossess > 0.4 on both sides, only 2 had evidence of significant anoxic brain injury. In both N18/N20 was =< 1.2. Though a trend observed with other morphological markers evaluated, none achieved statistical significance. Conclusions: This study confirms that there is additional prognosticating information that could be inferred from the results of SSEPs besides the commonly employed binary paradigms. N20/N18 amplitude ratio may provide additional information in subset of cases with equivocal findings and conceptually may be more immune to effect of ICU sedation than amplitudes.  Funding: This publication was made possible by funding support from grant 412064 from the American Epilepsy Society and Clinical and Translational Science Award grant UL2 TR000142 from the National Center for Advancing Translational Science. The authors wish to acknowledge the support of the Epilepsy Foundation, TUBITAK, and the Swebilius Trust.