Abstract Body: Background:

Cardiac arrest (CA) can cause coma in resuscitated patients. Accuracy of prognosis is limited. Presence of brain signals such as somatosensory evoked potential (SSEP) reveals integrity of the thalamocortical pathway, which is necessary for arousal but insufficient to predict positive outcome.

The N10 is a SSEP component that diminishes in CA. The N7, which can be obscured by stimulus artifact or other components, is associated with subcortical structures. Phase space area (PSA) quantifies SSEP morphology and is associated with CA recovery. It has been measured at least 5ms after stimulus.

Hypothesis: Early SSEP and PSA diminish under anesthesia and during CA but may recover afterwards.

Aims: By removing stimulus artifact, we capture early SSEP (N7 and early PSA) to enable improved prognostication.

Methods:

Six rats received 1.2, 1.8, and 2.5% isoflurane anesthesia for 30min. Two underwent 5 or 7min CA and were observed over 2hr. SSEP was recorded every 2s.

An adaptive filter removed the artifact. Peaks were measured near 7 (N7) and 15 (N10) ms post-stimulus. PSA, the area occupied by the phase space curve, was recorded over 0-9ms (“early”) and 0-30ms (“total”). Signal-to-noise ratio (SNR) is the ratio of N10 amplitude to standard deviation of the signal. T tests were used for comparisons between SNR values and between experimental conditions.

Results:

We identified 4153 N7 peaks with latency (mean±95% C.I.) 8.6±1.3ms in the filtered data, compared to 2464 with latency 8.5±1.4ms in the unfiltered data. SNR was greater in the filtered signal (2.3±1.5) than unfiltered (0.3±0.4).

N7 amplitude decreased at the highest anesthesia level; N10 decreased for each level. N7 and N10 decreased during CA and recovered after, although N10 remained below baseline during 2hr recovery. Early PSA was lower under the higher anesthesia levels, while total PSA decreased for each level. PSA decreased during CA and recovered after 5min CA; total PSA recovered to baseline and early PSA exceeded it after 2hr.Early PSA did not recover from 7min CA, while total PSA recovered but remained below baseline.

Conclusion: We demonstrate a novel technique to discover an elusive early-latency response, adding detail to a neural pathway involved in arousal. We reduce noise, capture N7 consistently, and measure PSA beginning at 0ms, providing additional diagnostic dimensions responsive to anesthesia and CA. Future study will investigate outcomes in resuscitated patients.