# An exploratory analysis of ACNS electroencephalography patterns in 34 comatose patients after in-hospital cardiopulmonary resuscitation

## TL;DR

This study explores how EEG patterns can predict neurological outcomes in patients who are comatose after in-hospital CPR.

## Contribution

The study applies the 2021 ACNS EEG terminology to evaluate its prognostic value in a small cohort of CPR patients.

## Key findings

- Increased slow wave and theta-dominant EEG patterns were linked to better neurological outcomes.
- Beta-dominant EEG patterns were associated with poor prognosis.
- EEG categorization showed high specificity and sensitivity in predicting patient outcomes.

## Abstract

This study aims to evaluate the prognostic value of electroencephalogram (EEG) patterns in comatose patients after in-hospital adult cardiopulmonary resuscitation (CPR).

Clinical and EEG data were retrospectively collected from 34 patients who underwent in-hospital CPR. The EEG data were classified into seven patterns according to the terminology defined by the 2021 American Clinical Neurophysiology Society (ACNS). All patients were further categorized into distinct groups based on their EEG characteristics, including background frequency and EEG categorization. The outcome of patients at discharge was assessed using the Glasgow Outcome Score (GOS) as 1–2 (poor) or 3–5 (good). Baseline characteristics and EEG patterns were compared between the outcome groups. Receiver operating characteristic (ROC) analysis was used to evaluate the predictive performance of the EEG background frequency and EEG categorization. Differences in the area under the curve (AUC) were compared using the DeLong test.

This study included 34 patients who underwent in-hospital CPR. Patients with poor outcomes had lower GCS scores [3.0 (interquartile range, 3.0–4.0) vs. 6.0 (interquartile range, 3.0–7.0), p = 0.006]. Statistically significant differences were identified in the duration of CPR (p = 0.030) and time to establish return of spontaneous circulation (ROSC) (p = 0.026). Our exploratory analyses indicated that increased slow wave pattern and theta-dominant background were associated with good neurological outcomes (p < 0.001 and p = 0.007, respectively), while a potential association was observed between beta-dominant background and poor prognosis (p < 0.001). For EEG categorization, the results revealed that Group (I) was more common among good-outcome patients and Group (III) was associated with an increased likelihood of clinical deterioration at discharge (p < 0.001 and p = 0.003, respectively). The presence of the EEG background frequency yielded an AUC of 0.889 (95% CI: 0.734–0.971, sensitivity 69.6%, specificity 99.9%), while EEG categorization yielded an AUC of 0.913 (95% CI: 0.765–0.982, sensitivity 82.6%, specificity 99.9%), with no significant AUC difference between the two indicators.

The 2021 version of the ACNS standardized terminology to analyze EEG patterns is useful for predicting the prognosis of comatose patients following CPR. Our preliminary findings suggest a potential association between EEG patterns and neurological outcomes, although this finding requires further validation in larger prospective studies.