Search

Assessing the integrity of neural functions in coma after cardiac arrest remains an open challenge. Prognostication of coma outcome relies mainly on visual expert scoring of physiological signals, which is prone to subjectivity and leaves a considerable number of patients in a ‘grey zone’, with uncertain prognosis. Quantitative analysis of EEG responses to auditory stimuli can provide a window into neural functions in coma and information about patients’ chances of awakening. However, responses to standardized auditory stimulation are far from being used in a clinical routine due to heterogeneous and cumbersome protocols. Here, we hypothesize that convolutional neural networks can assist in extracting interpretable patterns of EEG responses to auditory stimuli during the first day of coma that are predictive of patients’ chances of awakening and survival at 3 months. We used convolutional neural networks (CNNs) to model single-trial EEG responses to auditory stimuli in the first day of coma, under standardized sedation and targeted temperature management, in a multicentre and multiprotocol patient cohort and predict outcome at 3 months. The use of CNNs resulted in a positive predictive power for predicting awakening of 0.83 ± 0.04 and 0.81 ± 0.06 and an area under the curve in predicting outcome of 0.69 ± 0.05 and 0.70 ± 0.05, for patients undergoing therapeutic hypothermia and normothermia, respectively. These results also persisted in a subset of patients that were in a clinical ‘grey zone’. The network’s confidence in predicting outcome was based on interpretable features: it strongly correlated to the neural synchrony and complexity of EEG responses and was modulated by independent clinical evaluations, such as the EEG reactivity, background burst-suppression or motor responses. Our results highlight the strong potential of interpretable deep learning algorithms in combination with auditory stimulation to improve prognostication of coma outcome.

Early prognostication of long-term outcome of comatose patients after cardiac arrest remains challenging. Electroencephalography-based power spectra after cardiac arrest have been shown to help with the identification of patients with favourable outcome during the first day of coma. Here, we aim at comparing the power spectra prognostic value during the first and second day after coma onset following cardiac arrest and to investigate the impact of sedation on prognostication. In this cohort observational study, we included comatose patients (N = 91) after cardiac arrest for whom resting-state electroencephalography was collected on the first and second day after cardiac arrest in four Swiss hospitals. We evaluated whether the average power spectra values at 4.6-15.2 Hz were predictive of patients' outcome based on the best cerebral performance category score at 3 months, with scores ranging from 1 to 5 and dichotomized as favourable (1-2) and unfavourable (3-5). We assessed the effect of sedation and its interaction with the electroencephalography-based power spectra on patient outcome prediction through a generalized linear mixed model. Power spectra values provided 100% positive predictive value (95% confidence intervals: 0.81-1.00) on the first day of coma, with correctly predicted 18 out of 45 favourable outcome patients. On the second day, power spectra values were not predictive of patients' outcome (positive predictive value: 0.46, 95% confidence intervals: 0.19-0.75). On the first day, we did not find evidence of any significant contribution of sedative infusion rates to the patient outcome prediction (P > 0.05). Comatose patients' outcome prediction based on electroencephalographic power spectra is higher on the first compared with the second day after cardiac arrest. Sedation does not appear to impact patient outcome prediction.

BACKGROUND Electroencephalography (EEG) is essential to assess prognosis in patients after cardiac arrest (CA). Use of continuous EEG (cEEG) is increasing in critically-ill patients, but it is more resource-consuming than routine EEG (rEEG). Observational studies did not show a major impact of cEEG versus rEEG on outcome, but randomized studies are lacking. METHODS We analyzed data of the CERTA trial (NCT03129438), including comatose adults after CA undergoing cEEG (30-48 hours) or two rEEG (20-30 minutes each). We explored correlations between recording EEG type and mortality (primary outcome), or Cerebral Performance Categories (CPC, secondary outcome), assessed blindly at 6 months, using uni- and multivariable analyses (adjusting for other prognostic variables showing some imbalance across groups). RESULTS We analyzed 112 adults (52 underwent rEEG, 60 cEEG,); 31 (27.7%) were women; 68 (60.7%) patients died. In univariate analysis, mortality (rEEG 59%, cEEG 65%, p=0.318) and good outcome (CPC 1-2; rEEG 33%, cEEG 27%, p=0.247) were comparable across EEG groups. This did not change after multiple logistic regressions, adjusting for shockable rhythm, time to return of spontaneous circulation, serum neuron-specific enolase, EEG background reactivity, regarding mortality (rEEG vs cEEG: OR 1.60, 95% CI 0.43 - 5.83, p=0.477), and good outcome (OR 0.51, 95% CI 0.14 - 1.90, p=0.318). CONCLUSION This analysis suggests that cEEG or repeated rEEG are related to comparable outcomes of comatose patients after CA. Pending a prospective, large randomized trial, this finding does not support the routine use of cEEG for prognostication in this setting. Trial registration Continuous EEG Randomized Trial in Adults (CERTA); NCT03129438; July 25, 2019.

Background and ObjectivesEEG is widely used for prediction of neurologic outcome after cardiac arrest. To better understand the relationship between EEG and neuronal injury, we explored the association between EEG and neurofilament light (NfL) as a marker of neuroaxonal injury, evaluated whether highly malignant EEG patterns are reflected by high NfL levels, and explored the association of EEG backgrounds and EEG discharges with NfL.MethodsWe performed a post hoc analysis of the Target Temperature Management After Out-of-Hospital Cardiac Arrest trial. Routine EEGs were prospectively performed after the temperature intervention ≥36 hours postarrest. Patients who awoke or died prior to 36 hours postarrest were excluded. EEG experts blinded to clinical information classified EEG background, amount of discharges, and highly malignant EEG patterns according to the standardized American Clinical Neurophysiology Society terminology. Prospectively collected serum samples were analyzed for NfL after trial completion. The highest available concentration at 48 or 72 hours postarrest was used.ResultsA total of 262/939 patients with EEG and NfL data were included. Patients with highly malignant EEG patterns had 2.9 times higher NfL levels than patients with malignant patterns and NfL levels were 13 times higher in patients with malignant patterns than those with benign patterns (95% CI 1.4-6.1 and 6.5-26.2, respectively; effect size 0.47; p < 0.001). Both background and the amount of discharges were independently strongly associated with NfL levels (p < 0.001). The EEG background had a stronger association with NfL levels than EEG discharges (R2 = 0.30 and R2 = 0.10, respectively). NfL levels in patients with a continuous background were lower than for any other background (95% CI for discontinuous, burst-suppression, and suppression, respectively: 2.26-18.06, 3.91-41.71, and 5.74-41.74; effect size 0.30; p < 0.001 for all). NfL levels did not differ between suppression and burst suppression. Superimposed discharges were only associated with higher NfL levels if the EEG background was continuous.DiscussionBenign, malignant, and highly malignant EEG patterns reflect the extent of brain injury as measured by NfL in serum. The extent of brain injury is more strongly related to the EEG background than superimposed discharges. Combining EEG and NfL may be useful to better identify patients misclassified by single methods.