CONTINUOUS EEG IS SUPERIOR EEG MONITORING DURING COMA AFTER CARDIAC - PowerPoint PPT Presentation

CONTINUOUS EEG IS SUPERIOR EEG MONITORING DURING COMA AFTER CARDIAC ARREST JEANNETTE HOFMEIJER, MD PHD

EEG AFTER CARDIAC ARREST  Epilepsy detection  Outcome prediction 100% specificity for poor outcome prediction Requirements High specificity for good outcome prediction Background pattern EEG features Reactivity

POOR OUTCOME PREDICTION BACKGROUND PATTERN REACTIVITY

POOR OUTCOME PREDICTION BACKGROUND PATTERN REACTIVITY N = 1175 False positives = 0 Clear definitions IDENTICAL BURSTS- SUPPRESSION GPD-SUPPRESSION SUPPRESSION

POOR OUTCOME PREDICTION BACKGROUND PATTERN REACTIVITY N = 1175 N = 1131 False positives = 0 False positives = 45 Clear definitions Moderate definitions IDENTICAL BURSTS- SUPPRESSION GPD-SUPPRESSION SUPPRESSION

ABSENT EEG REACTIVITY HIGH FALSE POSITIVE RATE FOR POOR OUTCOME PREDICTION TP FN FP TN total <12h 20 2 12 16 50 12-24h 39 8 16 45 108 24-48h 33 12 15 25 85 >48h 12 5 1 5 23 All time points 51 19 14 65 149 Admiraal et al. Annals of Neurology (accepted)

EVOLUTION OF THE EEG FAVORABLE OR UNFAVORABLE 5h 12h 24h Sondag et al. Crit Care 2017 Hofmeijer et al. Neurology 2015 Sivaraju et al. Int Care Med 2015

THE EVIDENCE Ruijter et al. Annals of Neurology (accepted)

WHY CONTINUOUS EEG?  Identify more patients with poor or good outcome  Greater reliability of EEG conclusions  Detection of epilepsy  Future of brain monitoring

WHY CONTINUOUS EEG?  Identify more patients with poor or good outcome  Greater reliability of EEG conclusions  Detection of epilepsy  Future of brain monitoring

RESULTS PREDICTIVE VALUES FOR POOR OUTCOME Ruijter et al. Ann Neurol (accepted)

RESULTS PREDICTIVE VALUES FOR POOR OUTCOME >50% outside office hours Ruijter et al. Ann Neurol (accepted)

YIELD OF REPEATED MEASUREMENTS RELIABLE PROGNOSIS IN MORE PATIENTS

YIELD OF REPEATED MEASUREMENTS RELIABLE PROGNOSIS IN MORE PATIENTS

WHY CONTINUOUS EEG?  Identify more patients with poor or good outcome  Greater reliability of EEG conclusions  Detection of epilepsy  Future of brain monitoring

YIELD OF REPEATED MEASUREMENTS DECREASED UNCERTAINTY 18h 12h 24h 30h 24h 36h

WHY CONTINUOUS EEG?  Identify more patients with poor or good outcome  Greater reliability of EEG conclusions  Detection of epilepsy  Future of brain monitoring

YIELD OF CONTINUOUS MONITORING DETECTION OF ELECTROGRAPHIC SEIZURES  Electrographic seizures / status epilepticus: 10-20%  Non- convulsive ‘seizures’: 10 -15% Ruijter et al. Trials 2014 Knight et al. Epoilpesy Res 2013 Rittenberger et al. Neurocrit care 2012 Rossetti & Lowenstein. Lancet Neurol 2011

WHY CONTINUOUS EEG?  Identify more patients with poor or good outcome  Greater reliability of EEG conclusions  Detection of epilepsy  Future of brain monitoring

FUTURE OF BRAIN MONITORING BEDSIDE COMPUTER ASSISTED INTERPRETATION Tjepkema-Cloostermans et al. Crit Care Med 2017 Nagaraj et al. Clin Neurophysiol 2018

FUTURE OF BRAIN MONITORING BEDSIDE COMPUTER ASSISTED INTERPRETATION Tjepkema-Cloostermans et al. Crit Care Med 2017 Nagaraj et al. Clin Neurophysiol 2018

25% 50% uncertainty 25% Hofmeijer et al. Neurology 2015 Ruijter et al. Ann Neurol (accepted)

+ 25% 27.5% 50% uncertainty 45% 27.5% 25% Nagaraj et al. Clin Neurophysiol 2018

+ 25% 27.5% 30% 50% uncertainty 45% 40% 30% 27.5% 25% Tjepkema et al. Crit Care Med (accepted)

WHY NOT CONTINUOUS EEG? UNJUSTIFIED CONSERVATISM

CONCLUSION COMA AFTER CARDIAC ARREST cEEG  Móre reliable predictions in móre patients. For epilepsy detection cEEG is obligate Predictive values improve by bedside, computer assisted analysis Arguments to withhold cEEG are based on conservatism