Epileptiform patterns predicting unfavorable outcome in postanoxic patients: A matter of time?

Francesco Misirocchi; Giorgia Bernabè; Lucia Zinno; Marco Spallazzi; Alessandro Zilioli; Elisa Mannini; Stefania Lazzari; Valentina Tontini; Carlotta Mutti; Liborio Parrino; Edoardo Picetti; Irene Florindo

doi:10.1016/j.neucli.2023.102860

Epileptiform patterns predicting unfavorable outcome in postanoxic patients: A matter of time?

Neurophysiol Clin. 2023 Feb;53(1):102860. doi: 10.1016/j.neucli.2023.102860. Epub 2023 Apr 1.

Authors

Affiliations

¹ Unit of Neurology, Department of Medicine and Surgery, University Hospital of Parma, Parma, Italy. Electronic address: Francesco.misirocchi@unipr.it.
² Infermi Hospital, AUSL Romagna, Rimini, Italy.
³ Unit of Neurology, University Hospital of Parma, Parma, Italy.
⁴ Unit of Neurology, Department of Medicine and Surgery, University Hospital of Parma, Parma, Italy.
⁵ Unit of Neurology, University Hospital of Parma, Parma, Italy; Sleep Disorders Center, Department of Medicine and Surgery, University of Parma, Parma, Italy.
⁶ Unit of Neurology, Department of Medicine and Surgery, University Hospital of Parma, Parma, Italy; Sleep Disorders Center, Department of Medicine and Surgery, University of Parma, Parma, Italy.
⁷ Intensive Care Unit, University Hospital of Parma, Parma, Italy.

PMID: 37011480
DOI: 10.1016/j.neucli.2023.102860

Abstract

Objective: Historically, epileptiform malignant EEG patterns (EMPs) have been considered to anticipate an unfavorable outcome, but an increasing amount of evidence suggests that they are not always or invariably associated with poor prognosis. We evaluated the prognostic significance of an EMP onset in two different timeframes in comatose patients after cardiac arrest (CA): early-EMPs and late-EMPs, respectively.

Methods: We included all comatose post-CA survivors admitted to our intensive care unit (ICU) between 2016 and 2018 who underwent at least two 30-minute EEGs, collected at T0 (12-36 h after CA) and T1 (36-72 h after CA). All EEGs recordings were re-analyzed following the 2021 ACNS terminology by two senior EEG specialists, blinded to outcome. Malignant EEGs with abundant sporadic spikes/sharp waves, rhythmic and periodic patterns, or electrographic seizure/status epilepticus, were included in the EMP definition. The primary outcome was the cerebral performance category (CPC) score at 6 months, dichotomized as good (CPC 1-2) or poor (CPC 3-5) outcome.

Results: A total of 58 patients and 116 EEG recording were included in the study. Poor outcome was seen in 28 (48%) patients. In contrast to late-EMPs, early-EMPs were associated with a poor outcome (p = 0.037), persisting after multiple regression analysis. Moreover, a multivariate binomial model coupling the timing of EMP onset with other EEG predictors such as T1 reactivity and T1 normal voltage background can predict outcome in the presence of an otherwise non-specific malignant EEG pattern with quite high specificity (82%) and moderate sensitivity (77%).

Conclusions: The prognostic significance of EMPs seems strongly time-dependent and only their early-onset may be associated with an unfavorable outcome. The time of onset of EMP combined with other EEG features could aid in defining prognosis in patients with intermediate EEG patterns.

Keywords: Cardiac arrest; EEG; Epileptiform EEG; Neurological prognosis; Post-anoxic coma.

MeSH terms

Coma / diagnosis
Coma / etiology
Electroencephalography
Heart Arrest* / complications
Humans
Prognosis
Seizures / complications
Seizures / etiology
Status Epilepticus* / diagnosis
Status Epilepticus* / etiology