PURPOSE OF REVIEW: This paper aims to review and summarize the key contributions of EEG to prognostication after cardiac arrest (CA). RECENT FINDINGS: While there are more EEG patterns predicting poor than good outcome, even EEG patterns previously considered to be "very malignant" may result in survival with a meaningful neurological outcome depending on their underlying etiology as well as the continuity and reactivity of the EEG background. Regardless of the potentially confounding factors, EEG patterns are highly specific with a relatively low false-positive rate. The development of more complex and comprehensive approaches to quantitative EEG analysis could help improve the prognostic value of EEG, but this approach has its own limitations. Seizures and status epilepticus in the setting of CA predict poor outcomes, but it is not clear whether treating them prevents additional brain damage and results in improved outcome. Either continuous EEG or frequent intermittent EEGs should be obtained within the first 12-24 h of return of spontaneous circulation in order to capture highly dynamic and prognostic patterns. Even though EEG has high predictive value for outcomes after cardiac arrest, it should not be the only prognostic tool. Rather, to improve prognostication, EEG should be used in combination with the neurological examination and other ancillary tests.