Prognostic significance of clinical seizures after cardiac arrest and target temperature management

Abstract

AimClinical seizures are common after cardiac arrest and predictive of a poor neurological outcome. Seizures may be myoclonic, tonic-clonic or a combination of seizure types. This study reports the incidence and prognostic significance of clinical seizures in the target temperature management (TTM) after cardiac arrest trial. Our hypotheses were that seizures are associated with a poor prognosis and that the incidence of seizures is not affected by the target temperature. MethodsPost-hoc analysis of reported clinical seizures during day 1–7 in the TTM-trial including their treatment, EEG-findings, and long-term neurological outcome. The trial randomised 939 comatose survivors to TTM at 33°C or 36°C with strict criteria for withdrawal of life-sustaining therapies. Sensitivity, specificity and false positive rate for poor outcome were reported for different types of seizures. ResultsClinical seizures were registered in 268 patients (29%), similarly distributed in both intervention arms. Early and late seizures were equally predictive of poor outcome. Myoclonic seizures were the most common (240 patients, 26%) and the most predictive of a poor outcome (sensitivity 36.1%, false positive rate 4.3%). Two patients with status myoclonus regained consciousness, one with a good neurological outcome, generating a false positive rate of poor outcome of 0.2% (95%CI 0.0–1.0). ConclusionClinical seizures are common after cardiac arrest and indicate poor outcome with limited specificity. Prolonged seizures are a very grave sign but occasional patients may have a good outcome. The level of the target temperature does not affect the prevalence or prognostic significance of seizures.