O29 EEG criteria of non-convulsive status epilepticus

Abstract

Objectives A panel of experts, in 2013, at the 4th Colloquium on Status Epilepticus in Salzburg, proposed a set of EEG criteria for nonconvulsive status epilepticus (NCSE), known as the Salzburg-criteria for NCSE. The objective was to validate these criteria. Methods We did a diagnostic study where two raters blinded to all other patient data analysed EEG recordings from participants admitted three centres ( n = 220). Participants were placed into one of two groups: the clinical validation group which consisted of consecutive patients with clinical suspicion of NCSE referred to EEG recording ( n = 120), or the control group which consisted of consecutive patients with abnormal EEG-findings, but without clinical suspicion of NCSE ( n = 100). The reference standard was inferred from all clinical and para-clinical data, therapeutic response, and final outcome, as evaluated by two raters who were blinded to diagnosis according to the Salzburg-criteria. Results In the clinical validation group, sensitivity was 97.7% and specificity was 89.6% (overall accuracy: 92.5%). In the control group, specificity was 97%; sensitivity was not calculated in this group as the reference standard did not confirm NCSE in any of the control patients. Inter-rater agreement (IRA) was high, both for Salzburg-criteria ( k = 0.87) and for the reference standard ( k = 0.95). Therapeutic changes were seen significantly more often in the group of patients fulfilling Salzburg-criteria (84%) as compared to patients who did not (16%). Significance Salzburg-criteria for NCSE have high diagnostic accuracy and excellent IRA, making them suitable for implementation in clinical practice.