Abstracts

Rationale: Several EEG-criteria for nonconvulsive status epilepticus (NCSE) have been proposed, but none of them has been clinically validated yet. We assessed diagnostic accuracy of the EEG-criteria proposed by a panel of experts at the 4th London-Innsbruck-Colloquium on Status Epilepticus in Salzburg, 2013 (Salzburg-criteria for NCSE). Methods: Two raters analyzed EEG recordings, blinded to all other data, from 220 patients in three centers (Danish Epilepsy Centre, Dianalund Denmark; Aarhus University Hospital, Aarhus Denmark; Paracelsus Medical University, Salzburg Austria), comprising a clinical validation group and a control group. The clinical validation group consisted of 120 consecutive patients with clinical suspicion of NCSE referred to EEG (in Austria: January-February 2014; in Denmark: August 2014 - January 2015). The control group consisted of 100 consecutive patients with abnormal EEG-findings, but without clinical suspicion of NCSE (in Austria: January 2014, in Denmark: January 2015). Reference standard was inferred from all clinical and para-clinical data, therapeutic response and final outcome, evaluated by two raters, blinded to results of the Salzburg-criteria. We calculated sensitivity, specificity, positive and negative predictive value, overall diagnostic accuracy for Salzburg-criteria, and interrater agreement (IRA) for Salzburg-criteria and for the reference standard. Results: In the clinical validation group, sensitivity was 97·7% and specificity was 89·6% (overall accuracy: 92·5%). PPV was 84% and NPV was 98·6%. In the control group, specificity was 97·0%. IRA was almost perfect, 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 (42 out of 50 patients = 84%) as compared to patients who did not (11 out of 70 patients = 15·7%) (p< 0·001) Conclusions: Salzburg-criteria for NCSE have high diagnostic accuracy and excellent IRA. This is the first clinical validation study of diagnostic criteria for NCSE. Funding: No funding related to this project.