Abstracts

Rationale: Non convulsive status epilepticus (NCSE) is a neurological emergency with high mortality and morbidity. Diagnosis of NCSE without long-term EEG monitoring (LTM) is a challenge. Predicting the progression of generalized or complex partial status epilepticus to NCSE will help clinicians manage 'at-risk' patients by obtaining LTM and treating them appropriately at an earlier point. Methods: A prospective database of patients admitted to the Neurocritical Care unit at our institution was queried status epilepticus (SE) patients between November 2013 and January 2016. Patients with myoclonic SE after cardiac arrest were excluded. A chart review was done to collect demographics, history of epilepsy, etiology of seizures, type of seizure at onset of SE, development of refractory SE (RSE) and presence of NCSE at any point during hospitalization. The type of seizures at the onset of SE was further categorized as generalized convulsive SE (GCSE), complex partial SE (CPSE), myoclonic SE and NCSE. RSE was defined as ongoing seizure after administration of adequate dose of benzodiazepines and an antiepileptic drug (AED). Statistical analysis was carried out on MedCalc v16.4.3. Results: Our final cohort included 100 patients and 31% went on to develop NCSE. Their demographics have been listed in Table 1. The median age in patients who developed NCSE was higher (63 years [IQR 54.3-72.8] vs. 57 years [IQR 44.3-63]; p=0.02). Neither gender nor race predicted the development of NCSE. Vascular etiologies were the most common cause for SE in our patients (32%), while traumatic brain injury (20%) and tumor/post resection (16%) were the other common etiologies. There were no differences in rates of NCSE among various etiologies. A prior history of epilepsy was a strong negative predictor of NCSE. Among patients with a history of epilepsy, only 23% developed NCSE and 77% never went into NCSE (p < 0.01). Among patients with NCSE, there was a higher incidence of RSE (86.7% vs. 13.3%; p < 0.01). Patients presenting with CPSE had a much higher likelihood of progressing to NCSE (72.7% vs. 27.2%; p < 0.001). A logistic regression analysis with backward selection for the significant outcomes (Table 2) yielded CPSE at presentation as a predictor of developing NCSE (OR 4.4 [95% CI 1.3-14.8; p=0.02) and a history of epilepsy as a negative predictor (OR 0.2 [95% CI 0.1-0.6]; p < 0.01). The development of refractory status approached significance (OR 4.4 [95%CI 0.8-23.8; p=0.07). Conclusions: NCSE is a diagnostic and treatment challenge. NCSE is more likely to develop in patients presenting with CPSE and new onset seizures. More CPSE patients developed NCSE, probably due to delay in recognition and early aggressive treatment in these patients. The development of RSE may increase the chance of developing NCSE even though it did not attain statistical significance in our population. Larger prospective trials are required to elucidate these findings and define treatment parameters. Funding: No funding received.