To identify predictors of functional outcome for patients admitted to the intensive care unit with status epilepticus (SE) including admission characteristics and inpatient complications.

Status epilepticus is a neurological emergency associated with significant morbidity and mortality. Scales have been developed to help clinicians predict mortality in SE and rate its severity. There is a growing body of literature on complications associated with SE. We hypothesize that the underlying etiology of SE and other clinical and demographic data at the time of admission can be utilized to predict clinical outcome. 

A retrospective chart review of consecutive patients admitted to the neurological ICU at University of Maryland Medical Center from 1/1/2016-12/31/2016 for SE evaluation and management. Descriptive statistics are reported for admission characteristics and complications. A univariate analysis was used to explore predictors of clinical outcome using the modified Rankin Score (mRS) at discharge. Further analyses of this cohort are ongoing and will be reported.

There were 106 admissions during the study period, including 5 patients who were admitted twice. The mean age was 53.6±17.7 and 46.2% were female. 62.2% of patients had an in-hospital clinical event, 102 underwent continuous EEG (cEEG) and electrographic seizures were captured in 31.4%. Common complications included hypotension requiring vasopressor therapy (32.1%), infection (29.2%), skin ulcers (18.9%), severe hyperglycemia (16%) and need for gastrostomy or tracheostomy (17.9 and 14.2%, respectively). Admission GCS (F=5.324,p=0.024), EMSE score (F=14.202,p<0.001) and Cumulative Illness Rating Scale-neurological (F=8.404,p=.005) were significantly associated with mRS at discharge.

Patients admitted to our neurocritical care center for evaluation of SE had diverse clinical characteristics. Factors including admission GCS, severity of premorbid neurological illness and EMSE, a scale combining clinical factors and EEG, are predictive of poor functional outcome.  Further analyses are planned with a larger cohort, specifically predicting yield of cEEG based on baseline patient characteristics.