Comparing Patients With Isolated Seizures and Status Epilepticus in Intensive Care Units: An Observational Cohort Study

Objectives: To assess the frequency of status epilepticus (SE) among seizing critically ill adult patients and to determine clinical differences between patients with isolated seizures and patients with SE in the intensive care unit(ICU).

Methods: From 2015-2020, all consecutive adult ICU patients at a Swiss tertiary care center with isolated seizures or SE as reported by intensivists and/or consulting neurologists were identified by screening of all digital medical, ICU, and EEG records. Patients <18 years of age and patients with myoclonus due to hypoxic-ischemic encephalopathy but without seizures on EEG were excluded. The frequency of isolated seizures and SE as well as clinical characteristics at seizure onset associated with SE were the primary outcomes. Uni- and multivariable logistic regression was performed to identify associations with the emergence of SE.

Results: Among 404 patients with seizures, 51% had SE. Compared to patients with isolated seizures, patients with SE had a lower median Charlson comorbidity index(CCI)(3 versus 5, p<0.001), fewer fatal etiologies (43.6% versus 80.5%, p<0.001), higher median Glasgow Coma Scores (7 versus 5, p<0.001), fever more frequently (27.5% versus 7.5%, p<0.001), shorter median ICU and hospital stay (ICU: 4 versus 5 days, p=0.039; hospital stay: 13 versus 15 days, p=0.045), and recovered to premorbid function more often (36.8% versus 17%, p<0.001). Multivariable analyses revealed decreased odds ratios(OR) for SE with increasing CCI (OR=0.91, 95%confidence interval[CI] 0.83-0.99), fatal etiology (OR=0.15, 95%CI 0.08-0.29), and epilepsy (OR=0.32, 95%CI 0.16-0.63). Systemic inflammation was an additional association with SE after excluding patients with seizures as the reason for ICU admission (OR_{for CRP}=1.01, 95%CI 1.00-1.01; OR_{for fever}=7.35, 95%CI). While fatal etiologies and increasing CCI remained associated with low odds for SE after excluding anesthetized patients and hypoxic-ischemic encephalopathy, inflammation remained associated in all subgroups except patients with epilepsy.

Conclusions: Among all ICU patients with seizures, SE emerged frequently and seen in every second patient. Besides the unexpected low odds for SE with higher CCI, fatal etiology, and epilepsy, the association of inflammation with SE in the critically ill without epilepsy represents a potential treatment target and deserves further attention.