Benzodiazepines versus non-benzodiazepine antiseizure medications as first-line agents for status epilepticus: analysis of real word data from a 9-years prospective cohort

Background and objectives: The treatment of status epilepticus (SE) follows a stepwise approach, with benzodiazepines (BDZ) being the first-line therapy. This study analyzed real-word data on use of BDZ and non-BDZ antiseizure medications (ASMs) in SE treatment over 9-years to evaluate whether non-BDZ given as a first-line treatment affect 30-day mortality and other outcomes. Methods: We included SE cases in patients aged ≥14 years who were prospectively registered at Baggiovara Civil Hospital (Modena, Italy) between September 1, 2013, and October 31, 2021. First-line treatment choices were dichotomized as: (i) i.v. BDZ; (ii) other ASMs. A multivariate model with logistic regression and an adjusted stepwise method for variables was used. Then, a propensity-score matched analysis was performed with clinical variables unevenly distributed between the two groups to evaluate the independent association between first-line therapy and 30-day mortality and secondary outcomes. Results: Six hundred and thirty patients were included: 73.5% (463/630) received a BDZ as first-line therapy and 26.5% (167/630) were treated with non-BDZ. In the primary analyses of the whole cohort, 30-day mortality was 25.9 and 35.3% in patients receiving BDZ and non-BDZ, respectively (p = 0.027). However, multivariate analysis adjusted for potential confounders showed that non-BDZ treatment was not independently associated with increased 30-day mortality. Patients who received BDZ as a first-line treatment had less orotracheal intubation and anesthetics within 24 h of SE onset; less frequent progression to refractory and super-refractory status epilepticus; less admission to and shorter stay in intensive care units; shorter time to SE cessation. In the propensity cohort (140 patients, mainly non-convulsive SE; NCSE), 30-day mortality was 30.7% (43/140), with no difference between BDZ-treated patients (30%; 21/70) and those who received non-BDZ (31.4%; 22/70) (p = 1.000). No difference in secondary outcomes was found, except for a shorter time to SE cessation among BDZ-treated patients. Conclusion: The use of non-BDZ first-line treatment was found to be frequent, approaching 25%. Our propensity-score matching analysis shows that in some patients, mainly with NCSE, the overall prognosis of SE was not affected by first-line use of non-BDZ drugs. In these cases, SE prognosis might only be partially dependent on the first medications administered and could be more influenced by other biological variables.