The crude mortality difference between first-line treatment groups did not persist as an independent association after adjustment.
The retrospective single-center study included 630 status-epilepticus episodes in patients aged fourteen or older. Thirty-day mortality was 25.9% after first-line benzodiazepines and 35.3% after non-benzodiazepine agents. Multivariable analysis did not identify non-benzodiazepine treatment as independently associated with higher mortality. In a matched cohort of 140 patients, mainly with nonconvulsive status, mortality was 30.0% versus 31.4%. Time to seizure cessation still favored benzodiazepines.
Clinicians selected treatment in routine care. Seizure type, severity, recognition and contraindications can influence both choice and outcome. Matching balances measured features, while leaving possible unmeasured confounding. The largely nonconvulsive matched sample should not be generalized to every convulsive emergency or interpreted as proof of treatment equivalence.
In a treatment-delay or causation review, preserve the seizure phenotype and the timing of recognition, administration and cessation. The original crude percentages alone cannot establish that one drug choice caused death. Equally, a nonsignificant adjusted mortality comparison does not erase the importance of timely seizure control. This source demonstrates how population selection and adjustment change an observational conclusion without resolving an individual clinical decision.
Source: PMC12649711, supplied full-text paper. Findings are paraphrased; forensic implications are an interpretation.
Related: Adult Status-Epilepticus Meta-Analysis Favored Lorazepam With Very Low Certainty.