Background <p>Convulsive status epilepticus (CSE) is the most common neurological emergency in children. Without prompt, safe and effective intervention, CSE can cause adverse neurological sequelae or death. There is increasing use of intramuscular midazolam for the treatment of childhood CSE. The aim of this systematic review is to investigate the efficacy and safety of intramuscular midazolam versus midazolam by other routes or other non-intravenous benzodiazepines in childhood CSE.</p> Methods <p>MEDLINE, PubMed, Embase, Cochrane CENTRAL and Google Scholar were searched from 1 January 1980 to 29 August 2025. The primary outcome of interest was cessation of seizures within 5–10 min of treatment. Secondary outcomes included time to seizure cessation and incidence of respiratory depression/hypotension.</p> Results <p>Five studies were identified comparing intramuscular midazolam with buccal midazolam, intranasal midazolam or rectal diazepam. Data were reported for 428 children treated in the emergency department with intramuscular midazolam (<i>n</i> = 196), intranasal midazolam (<i>n</i> = 34), buccal midazolam (<i>n</i> = 115) or rectal diazepam (<i>n</i> = 83) and 196 seizure episodes in 185 children treated at home with intramuscular midazolam (<i>n</i> = 69), intranasal midazolam (<i>n</i> = 60) or buccal midazolam (<i>n</i> = 67). In single studies, there was no difference between intramuscular midazolam and intranasal midazolam or rectal diazepam in achieving seizure cessation, but intramuscular midazolam was associated with shorter time to seizure cessation. Meta-analysis was only possible comparing intramuscular midazolam with buccal midazolam. Intramuscular midazolam was superior to buccal midazolam for time to seizure cessation (<i>p</i> &lt; 0.00001). While there were no inter-group differences in other outcomes, point estimates favoured intramuscular midazolam. Few cases of respiratory depression or hypotension were reported, with no inter-group differences. No studies involved treatment by paramedics, sample sizes were modest and none were in Western high-income countries. Quality of evidence was low or very low.</p> Conclusions <p>There is insufficient clinical trial evidence for safety and efficacy of intramuscular midazolam compared with midazolam by other routes or other non-intravenous benzodiazepines for childhood CSE.</p>