Purpose Convulsive status epilepticus (CSE) has a high morbidity and mortality in developing countries. This study aims to determine what factors affect the outcome of CSE in Uganda.

Methods Children with CSE admitted to Mbarara Regional Referral Hospital were consecutively enrolled over 8 weeks in May-June 2011. Data on patient demographics, previous neurodisability, aetiology and management of CSE were gathered. Children were seen at 4-6 weeks after CSE to determine outcome in terms of death, neurological deficit or subsequent epilepsy.

Results 36 children were included. Average seizure duration was 72.2 hours. Only 22% of children presented within 3 hours of seizure onset, 75% of these were local. Most children had treatment in the community before attending hospital. Most attended a traditional healer or private clinic (neither of which used antiepileptics to treat CSE), only 28% attended government health centres, where diazepam was sometimes given. CSE duration was only 10.1 hours when diazepam was given in the community, vs. 96.6 hours when it was not. 72% of seizures were due to acute symptomatic causes (meningitis or cerebral malaria). 50% seizures were focal.

25% died, these children had a mean seizure duration of 138.7 hours. All deaths were in children with acute symptomatic CSE, and tended to be younger. The duration of seizure was considerably longer in children who died than in children who survived (65.8 hours vs. 42.5 hours) when cause of CSE was controlled by selecting only cases with acute symptomatic CSE. 52% (12/23) of children who were followed up developed a new neurological deficit (mean CSE duration 144.3 hours), including focal motor deficits or movement disorders (47%), learning disability (43%), and behaviour difficulties (9%). 21% of children developed subsequent epilepsy (mean SE duration 416.0 hours). None of the 8 children whose seizures lasted less than 3 hours died or, when known, had subsequent neurological deficits or epilepsy. The average duration of CSE for children who suffered no adverse events was 20.6 hours.

Conclusion Aetiology and duration of CSE are powerful predictors of outcome. Duration of CSE is modifiable by community education and the use of pre-hospital antiepileptic drugs.