Status epilepticus in sub-Saharan Africa: New findings


  • Charles R. Newton,

    Corresponding author
    1. Neurosciences Unit, Institute of Child Health, University College London, London, United Kingdom
    2. Department of Psychiatry, University of Oxford, Oxford, United Kingdom
    • Centre for Geographic Medicine Research (Coast), Kenya Medical Research Institute, Kilifi, Kenya
    Search for more papers by this author
  • Symon M. Kariuki

    1. Centre for Geographic Medicine Research (Coast), Kenya Medical Research Institute, Kilifi, Kenya
    2. Department of Psychiatry, University of Oxford, Oxford, United Kingdom
    Search for more papers by this author

Address correspondence to Professor Charles Newton, Centre for Geographic Medicine Research (Coast), Kenya Medical Research Institute, PO Box 230, Kilifi, Kenya. E-mail:


Status epilepticus (SE) is common in sub-Saharan Africa, particularly in children. Most cases in children are caused by infections, particularly malaria in endemic areas. The outcome is worse than in the West, probably because of delays in initiating treatment, and lack of skills and facilities for the management of SE. However some of the causes, for example, falciparum malaria, offer challenges in the diagnosis and the treatment of SE. Exposure to falciparum malaria increases the risk of SE. Much of the SE in Africa could be prevented by reducing the incidence of infections. The outcome could be improved through education, development of locally appropriate guidelines, and provision of appropriate facilities.

Status epilepticus (SE) occurs either in the context of febrile seizures, acute symptomatic seizures, or as a complication of epilepsy. In sub-Saharan Africa (SSA), acute symptomatic seizures and epilepsy are more common than in resource rich countries; and in each of these conditions, SE appears to be more frequent, although there are few published data. Infections are important causes of SE. This article briefly reviews recent studies on status epilepticus in SSA, with a particular emphasis on the role of parasitic infections.

Definition of Status Epilepticus

The definition of status epilepticus in SSA is often problematic, since many people do not have watches or clocks, and the culture can influence the perception of time. Although the duration of seizures can be measured accurately in patients admitted to the hospital, assessing the duration of seizures prior to admission is often difficult. Thus the definition of SE may consist of those who fulfill the International League Against Epilepsy (ILAE), but also those with a history suggestive of probable SE (Sadarangani et al., 2008). Recall bias of duration of seizures is common, but the use of culturally accessible events, for example, boiling a pot of maize, may help (Kariuki SM, unpublished data).

Burden of Epilepsy in SSA

The prevalence of epilepsy in SSA is twice that of other continents, such as Europe, Asia, and North America (Preux & Druet-Cabanac, 2005; Ngugi et al., 2010). The larger burden is thought to be caused by an increase in the incidence of risk factors such as perinatal insults and parasitic diseases (Newton & Garcia, 2012). A recent study of five sites across Africa has identified neonatal insults and parasitic infections as major risk factors for epilepsy on this continent (Ngugi et al., 2013). Neonatal insults were responsible for one third of the cases in children and parasitic diseases for one third in adults.

Burden of Status Epilepticus

There are few epidemiologic studies of SE of other continents, such as SSA. One study in Kilifi on the coast of Kenya, recorded a minimum incidence of convulsive SE (CSE) as 35/100,000/year in children (0–13 years) based on hospital admissions (Sadarangani et al., 2008), which is 2–8 times (depending upon the definition used) that found in London (Chin et al., 2006). This rate is the minimum incidence, since a proportion of children will die in SE before reaching hospital, and others will be treated in private clinics. In this study, 15% died in hospital, but 21% died after being discharged from hospital (Sadarangani et al., 2008), often from another episode of SE (Prins A, Mung'ala Odera ECV, Sadarangani M, Holding CSP, Fegan G, Newton CR, unpublished data). Seventy-one percent of the cases of SE were caused by infections, mainly falciparum malaria (58%), but only 3% were previous cases of epilepsy.

In the multicenter cross-sectional study in 586,607 people (Ngugi et al., 2013) in SSA, 25% of the people with active convulsive epilepsy had a history of SE (Kariuki SM, unpublished data). The frequency of SE in the people with epilepsy differed significantly across the sites, although in some sites the recall and/or documentation of SE may be poor. In this study, 69% of the SE was reported to be associated with a febrile illness (Kariuki SM, unpublished data). SE was significantly more common in generalized epilepsy than focal epilepsy.

In Kilifi, Kenya, one of the sites for this multicenter study, only 40% of children who reported SE in the community were admitted to Kilifi District Hospital (Kariuki SM, unpublished data). About 58% of SE was associated with a febrile illness. Seizures in SE occurred significantly earlier in those with a history of febrile seizures compared to those without, and later in those with epilepsy compared to those without. Focal seizures and focal electroencephalographic features were present in 54% and 36% of those with SE, respectively. Important risk factors were perinatal complications, parasitic infections (particularly exposure to falciparum malaria), history of seizures in the family, neurologic impairment, and visit to traditional healers.

Malaria and Status Epilepticus

In malaria endemic areas, falciparum malaria is the most common cause of SE, particularly in the context of acute symptomatic seizures in children. Attribution of SE to malaria in endemic areas is problematic, since up to 70% of asymptomatic children in the community will have a peripheral parasitemia. Furthermore, there is debate as to whether children with malaria have febrile seizures or acute symptomatic seizures, since malaria is the most common cause of febrile illness in children aged 6 months to 6 years in malaria endemic areas, but parasites sequester within the cerebrovasculature.

In Kilifi, we have calculated the malaria-attributable fraction using logistic regression to estimate the proportion of CSE that can be attributed to malaria (Kariuki et al., 2011). In these models, we estimated that 92% (95% confidence interval 91–93%) of children, who were admitted to hospital with convulsive SE and falciparum parasitemia, could be attributed to malaria. This estimation was confirmed by monitoring the reduction in the incidence of CSE attributable to malaria with the decrease in the incidence in malaria that occurred in this area over a 7-year period.

The pathologic hallmark of falciparum malaria is the sequestration of parasitized red blood cells in the microvasculature of the deep organs, particularly the brain. Although the parasites are rarely seen within the parenchyma of the brain, the sequestration causes a wide range of inflammatory and metabolic perturbations that affect the brain function. Seizures associated with malaria are different from the febrile seizures seen in nonmalaria endemic areas, in that the proportion of focal seizures, repetitive seizures, and prolonged seizures are much more common in the malaria-associated seizures than febrile seizures (Waruiru et al., 1996).

Antiepileptic Drugs for the Treatment of SE in SSA

The pharmacologic management of SE is problematic in SSA. The antiepileptic drugs available for the treatment of SE are limited in many African hospitals, mainly to diazepam and phenobarbital. Furthermore, the supply of parenteral phenobarbital has been erratic in some countries (Wilmshurst & Newton, 2005), and this lack of supply has initiated studies or the administration of oral phenobarbital in the management of SE (Wilmshurst et al., 2010). Intravenous access is often difficult in low-resource settings, so alternative routes of administration have been examined. In particular, intramuscular administration is widely acceptable in SSA, and is the route of administration of many drugs, for example, antimicrobials, in understaffed African hospitals.

Prevention of Status Epilepticus

Because SE is common in SSA and is associated with greater prevalence of neurologic deficits and mortality, measures to prevent SE have been examined. In a double-blind placebo-randomized control study of 340 Kenyan children admitted with cerebral malaria, a single dose of intramuscular phenobarbital (18 mg/Kg) significantly reduced the proportion of SE from 14% to 5% (Crawley et al., 2000). However phenobarbital was associated with increased mortality (18% in the phenobarbital group vs. 8% placebo), probably because of the respiratory depression in association with use of diazepam. Fosphenytoin may be useful, since it can be administered intramuscularly and has less respiratory depression than phenobarbital. In Kenyan children with SE, intramuscular fosphenytoin (18 mg/kg) achieved concentrations above 10 mg/L at 20 min and stopped the seizures in 64% in those with SE (Ogutu et al., 2003). In a recent double-blind placebo-randomized control study, a single dose of intramuscular fosphenytoin (phenytoin equivalents 20 mg/kg) administered to 173 children with acute encephalopathy did not significantly reduce status epilepticus or prolonged seizures (Gwer SA, Idro RI, Fegan G, Chengo EM, Mpoya A, Kivaya E, Crawley J, Ndirangu SM, Kihara MN, Ogutu BR, Kirkham FJ, Newton CR, unpublished data). Other drugs, for example, levetiracetam need to be tested.

Treatment of Status Epilepticus

The mortality of SE is higher in SSA than in the resource-rich countries. This may be caused by the delay in presenting to hospital, poor management of SE in hospital, and underlying etiology (Wilmshurst et al., 2011). Patients who have SE may be taken to traditional healers first (Kariuki SM, unpublished data), and/or their presentation to hospitals may be delayed even further by lack of infrastructure or transport. Many hospitals in SSA lack facilities such as infusion pumps, ventilators, and staff to manage SE appropriately (Wilmshurst et al., 2011).

In a randomized controlled trial in Malawian children, intranasal lorazepam (100 μg/kg) failed to stop acute seizures in 25%, compared to intramuscular paraldehyde (0.2 ml/kg), which failed to stop 39% (Ahmad et al., 2006). This difference may be attributed to the relatively low dose of paraldehyde used. In a randomized controlled trial in Ugandan children admitted with acute seizures to an urban hospital, rectal diazepam (0.5 mg/kg) failed to stop seizures in 43% within 10 min compared with 30% who received buccal midazolam (0.5 mg/kg) (Mpimbaza et al., 2009). Focal seizures, cerebral malaria, and blood glucose ≥11 mmol/l at presentation were independent predictors of treatment failure defined as seizures lasting >10 min or recurrence within 1 h of treatment (Mpimbaza et al., 2009). However, in an audit of the treatment of seizures in Kenyan children, 64% seizures were not terminated with intravenous diazepam (0.3 mg/kg) compared to 58% with intramuscular paraldehyde (0.4 ml/kg) (Ikumi ML, Muchohi SN, Ohuma EO, Ogutu BR, Kokwaro GO, Newton CR, unpublished data). The later study reported the findings from a District General Hospital in rural Kenya, and the lack of response could be the result of the prolonged duration of seizures before treatment is started, or the cause of the seizures, for example, malaria (Ikumi ML, Muchohi SN, Ohuma EO, Ogutu BR, Kokwaro GO, Newton CR, unpublished data). Seizures caused by falciparum malaria appear to be resistant to diazepam, possible because of decrease in the binding capacity of GABAA receptors in malaria (Ikumi et al., 2008).


Status epilepticus is common in SSA, particularly in children in whom it is often caused by infections. The outcome is worse, possibly due to the etiology, but also delays in accessing treatment and lack of skills and facilities. Falciparum malaria is an important cause of SE in endemic areas, but presents unique problems in defining malaria-associated SE and treatment.


Charles Newton is supported by the Wellcome Trust (083744). Symon Kariuki is supported by the Wellcome Trust Research Training Fellowship (099782).


The authors have no conflicts of interest to declare. The authors confirm that they have read the journal's position on issues involved in ethical publication and affirm that this report is consistent with these guidelines.