Article Text

Frequency, causes, and consequences of burns in patients with epilepsy
  1. L KINTON,
  1. The National Hospital for Neurology and Neurosurgery, National Society for Epilepsy
  2. Chalfont St Peter, Buckinghamshire, UK
  1. Dr JS Duncan, Epilepsy Research Group, National Hospital for Neurology and Neurosurgery, Queen Square, London WC1N 3BG, UK. Telephone 0044 171 837 3611 ext 4259; fax 0044 171 837 3941; email j.duncan{at}

Statistics from

The increased incidence of burns in people with epilepsy has long been recognised. Previous surveys (via questionnaires in clinic or burns unit admissions) have identified cooking, showering, and heaters as the most common causes. The duration of epilepsy and frequency of seizures have been recognised as the greatest risk factors,1-4 compounded by lack of awareness and education among people with epilepsy about the risk of burns.3

The aim of this study was to determine the frequency, causes, and consequences of burns and scalds in patients with epilepsy. The population comprised patients with chronic epilepsy who were resident at the Chalfont Centre for Epilepsy (CCE). The residential part of the CCE consists of a series of houses that provide for varying degrees of independence in terms of self care. There is also a short stay tertiary referral inpatient assessment facility and a medical and nursing unit on site where any injury is documented and assessed.

The daily records of the Medical and Nursing Unit for the year June 1995–6 were examined and any record of a burn extracted and followed until it was recorded as healed. The records for the day of the burn were then examined to determine the cause of the burn and whether it was seizure related.

The residential houses were divided into three groups according to the level of self care of the patients within each. Dependency was largely governed by physical infirmity rather than by severity of the seizure disorder.

  • Dependent (98 residents)—all meals and hot drinks provided and help given with personal care.

  • Intermediate (94 residents)—all meals provided, some residents make their own hot drinks and there is a variable level of independence with personal care.

  • Independent (111 residents)—independent with regard to hot drinks, personal care, and some meals.

The number of residents in each category were calculated at the end of the year and remained relatively constant during the 1 year period. Seizures occurring in all patients were recorded prospectively in seizure diaries.

The number of seizures in the same 1 year period (June 1995 - June 1996) were calculated from the case records of the 303 residents. The number of seizures in the 28 bed assessment unit could not be calculated for the year, as there was a high patient turnover with a median stay of 32 days. The number of seizures from the residential houses were therefore extrapolated to allow for the number of patients in the assessment unit. No burns occurred in the assessment unit, which was classified as dependent with regard to the level of care made available to the inpatients.

A χ2 test was applied to the number of burns occurring in houses of differing levels of dependence to examine the significance of observed variation. This was then repeated correcting for the median number of seizures per person, in the three groups.

The results are shown in the table. Whereas the frequency of burns is regarded as accurate, the number of seizures may be underreported as nocturnal or minor seizures may not have been seen and recorded. Information as to whether an injury was seizure related or not was unavailable in three, and the cause of the burn was unavailable in two cases. Of the severe burns, one required skin grafting with healing occurring over 3 months and one patient attended an accident and emergency department, but was not admitted, and the burn took 4 months to heal. Two of the severe burns were complicated by methicillin resistant Staphyloccus aureus (MRSA).

infection, but healed uneventfully. Two other patients had burns that were also complicated by an infection but without growth of a particular organism. One received antibiotics and recovered quickly, the other was initially treated only with topical therapy and took longer to recover.

Most burns were related to seizures (19 of 34). This was particularly true of the more severe burns (five of six). A total of 18 631 seizures were recorded during the year meaning that about one seizure in every 980 resulted in a burn and one in every 3105 seizures resulted in a severe burn.

Factors influencing burns in patients with epilepsy

Most burns were caused by hot water injuries (25 of 34). More severe burns were mostly caused by larger quantities of water. Cookers, hot pipes, and heaters were rare causes of burns in this study.

There were significantly more burns in the more independent houses (p<0.05). The median number of seizures per person in the more independent houses was also greater as residents were placed in houses according to their physical disability rather than on the severity of their seizure disorder. Correcting for seizure frequency abolished the significance of the effect of level of dependence on the number of burns (p>0.1).

Burns are a serious but underrecognised complication of epilepsy. Previous studies have used two different approaches. The first has been to give patients attending an epilepsy clinic a questionnaire about any burns ever sustained.3 4 This method relies on the memory of the patients involved and therefore biases recording towards serious injuries. These studies also gave little idea of frequency, as there was no time limit imposed. The second approach has been to consider patients with epilepsy admitted to burns units.1 This again restricts the survey to severe burns. Neither method gives any idea of frequency as there is no set population with no calculation of the number of seizures over a period of time, hence no idea of level of risk.

The residents at the Chalfont Centre for Epilepsy live as normal a life as their level of disability allows. This means that some residents are not exposed to the same risk of burns as those living in the community. A consequence, however, is that the effectiveness of simple interventions can be gauged in a large population.

This case record survey of burns at the Chalfont Centre for Epilepsy is unique in that it included all burns and had an accurate record of the cause in most cases. The seizure charts kept prospectively by staff and residents allowed the frequency of burns per seizure to be calculated. The data are considered to be reasonably accurate as all but the most trivial burns were recorded. Brief, inconspicuous seizures may not have been recorded, particularly simple partial, absence seizures or myoclonic jerks. All complex partial, secondary generalised, and generalised tonic-clonic seizures would, however, have been noted. From the 303 patients there were a total of 34 burns, at least 19 of which were seizure related, and 18 631 seizures were recorded during the 1 year period. Only one required skin grafting but six required over 3 weeks to heal. Therefore only one in every 980 seizures resulted in a burn with one in 3105 resulting in a burn taking longer than three weeks to heal.

Although there were significantly more burns in independent houses, there were also a greater number of seizures per person on average in this group. As most burns were related to seizures, correcting for the number of seizures per person abolished the significance of level of independence on number of burns.

It is of note that almost no burns occurred secondary to showers or heating devices, which is in stark contrast with all previous reports. Five years ago thermostatic regulators were put on to hot water supplies and covers were installed to protect hot water pipes and heating appliances. These changes followed several severe burns in previous years from the above sources.

A further difference between this study and previous, community based studies was that there were very few cooking related burns in this study. This is almost certainly due to the fact that even the most independent residents do relatively little cooking as a canteen is available at lunchtimes and residents are encouraged to use microwave ovens.

Lastly, most burns were caused by hot drinks; either severe burns from large containers of hot water such as kettles or teapots, or milder burns from individual cups. The scope for reducing future morbidity from burns in persons with epilepsy is therefore in this area. The use of microwave ovens to heat individual cups of hot water is one option as these were already in widespread use for food preparation. The use of insulated flasks instead of teapots would prevent surface burns and the exit valve would prevent hot water being spilt. Further, hot water heaters, which are only able to dispense a fixed amount of water, are available and would be an alternative to kettles.

Burns are therefore a relatively rare but potentially serious cause of morbidity in people with epilepsy. The more severe burns are also largely preventable by simple interventions, which do not interfere greatly with a person’s independence.


View Abstract

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.