Epilepsy: a progressive disease?BMJ 1997; 314 doi: https://doi.org/10.1136/bmj.314.7078.391 (Published 08 February 1997) Cite this as: BMJ 1997;314:391
Still no answer to the controversy over whether seizures beget more seizures
- Bernard Sadzot, Associate professora
It is amazing how after years of scientific research and therapeutic progress many simple and basic questions about the epilepsies remain unanswered. Even the natural course of epilepsy is not well known because of the widespread use of antiepileptic drugs. Gowers suspected that seizures trigger subsequent seizures or, in other words, that there is a facilitation phenomenon.1 This idea still survives but has yet to be substantiated. We also do not know if antiepileptic drugs alter the natural course of epilepsy and prevent chronicity and intractability developing. Are the risks of further seizures in newly diagnosed patients (especially children) worth the side effects associated with antiepileptic drugs? What is the optimum time to begin treatment? Clinical studies have produced diverging and sometimes conflicting views on these problems.
In this week's BMJ, van Donselaar et al (p 401) followed 204 untreated children aged 1 month to 16 years who experienced one (123 patients) or more idiopathic or “remote symptomatic” tonic-clonic seizures (including seizures with partial onset).2 They analysed the time between seizures until the start of treatment, the fourth untreated seizure, or the end of the two year follow up. Only a few patients showed an accelerating pattern with decreasing time between seizures. Most showed a varying or slowing pattern.
These results contrast with those of Elwes et al in 1988.3 In a retrospective, hospital based study, they examined the time between primary or secondary generalised tonic-clonic seizures in untreated patients. Many of them had an accelerating seizure pattern in the early stages of the disease. Even if their population of patients is not comparable with that studied by van Donselaar et al, their findings are in keeping with the observation that remission rates with antiepileptic drugs seem inversely correlated with the number of seizures before treatment.4 5 This implies that early treatment may be important to prevent epilepsy from evolving into a chronic and more intractable state.
The concept of evolving epileptogenesis has been fuelled by studies on the “kindling” model of epilepsy. The term kindling refers to a dynamic phenomenon whereby repeated administration of an initially subconvulsive electrical stimulus results in progressive intensification of seizure activity culminating in a generalised seizure.6 This increased sensitivity to the stimulus (and the reduction of seizure threshold) is permanent. Kindling can be established in numerous mammalian species by electrical stimulus and various pharmacological agents, including excitatory amino acids like kainic acid. However, kindled animals rarely develop either spontaneous seizures or the brain lesions observed in human epilepsy.
Whether similar mechanisms operate in the development of epilepsy in humans has not been established, but there are certainly valid arguments to suggest that this may be the case. Adults with complex partial epilepsy arising from one temporal lobe may progress so that both temporal lobes eventually trigger complex partial seizures. In a series of patients with temporal brain tumours, Morell has quite convincingly shown that secondary epileptogenesis takes place in humans, leading to the development of mirror foci.7 Furthermore, he provided evidence that the likelihood of a secondary focus becoming permanent rises with increasing frequency of seizures, which again underscores the importance of rigorous seizure control. In a retrospective and longitudinal study of a large number of electroencephalograms, Hughes found that the incidence of bilateral foci, as opposed to unilateral foci, increased with age at a rate of almost 1% a year.8 Clinical signs of the development of bilateral temporal foci were seen in 34% of patients. The frequent occurrence of multiple foci, usually in synaptically connected sites, in the temporal lobes of patients with epilepsy also raises the possibility of kindling-like phenomena in humans.9
However, the situation is not as clear as it may seem. Van Donselaar et al's study is not the only one that does not support the notion that repeated seizures aggravate the epileptic process. Epidemiological surveys have been conducted in developing countries where epileptic patients remain untreated for years after the onset of seizures. The prospect of achieving remission spontaneously10 or after treatment, even in patients with prolonged epilepsy, remains high.11 In industrialised countries also untreated epilepsy may run a benign course in some patients.12 Community based studies suggest that in most patients the long term prospect of seizure control is good and that the remission rate improves with increasing duration of follow up.13
Epilepsy is certainly not a homogeneous disease, and the work of van Donselaar et al suffers from patient and seizure heterogeneity, which is a serious shortcoming. The clinical diversity of the epilepsies implies a diversity of aetiologies and cellular mechanisms. A primary generalised tonic-clonic seizure in a baby may have little in common with a secondary generalised seizure in a 16 year old. Their results are certainly not applicable to other types of epilepsies such as primary generalised absence, juvenile myoclonic epilepsies, or benign rolandic epilepsy, which carry a different prognosis.
Van Donselaar et al's study is therefore another piece of the puzzle but it does not provide a definitive answer to the old controversy: “Do seizures beget seizures?” The question is important because the answer would influence our treatment strategy after a first seizure. Further prospective studies on larger groups of more homogeneous patients are needed, and factors that contribute to the development of seizure chronicity and intractability should be identified. For the time being, it remains reasonable to delay treatment until after a second or third seizure, unless clinical features such as more than one seizure type or a neurological deficit augur poor seizure control.5
Log in using your username and password
Log in through your institution
Register for a free trial to thebmj.com to receive unlimited access to all content on thebmj.com for 14 days.
Sign up for a free trial