Understanding epilepsy is a necessary part of the management of individuals with cerebral palsy.⁴⁵ Epilepsy is common in all types of cerebral palsy and, in some varieties, it is the rule.

The definitions of epileptic seizures, symptomatic epileptic seizures, and epilepsy are the same as those used in other chapters throughout this book. The definition of cerebral palsy (CP) is a little more difficult, insofar as some variation occurs in the way in which the term is used between different authors and different studies. The most frequent definition of cerebral palsy is “a disorder of posture and movement due to a static lesion of the developing brain.” Most authors include cerebral malformations as examples of static lesions of the brain but, in at least one study, cerebral malformations were excluded.²⁰ Many authors include the aftereffects of acute cerebral insults and injuries in infancy and even in early childhood, whereas others would exclude from the CP definition any condition presumed not to have been present before the age of 4 weeks.²⁰ There is considerable force in the argument for limiting the definition of cerebral palsy in this way, insofar as the prognosis with respect to epilepsy differs between those whose brain lesion is of perinatal or prenatal origin and those children with unequivocally postnatal cerebral insults.

A new definition of cerebral palsy was proposed in 2005⁵: “Cerebral palsy (CP) describes a group of disorders of the development of movement and posture, causing activity limitation, that are attributed to nonprogressive disturbances that occurred in the developing fetal or infant brain. The motor disorders of cerebral palsy are often accompanied by disturbances of sensation, cognition, communication, perception, and/or behavior, and/or by a seizure disorder.” From the point of view of this chapter, it is disappointing to see the term “seizure disorder” employed. This sloppy and ambiguous term deserves to be deleted from scientific discourse.³⁸

Authors have exhibited some variability on the definition of the various subtypes of CP but, on the whole, these differences in classification do not materially affect the understanding of epilepsy in the context of CP. The main divisions in the Edinburgh classification were²³: (a) hemiplegia, (b) bilateral hemiplegia, (c) diplegia (which might vary from virtually paraplegic to tetraplegic), (d) ataxic CP (including ataxic diplegia and ataxia), (e) dyskinesia (including dystonia and athetosis), and (f) any other form of CP, including mixed forms. Many authors now include ataxic diplegia within the diplegia categories. Certainly, for the purposes of discussion of epilepsy, dystonic CP should be strongly distinguished from athetoid CP, insofar as the liability to epilepsy may be high in the former and low in the latter type. The group described by the Edinburgh School as bilateral hemiplegia will have severe mental retardation, as discussed in the section Risk Factors for Epilepsy.¹³

Recent studies^31,37 of the etiology of CP (with or without epilepsy) have shown a much higher diagnostic yield than was thought likely in the past. This should make one wary of the CP label without firm foundation.

Many conditions can masquerade as cerebral palsy but turn out to be something else.^21,40 Most such conditions are individually rare, but parents do not commonly judge the rarity of a condition as a justification for pediatricians or neurologists to make an incorrect diagnosis. Fortunately for the present issue, cerebral palsy imitators complicated by epilepsy or epileptiform attacks are few. Pelizaeus-Merzbacher disease is one such confusing condition.⁶ One might argue that this is actually an example of cerebral palsy but, of course, the important point is that it is an X-linked disorder and may recur in a future male child.

Another condition with an unequivocally progressive cerebral pathology is the Aicardi-Goutières syndrome.³⁹ Affected children may behave much like patients with cerebral palsy without regression.³⁹ Confirmation of the diagnosis of Aicardi-Goutières syndrome depends on finding elevated cerebrospinal fluid (CSF) and serum α-interferon in the absence of any congenital viral infection.

Another confusing condition is hyperekplexia⁴⁷ in which affected neonates may be very stiff and troubled by severe nonepileptic convulsions. When a family history of dominantly inherited startle disease is present, the diagnosis presents no difficulty, but in sporadic cases it can be difficult.

Rare, potentially treatable inborn errors of metabolism leading to cerebral palsy and epilepsy have been described. These include defective serine biosynthesis²⁴ deficiency,⁴ and GAMT deficiency.^28a

New conditions will continue to be recognized, and it behooves the child neurologist to question the diagnosis in every patient with “cerebral palsy.” This is particularly so in so-called ataxic cerebral palsy, which some have suggested should not be called cerebral palsy at all,²⁶ insofar as genetic etiologies abound.

Having cerebral palsy is no insurance against exhibiting or suffering from the various nonepileptic attacks that may affect the general population.³⁸ We suspect that children with cerebral palsy are more at risk of having epilepsy misdiagnosed in this way because epilepsy is common in cerebral palsy and so expected to be seen. For example, a young child with spastic diplegia may have what we term reflex anoxic seizures³⁸ after falling over and bumping his head just like anyone else who is similarly genetically predisposed. In addition, patients
with cerebral palsy may have a predilection for various movements having a superficial resemblance to epileptic seizures. In all these situations, keen clinical judgment must be used. The interictal electroencephalogram (EEG) cannot be relied on to assist in the differential diagnosis. Although it has been attested that spikes or epileptiform discharges are more common in children with cerebral palsy who have epileptic seizures, cerebral palsy individuals who have never had epileptic seizures may not uncommonly show spike discharges also.³⁵ When an individual with cerebral palsy has had one or more previous epileptic seizures, it is even less rational to expect that an EEG examination will help to diagnose a new unexplained paroxysmal event.³⁸ On the other side of the coin, those with startle epilepsy³⁶ may not show spikes at all in the ictal EEG; however, such individuals always have other epileptic seizures without the startle provocation.²⁸

The early studies of Ellenberg and Nelson¹⁴ indicated that knowledge of the etiology of epilepsy not associated with cerebral palsy was very limited. A number of later studies have addressed such a question in various ways.

Goulden et al.²⁰ studied a cohort of mentally retarded individuals born in Aberdeen, Scotland, between 1951 and 1955. By 22 years, 15% of these had epilepsy. With their definition of cerebral palsy as “having a presumed prenatal or perinatal onset,” the cumulative risk for epilepsy was 28%, 31%, and 38% at 5, 10, and 22 years of age, respectively. This compared with a much lower risk for individuals with mental retardation and no associated disabilities, and a much higher risk of epilepsy in those with postnatal brain injury (defined as a significant brain insult after 28 days of life which might reasonably account for the child’s later functioning). This particular study included a separate group with cerebral malformations but, in fact, the only malformation determined was described as an occipital meningocele.

A careful Italian study⁹ focused on the risk factors for the co-occurrence of partial epilepsy, cerebral palsy, and mental retardation. In a studied population of 64 children with these three conditions, neuroimaging identified 32 with cerebral malformations and 32 without but with encephalomalacia, periventricular leukomalacia, or diffuse atrophy. These two groups were compared with a much larger population of normal children. The definitions in this study were as follows: