Epilepsy is an ancient disease that is already mentioned in the Babylonian medical treatise that dates back to 1067-1046 B.C., which calls it miqtu (disease that makes one fall) and accurately describes its main clinical expressions (Fig.1). In time epilepsy was designated with more or less imaginative names that affected the attitude of the culture that created them. Descriptive terms such as those in Babylonian tablets or the medieval and renaissance culture’s ‘falling sickness’ are alternated with definitions such as ‘sacred disease’ in Greece, referring to the disorder’s supposed supernatural origin.

The medical term ‘epilepsy’ carries traces of this mythical vision: from the Greek verb epilambáno (take, invade, seize) that recalls the image of possession. It would on the other hand be wrong to think that a long-standing past ruled by prejudice was at last surpassed already in the 5th century B.C. only thanks to the progress of modern science. Hippocrates passionately contested the prejudice in the following terms: “It” (epilepsy) “is not in my opinion anything more divine or more sacred than other diseases; it has the same nature as the others”.

Besides scientific progress has not abolished the prejudice and stigma attached to this disease that still rouses a superstitious fear in many, causing unnecessary pain to those who suffer from it. Scientific and humanitarian associations throughout the world are fighting against the prejudice and superstition with information campaigns; worth mention is the Global Campaign against Epilepsy organized and conducted by the International League Against Epilepsy (federation of national scientific societies of 86 countries, Fig. 2) in collaboration with the International Bureau for Epilepsy (federation of voluntary service associations) and the World Health Organization. Modern medicine considers epilepsy a condition that is characterized by the presence of fits (crises) that are repeated apparently spontaneously in time.
Its physiological and pathological basis is the persistent presence of too excitable nerve cells (neurones), which generate the occasional epileptic impulses. It has been calculated on a prevalence basis (from 6-8% in industrialized countries to almost twice this in developing countries) that 50 million people in the world suffer from epilepsy.

CLASSIFICATION AND CLINICAL FORMS
Diverse clinical epileptic forms are distinguished by aetiology, clinical symptoms and prognosis. If the epileptogenic neurone impulses begin and remain localized in a limited neurone population, crises are defined partial and their symptoms are consistent with the functions of the special cortical areas involved. If the impulses begin locally but spread more or less rapidly to vast cortical areas, there can be a secondary generalization (often with seizures).
The originally generalized crises are instead supported by a series of epileptic impulses that from the start simultaneously involve wide cortical areas in both hemispheres. Many of the crises described as generalized (mainly convulsive) are in practice partial crises generated by local impulses that spread so rapidly as not to enable the recognition of the partial initial phenomena.
Table 1 reports the classification of epileptic crises processed by the International League Against Epilepsy’s specially appointed commission. Concomitant electroencephalographic alterations (EEG, and critical EEG) play an essential role in the classification of crises and the specific form of epilepsy.
They differ greatly in the many types of crises (refer Fig. 3 and Fig.4).
Correlations between clinical symptoms and related EEG can be accurately analysed in a combined video-EEG recording with concomitant simple tests to assess intra and post-critical deficits. The diagnosis of a specific form of epilepsy is based on the type (or on the association of diverse types) of crises the individual presents and also on the intercritical EEG chart’s characteristics, the age of symptom onset, the presence-absence of clinical or radiological signs of damage to the central nervous system, the presence-absence of recognizable causes and the family history. On the basis of these elements one can formulate a diagnosis of the epileptic syndrome, an essential premise to programme further investigations targeted at defining its aetiology and planning medication.
Table 2 refers a simplified version of the classification of epileptic syndromes processed by the International League Against Epilepsy. Idiopathic epilepsies are characterized by: an age-related beginning, normal psychological and motor development and absence of cerebral damage. The study of the EEG anomaly patterns while awake and asleep is essential towards the diagnosis and hence towards the prognosis of the many forms. The individual forms’ clinical features are reported in detail in the studies mentioned in the references.
Symptomatic/cryptogenic epilepsies represent the majority of epilepsies that are protracted in time and which tend to resist medication. They can begin at any age. They include both forms whose cerebral lesions can be spotted and diagnosed and cases where these can only be theorized. Crises can be monomorphous or polymorphous; the symptoms and EEG picture depend on epileptogenic foci’s site. Epilepsy of the mesial temporal lobe holds a special place among partial forms; it presents a typical biphasic pattern with an acute initial episode (often a protracted febrile seizure) followed, after a more or less prolonged latent period of even many years, by a chronic phase marked by partial crises that are often refractory to pharmacological treatment.
This form is one of the major problems in adult epileptology due to its frequency and difficulty to treat. In most patients generalized forms begin during infancy.
They can start without known causes in previously normal children (cryptogenic forms) or in the presence of elements that give evidence of a past cerebral damage of known or unknown nature (symptomatic forms). This chapter discusses certain infantile epileptic encephalopathies having serious prognoses both from a crisis control viewpoint and that of somatic and psychic development. Crises that set in while the temperature is rising in neurologically healthy patients below five years of age are generally defined febrile seizures. Hence we must distinguish between real epileptic fits that can be facilitated by hyperthermia in epileptic patients and those carrying cerebral damage.

AETIOPATHOGENESIS
The study of experimental epileptic models has highlighted the epileptogenic potential of altered neurone excitability mechanisms, which depend on ionic flows through cell membrane channels. A great progress in knowledge has resulted from the recent definition of channel molecular structure and the genes that encode the proteins that form it (Fig. 5).
Today we know with high precision the consequences of structural modifications of channels on their control of ionic flows that are regulated by varying the membrane’s electric potential (voltage-dependent channels) or by means of neurotransmitters (receptor associated channels).
These studies have greatly furthered the understanding of mechanisms that generate epilepsies and opened new treatment perspectives. Epileptic crises can be triggered by dysfunctions in neurone populations with no morphological alterations (idiopathic epilepsies). However, in a large number of cases crises are subsequent to recent or previously existing cerebral damage that has been established (symptomatic epilepsies) or presumed (cryptogenic epilepsies).
The evolution of neurological and radiological diagnostics and the extensive use of MRI have enabled to spot misunderstood cerebral damage in a considerable number of patients. An important example is the identification of localized malformations such as focal cortical displasias (Fig.6).
To be effective, image testing must be targeted at a specific question formulated on the basis of the epileptic form’s clinical assessment. Image functional tests such as positrone emission tomography (PET), functional MRI and spectroscopy are a further development.
They have important applications in selected cases (i.e. patients with focal epilepsy who are candidates for surgery). We must number malformations (displasias, lissencephalia, phacomatosis, vascular malformations), foetal and perinatal encephalopathies on an anoxic and haemorrhagic background, infectious and post-traumatic encephalopathies, vasculopathies (vasculitis, embolic events), primitive or secondary tumours of the nervous system, chromosome diseases (trisomy 18 syndrome, Down, Angelmann and Prader-Willy syndrome) and genetically caused progressive encephalopathies (mitochondrial encephalopathies, organic acidurias, aminoacidopathies and peroxisomial diseases, besides those reported in table 2) among the most important causes of symptomatic epilepsies. No cerebral damage can be proved in idiopathic epilepsies; a genetic origin can be proved or is highly probable. So far the mutations responsible have been spotted only in rare forms transmitted by the dominant gene; these are expressed with repeated seizures during the precocious neonatal or infantile period with febrile seizures or partial crises that persist in adulthood.
Though they concern only a limited population of patients, they are of great interest as they involve genes that encode by subunits of voltage-dependent ionic channels or that are associated to receptors. Mutations spotted concern muscarinic receptors in familial forms of frontal epilepsy, K+ channels during benign neonatal familial crises, Na+ channels in a special form of generalized epilepsy with febrile crises and GABA receptors in one of its variations and in a sub-form of myoclonic epilepsy in youth. Concerning the most common forms of idiopathic epilepsy, their molecular characterization has not been achieved and it is possible that diverse mutations can determine similar phenotypes in various families or geographical areas.
Concerning symptomatic epilepsies the epileptogenic dysfunction is caused by alterations the lesion induces on nerve cells in surrounding tissue (as in tumours) or nerve cells present in the pathological tissue (as in displasias).
Experimental data and circumstantial evidence in human symptomatic/cryptogenic epilepsies suggest that epileptic activity can alone induce alterations in the cerebral tissue involved, causing a progressive development for the worse of the epileptogenic area till it becomes immune to medications.
A typical example of the epileptogenic process’ development potential is the aforementioned epilepsy of the mesial temporal lobe. The patient’s age and his clinical history must always direct diagnostic investigations. In fact, the great majority of static antenatal o perinatal epileptogenic cerebral damage is often expressed with crises that begin precociously; many “progressive” epileptogenic encephalopathies begin in a relatively well-defined age group (infancy or youth).
On the contrary, some acquired cerebral lesions (such as tumours) can more often be recognized in adults. However epilepsies that begin quite late (second-third decade of life) can be caused by gene mutations or lesions such as focal structural malformations or vascular malformations present since birth.

MEDICATION
Epileptic treatment has been enriched in recent years by introducing newly formulated drugs studied to act specifically on the epileptogenic mechanisms highlighted by basic studies.
Three of them in particular (oxcarbazepine, gabapentin and lamotrigine) inhibit Na+ flow; two strengthen GABA-mediated neurotransmission (vigabatrin and tiagabine); one reduces Ca2+ flow levetiracetam; and, two (topiramate and felbamate) have multiple actions on Na+ flow and GABA and Glutamate-mediated neurotransmission.
The new drugs are useful in all cases that are refractory or intolerant to traditional drugs. Generally speaking there is no proof that these drugs are more powerful than those already used for decades, which are hence preferred as frontline treatment because better known through long practical experience.
The objective of treatment with antiepileptic drugs is to stop a crisis or, when this is not possible, to limit their number and gravity, avoiding the onset of side effects. A premise for medication is to check a crisis’ tendency to repeat itself in time, in other words to make sure it is epilepsy and not an occasional crisis.
Once the diagnosis of epilepsy has been made, it is essential to correctly diagnose the syndrome because it influences the choice of drug and also because the prognosis is important towards the therapeutic course. As a general rule it is best to begin after not more than two or three crises.
A waiting strategy can be followed in partial idiopathic infantile forms (especially in forms with rare crises and especially at night), whose tendency to spontaneous remission is certain. The timely beginning is particularly important in forms of cryptogenic and symptomatic epilepsy where it is important to contrast the tendency to develop towards progressive refractoriness.
This context does not permit us to study each drug’s specific indications in detail for the many forms of epilepsy.
Generally the first choice drug in idiopathic epilepsies is valproate; carbamazepine can be used in partial forms; ethosuximide, lamotrigine and phenobarbital in relatively rare cases when valproate fails.
Carbamazepine, phenitoine and vigabatrin are not advised in generalized idiopathic cases. In symptomatic/cryptogenic epilepsies with partial crises the first choice is carbamazepine and all other drugs can be used in relatively frequent non-responsive cases (about 50%).
The treatment of generalized infantile symptomatic/cryptogenic forms is still more difficult; they require specialized skills. I only wish to mention that vigabatrin is specially recommended in West’s syndrome, though it is often necessary to resort to steroids.
As a rule, medication must begin with only one drug and be maintained with a very simple schedule. The suspension of treatment can be considered after a period of total absence of crises that can vary from two to five years, depending on what is known about the natural history of the specific form of epilepsy in question and a careful assessment of risk factors for a possible recurrence. As a rule an epileptic crisis does not require urgent drug treatment.
The problem arises only when crises follow at a short distance or are prolonged, thus presenting an epileptic condition. The generalized convulsive condition that can rapidly compromise the patient’s vital functions is particularly serious. Frontline drugs are benzodiazepines, recommended for all epileptic disease types. If these are not effective, an intravenous loading dose of phenitoine can be used or valproate can be administered intravenously if it is a generalized form. If the emergency does not rapidly recede with antiepileptic drugs, the patient must be transferred to the intensive care unit and anaesthetic drugs or gas must be resorted to. Drug kinetic parameters reported in the table must be particularly considered when defining a therapeutic pattern and especially the data relevant during the metabolism-excretion stage, which is the most common moment of drug kinetic interaction between antiepileptic drugs and other drugs, monitoring, if necessary, plasma levels on precise indications on the final balance between absorption and excretion.
Monitoring is necessary in all physiological conditions (as in pregnancy) or pathological ones that can interfere on absorption, distribution and removal (i.e. association of a drug with a high probability of interaction, serious intercurrent illnesses).
Plasma level monitoring must be adjusted to the specific clinical conditions. As only the free drug molecules (in other words those not bound to plasma proteins) pass the haematoencephalic barrier, becoming active in the nervous system, it may at times be necessary to also monitor the free molecules in specialized laboratories.
Therapeutic decisions in women can present particular problems. Antiepileptic treatment can reduce the effectiveness of contraceptives as antiepileptic drugs that induce hepatic enzymes can accelerate the catabolism of estrogen-progestine drugs. The increased risk of crises during menstruations in some forms of epilepsy (catamenial epilepsy) can make drug protection periodically inadequate.
Pregnancy must be carefully followed both in the pharmacological and gynaecological spheres in women treated with antiepileptic drugs. Some drugs’ metabolism (primidone and carbamazepine) can change in pregnancy with the subsequent onset of side effects due to increased plasma levels that must hence be monitored regularly. Some studies have noticed a moderate and general increase in malformations in those born from patients under antiepileptic treatment.
The greatest suspicions concern valproate, whose administration seems related with a greater incidence of anomalies in midline structures (variable from the bifid spine to anencephalia), when compared to the population at large.
There are no reports of specific teratogenesis concerning recently introduced drugs; however the scarcity of information available on human case studies advises against their use. In any case the therapeutic regime must be rationalized before pregnancy begins, checking that plasma levels are not too high and there are no peaks of concentration during the day (this is especially valid for valproate) and preventively supplementing folic acid 4-5 mg/day.
The prophylactic use of antiepileptic drugs in the presence of crises triggering risk factors (in particular head injuries) did not yield satisfactory results.
The advisability of drug prophylaxis for febrile seizures after a first episode is debated; experts instead agree on recommending impromptu rectal administration of benzodiazepine in the presence of protracted febrile seizures that last over 15 minutes.
Side effects involving the nervous system have been reported for all antiepileptic drugs (i.e. drowsiness, mood changes, vertigo, ataxia etc.) and gastrointestinal side effects too. Generally their intensity is proportionate to the plasma dosage/level of antiepileptic drugs (dose-dependent side effects). A special effect typical of vigabatrin is a concentric reduction in the visual field caused by an effect of the drug on the retina; this is frequent but usually it is not subjectively noticed.
Besides, in susceptible individuals every drug can potentially give rise to idiosyncratic reactions even with very low plasma dosages/levels (dose-independent side effects).
Among the most important reactions typical of antiepileptic drugs we must mention the rare but serious skin reactions (Stevens-Johnson and Lyell’s syndrome) caused by carbamazepine, diphenylhydantoin, oxcarbazepine, barbiturates and lamotrigine; valproate and felbamate hepatotoxicity, which is also rare, and haematological reactions that can occur exceptionally with all drugs, but are particularly feared in patients treated with felbamate.
A satisfactory crisis control with a minimum incidence of side effects is achieved in 70% of cases; the remaining 30% of patients is immune to medication.
Attempts to transfer experimental experiences that imply the possibility of using antiepileptic drugs to prevent progress towards an immune condition in certain symptomatic/cryptogenic epilepsies to a clinical field have so far been disappointing. This is a topic of great importance that is a challenge for epileptological studies in the next years. However there already exist other types of medical and surgical treatment, which must be considered the solution for patients immune to antiepileptic medication.

OTHER MEDICAL TREATMENT
ACTH and cortisone based drugs: these are recommended in West’s syndrome and, more rarely, in serious forms of precocious infantile epilepsy that is non-responsive to antiepileptic drugs. Vitamin B6: it is primarily recommended in the treatment of rare pyridoxine-dependent epilepsies with a precocious infantile onset. Ketogenic diet: it is based on the administration of a diet rich in fatty acids and poor in carbohydrates and proteins (with a 4:1 ratio) that leads to an increase in blood ketones. It has proved effective in individuals with serious drug-resistant epilepsies especially in infancy.
Immunoglobulins: data on effectiveness is inconsistent in serious, drug-resistant forms and with a high critical frequency. Plasmaphaeresis: it has been used in patients struck by partial constant epilepsy associated with typical or atypical Rasmussen’s encephalitis. It aims at removing fractions of antibodies that may have a cytotoxic or excitotoxic effect (anti-glutamate receptor antibodies).

SURGICAL TREATMENT
In symptomatic epilepsies caused by cerebral lesions, which alone present the indications for surgery (developmental lesions or vascular malformations with a risk of bleeding), this is generally targeted at removing the lesion and not at a possible fading of critical symptoms. In epilepsies with focal crises associated with non-developmental cerebral damage or in which no cerebral damage can be detected, surgery, targeted at controlling crises by surgically removing the epileptogenic area can be considered when the following premises are present:
1) crises have proved resistant to drugs recommended for partial epilepsies;
2) there is the certainty (or a high probability) that all crises begin in the same cerebral area (epileptogenic area);
3) it can be foreseen that the epileptogenic area’s removal will not cause significant neurological deficits.
These premises must be rigorously checked through a targeted clinical, neurophysiological, neuropsychological and neuro radiological study.
The clinical, neurophysiological study must comprise the video recording of crises through prolonged EEGs to enable their exact topographical definition. When all the clinical data, EEG, neuroradiological and psychological data do not solve the question of the precise site of origin or present the suspicion of multiple foci, an in depth exploration with electrodes must be performed in highly qualified structures.
The results of surgical treatment for epilepsy are excellent with these premises and can enable the complete control of critical symptoms in most patients with partial epilepsies and those that are immune to medication.
Hemispherectomy can be considered in certain cases of infantile epilepsy associated with serious contralateral neurological deficits (hemiplegia or serious hemiparesis) of the damaged hemisphere. The “palliative” method based on partial helotomy and targeted not at stopping epileptic impulses but in limiting them in individuals suffering from very serious epilepsies must be considered with greater caution.
The following palliative indications for surgery can be considered in a limited number of immune patients who cannot be candidates for the choice surgical treatment: helotomy, intracerebral or nervus vagus stimulation techniques.

CONCLUSIONS
In few fields of medicine the integration between basic science and clinical experience has been so close and effective as in epileptology. Thanks to the careful observation of clinical signs and symptoms and the progress of neurophysiology, molecular biology and pharmacology, neurologists today have powerful pharmacological and surgical weapons to effectively solve problems in most patients.
However the problem concerning the development of new antiepileptic drugs to control crises that are immune to medication and, generally speaking, the study of new “antiepileptogenic” strategies to prevent the epileptogenic process’ establishment and to change its course remains open.
This is what we expect from future studies, stressing that we will not be satisfied with the progress of medicine if at the same time there is no change in prejudice and exclusion dynamics that still weigh on people who suffer from epilepsy.

Translated by Interpres sas

Giuliano Avanzini
Direttore Dipartimento di Neuroscieze cliniche Istituto Nazionale Neurologico C.Besta, Milano
Presidente della International League Against Epilepsy (ILAE)