Diagnosis and management of epilepsy in adults
Section 2: Diagnosis

2.1 Who should make the diagnosis of epilepsy?

The diagnosis of epilepsy has important physical, psychosocial and economic implications for the patient. It is therefore important that the diagnosis is correct. It has been shown that a significant proportion of epilepsy diagnoses made by non specialists is incorrect.2, 3 Epilepsy may be difficult to diagnose in the early stages5 especially in the absence of a witnessed account. Differentiation of epileptic seizures and stereotyped behavioural phenomena can be difficult in people with a learning disability. Evidence level 2+


An epilepsy specialist has been defined as a consultant with expertise in epilepsy as demonstrated by training and continuing education in epilepsy, peer review of practice and regular audit of diagnosis. Epilepsy must be a significant part of their clinical workload (equivalent to at least one session a week).6 Evidence level 4

2.2 Classification

Classification of seizure types and epilepsy syndromes should always be attempted, as both may have implications for management and prognosis.

2.2.1 CLASSIFICATION OF EPILEPTIC SEIZURES

International classification of epileptic seizures:7

I. Partial seizures

  1. simple partial seizures (no loss of consciousness)
  2. complex partial seizures
    1. with impairment of consciousness at onset
    2. simple partial onset followed by impairment of consciousness
  3. partial seizures evolving to generalised tonic-clonic (GTC) convulsions.

II. Generalised seizures
(convulsive or nonconvulsive with bilateral discharges involving subcortical structures)

  1. absence
  2. myoclonic
  3. clonic
  4. tonic
  5. tonic-clonic
  6. atonic.

III. Unclassified epileptic seizures
(usually used when an adequate description is not available).

2.2.2 CLASSIFICATION OF EPILEPSY SYNDROMES

It is important to make the distinction between idiopathic generalised epilepsies (IGEs) and focal (localisation-related) epilepsies, as this affects treatment choices, investigation, prognosis and counselling. Identifying the aetiology is important in focal epilepsies. Evidence level 2+,4

The onset of IGEs is unusual over the age of 25.8 The most common IGEs in adolescence are juvenile myoclonic epilepsy (generalised tonic-clonic seizures with myoclonic seizures on waking, sometimes with absence seizures, with photoparoxysmal response in 30% of cases), early morning tonic-clonic seizures in adolescence, and juvenile-absence epilepsy. These phenotypes may overlap.9, 10, 11, 12, 13 Evidence level 2+,4

Features suggesting idiopathic generalised epilepsies:

Features suggesting focal epilepsies:


Tonic-clonic seizures without any focal features or any positive features of an idiopathic generalised epilepsy cannot be confidently classified.

2.3 Clinical factors and diagnosis

Attack disorders such as faint and epilepsy produce their effects because some element of physiology becomes disordered, temporarily disturbing the function of the brain. For a test to positively identify the nature of an attack disorder, an attack must be recorded, and the disturbed physiology detected. As this is usually impractical, the routine diagnosis of attack disorders is largely clinical, based on history. The history should make clear what occurred before, during and after the attack, from both patient and eyewitness points of view. A number of clinical features are common to different types of attack, so diagnosis should be based on the ensemble of the clinical features, not on single features. A generalised tonic-clonic seizure may be the presenting symptom in people with previously unrecognised epilepsy and a detailed history should be taken to uncover previous myoclonic, absence or partial seizures.10, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25 Evidence level 2+,3,4

Table 1 is a guide to how history may assist in diagnosis. Syncope is the most common attack disorder presenting to hospital emergency departments.26 Non-epileptic attack disorder may be defined as attacks which clinically resemble or may be mistaken for epileptic seizures, but which are not accompanied by any pathological disturbance of brain function, and which have a presumed or known psychological cause. Other disorders may also have to be considered (see Annex 1). Evidence level 3

Table 1: Useful clinical features in differential diagnosis of epilepsy

Note: injury, incontinence and pallor can occur in syncope, epilepsy and pseudoseizures and are poor discriminators for this differential diagnosis.

2.4 Use of EEG in the diagnosis and classification of epilepsy

2.4.1 ELECTROENCEPHALOGRAPHY

Electroencephalography (EEG) is often helpful in the diagnosis and classification of epilepsy.10, 27 However, it is essential to understand the scope and limitations of the technique when requesting an EEG examination and subsequently evaluating an expert report on the recording.2 Non-specific EEG abnormalities are relatively common, especially in the elderly, patients with migraine, psychotic illness and psychotropic medication. Non-specific abnormalities should not be interpreted as supporting a diagnosis of epilepsy. Evidence level 2+

A normal EEG does not exclude a diagnosis of epilepsy. A single routine EEG recording will show definite epileptiform abnormalities in 29-38% of adults who have epilepsy. With repeat recordings this rises to 69-77%.10, 28, 29, 30, 31 The sensitivity is improved by performing an EEG soon after a seizure, and by recordings with sleep or following sleep deprivation.10, 32, 33 Evidence level 2+

Incidental epileptiform abnormalities are found in 0.5% of healthy young adults, but are more likely in people with learning disability and psychiatric disorders, patients with previous neurological insult (e.g. head injury, meningitis, stroke, cerebral palsy) and patients who have undergone neurosurgery.34, 35, 36 Evidence level 2+

In a patient in whom the clinical history suggests an epileptic seizure but is not conclusive, the prevalence of epilepsy will be high. The finding of epileptiform abnormalities is specific, and the diagnostic value of the test is good. In a patient in whom the history is typical of some other disorder, such as syncope, the prevalence of epilepsy will be low, and any epileptiform abnormalities are more likely to be incidental. The test should not be performed in this circumstance. Evidence level 2+

EEG can aid classification of epileptic seizures and epilepsy syndromes. The finding or not of a photoparoxysmal response can allow appropriate advice to be given.37 If performed within the first few weeks after a first seizure, EEG has prognostic value; patients with epileptiform abnormalities are more likely to have a second attack.38 Evidence level 2+

2.4.2 LONG TERM EEG MONITORING

When clinical information and standard investigations do not allow a confident diagnosis, referral for the recording of attacks should be considered. The attacks should usually be occurring at least once a week. If non-epileptic attack disorder is a possibility, then monitoring in patients with less frequent attacks may be worthwhile, as they are often easily provoked.39, 40, 41, 42, 43, 44, 45, 46 The best method is inpatient video EEG recording.47 Twenty four hour ambulatory EEG recording has the advantage of recording attacks in the patient’s usual setting, but does not usually allow correlation of EEG and video data. Home video recordings can be useful, but rarely capture the onset of attacks.47, 48 These investigations should include single channel electrocardiography recording.49 Evidence level 2+,3,4

2.4.3 OTHER INVESTIGATIONS

Electrocardiography (ECG) should be performed in the assessment of all patients with altered consciousness, particularly those in older age-groups, when disorders of cardiac rhythm may simulate epilepsy. Twenty four hour ambulatory ECG and other cardiovascular tests may also be helpful.49 Evidence level 3






2.5 Brain imaging

Brain imaging detects lesions in 21-37% of patients presenting with epilepsy. Such lesions require treatment in only a small minority, but their detection may have implications for future management should the epilepsy become intractable. Idiopathic generalised epilepsies are not associated with an increased prevalence of brain lesions.10, 50, 51, 52, 53 Evidence level 2+

2.5.1 MAGNETIC RESONANCE IMAGING

Magnetic resonance imaging (MRI) scanning is the current standard of reference in the investigation of patients with epilepsy. Routine MRI brain scanning using simple standard sequences will detect lesions (eg small tumours, vascular malformations and cortical dysplasia) that are not detected by computed tomography (CT) scanning. MRI carried out for the assessment of drug-resistant epilepsy requires specialised protocols and expertise (eg to detect hippocampal sclerosis).53, 54, 55, 56, 57, 58, 59, 60, 61 Evidence level 2++,2+,4

2.5.2 COMPUTED TOMOGRAPHY SCANNING

CT scanning has a role in the urgent assessment of seizures, or when MRI is contraindicated (eg when patients have pacemakers or metallic implants). A non-contrast CT scan will fail to identify some vascular lesions and tumours. CT has only a limited role in the assessment of intractable epilepsy.54, 61, 62 Evidence level 2++,2+



[Contents][Back][Top] [Next]

Scottish Intercollegiate Guidelines Network
Home