Intended for healthcare professionals


NICE guidance on newer drugs for epilepsy in adults

BMJ 2004; 328 doi: (Published 27 May 2004) Cite this as: BMJ 2004;328:1273
  1. Emilio Perucca, professor (perucca{at}
  1. Clinical Pharmacology Unit, University of Pavia, Piazza Botta 10, I-27100 Pavia, Italy

    Individualisation of therapy remains the key to successful treatment

    With the introduction of felbamate, gabapentin, lamotrigine, levetiracetam, oxcarbazepine, tiagabine, topiramate, vigabatrin, and zonisamide, the number of antiepileptic drugs marketed in different parts of the world has virtually trebled between 1989 and 2002.1 The use of newer antiepileptic drugs is increasing steadily. In 2002 new antiepileptic drugs accounted for 20% of total prescriptions and for 69% of total costs for antiepileptic drugs in the United Kingdom (£99m of £142m).2 Is the shift towards new antiepileptic drugs justified, and what are the indications for these drugs in the modern treatment of epilepsy? These questions are addressed in the latest guidance from the National Institute for Clinical Excellence (NICE).2

    The guidance, which is based on data from randomised trials, other published information, and feedback from professional, specialist, patient, and carer groups and drug manufacturers, focuses on the place of new antiepileptic drugs in adults. Zonisamide, not yet available in Europe, and felbamate—a last resort agent because of toxicity to bone marrow and liver—were not considered, but this does not affect the general applicability of the document. The conclusions reached are: firstly, monotherapy data in newly diagnosed patients do not show the existence of differences in effectiveness between newer and older antiepileptic drugs; secondly, although side effect profiles and interaction potential differ among drugs, adequate evidence does not exist to support the claim that newer drugs are generally associated with improved quality of life; thirdly, integrated cost effectiveness analysis shows a high degree of uncertainty about the costs and benefits associated with individual drugs; fourthly, seizure freedom, the most important outcome, is infrequently achieved during combination therapy in patients refractory to monotherapy, and insufficient evidence exists to determine whether any one of the newer antiepileptic drugs is superior to others in providing long term freedom from seizures. Overall, these conclusions agree with those reached in previous publications, including Cochrane reviews.1 3 Some of the studies showing improved outcome with newer antiepileptic drugs may in fact have been flawed by bias in study design or analysis.4

    The guidance recommends that people with a first seizure be referred to a specialist for accurate diagnosis. If treatment is indicated (which is normally the case after two seizures have occurred), this should be initiated with a single drug, and combination therapy should be attempted only when at least two sequentially tried individual drugs have failed. Treatment should be monitored regularly and revert ultimately to the regimen (monotherapy or polytherapy) that proved most acceptable in terms of seizure control and side effects. As for the place of newer antiepileptic drugs, the guidance recommends that these be reserved for people who have not benefited from older agents such as carbamazepine or valproate. However, new antiepileptic drugs may be used as first line treatments when older antiepileptic drugs are contraindicated or are already known to be poorly tolerated by the individual, or when they are expected to interact with other drugs the patient is taking (most notably oral contraceptives) or when the patient is a woman of childbearing potential. In the latter situation, the guidance highlights the need for counselling on contraception (phenytoin, carbamazepine, barbiturates, oxcarbazepine, and felbamate reduce the efficacy of the contraceptive pill, whereas valproate, lamotrigine, gabapentin, lamotrigine, levetiracetam, tiagabine, vigabatrin, zonisamide and, at doses up to 200 mg/day, topiramate, do not5) and risks for the unborn child, specific caution being given against valproate because of risks for the unborn child.

    The recommendations for women of childbearing potential are the most innovative. If a woman is taking (or planning to take) an oral contraceptive it makes sense to avoid enzyme inducers such as carbamazepine. The specific warning against valproate is stronger than in earlier guidelines,67 but may be justified by recent reports implying a greater risk of fetal malformations after prenatal exposure to valproate compared with monotherapy with other antiepileptic drugs.812 The latter findings, however, cannot be regarded as conclusive because of limited power in some studies and possible influence of confounders, and therefore evidence on which to base recommendations is far from ideal. Presumably, NICE also had to take into account the summary of product characteristics for valproate in the United Kingdom, which says that valproate should not be given to women of childbearing potential without specialist neurological advice and that for partial seizures it should be used only in women resistant to other treatments.

    The NICE guidance is likely to have little impact on the trend favouring wider use of the newer antiepileptic drugs. As with all guidance, it will need to be adapted to local settings (including scarce resources in many countries), and it cannot substitute sound clinical judgment in individual cases. The guidance does not address the role of individual newer antiepileptic drugs, which differ in their tolerability profile and efficacy spectrum in specific epilepsy syndromes. The key to success in treating epilepsy is tailoring a drug's properties to the patient's characteristics,1 7 and in this respect newer antiepileptic drugs are welcome because they widen the range of options. The guidance is due for revision in 2006, when hopefully new evidence will be available from important studies, including pregnancy registries, and large randomised trials of newer and old antiepileptic drugs in newly diagnosed epilepsy (SANAD)2 and in seizures with onset in old age (VA cooperative study #428).13


    • Competing interests EP has received research grants and fees from the manufacturers of carbamazepine and oxcarbazepine (Novartis), gabapentin and phenytoin (Pfizer), lamotrigine (GSK), levetiracetam (UCB Pharma), topiramate (Johnson and Johnson), valproate (Sanofi Synthelabo) and vigabatrin (Aventis).


    View Abstract