In other reports, phenobarbital aggravated absence seizures in children (Acta Scand, 1996; 94: 367-77), and induced tonic seizures (muscle tightening) in those with Lennox-Gastaut syndrome, a severe childhood epilepsy with multiple types of seizures (Epilepsia, 1972; 13: 421-35). Most recently, other AEDs - carbamazepine, tiagabine and valproate - have been implicated in aggravating seizures because the prescriptions were too high. In the case of valproate, worsening seizures may also indicate the presence of severe liver or brain toxicity - even when the patient doesn’t show high levels of the AED in the blood.

Drugs can also worsen seizures by ‘selective aggravation’, when the wrong AED is given for the type of epilepsy. So far, the vast majority of these cases involve carbamazepine and children with generalised seizures.

Carbamazepine can cause and aggravate a vast array of seizures, including absence, atonic, tonic and myoclonic seizures. A severe consequence is that the patient is thrown into a state of perpetual seizure.

The sheer range of individuals who do poorly with carbamazepine is frightening. In one study, 59 epileptic children were given this drug and followed-up with regular electroencephalograms (EEGs). Thirty-three of them did well and no one needed to be taken off the drug. But 26 of the children (46 per cent) began having seizures or their seizures got worse. Nearly half of them had to stop taking the drug, and most of those who continued taking it needed to use other AEDs as well (Epilepsia, 1994; 35: 1154-9).

In another study of 170 children with juvenile myoclonic epilepsy (JME), 40 received carbamazepine or phenytoin early in their treatment. Nearly 57 per cent of these children had their symptoms worsen with these drugs. JME usually responds best to valproic acid (from which valproate is derived) and is the drug of choice for this condition (Neurology, 2000; 55: 1115-21).

The latest ‘wonder’ drug Adverse reactions are not limited to the older drugs. Most recently, lamotrigine has been hailed as a treatment with fewer adverse effects than other drugs and a good record of effectiveness. But it has now come to light that the drug may actually eliminate the ‘warning signals’ associated with seizures. Many people who have epileptic episodes get a warning feeling - an ‘aura’ - that a seizure is coming. This warning, lasting from a few seconds to minutes, that an epileptic seizure may kick off shortly, gives the sufferer time to avoid injury by moving out of the way of potential hazards such as objects they could collide with, stairs or motor vehicles.

Without such a warning, people with epilepsy could suddenly and unexpectedly fall and injure themselves. Indeed, several patients who’d lost their auras once they’d switched to lamotrigine suffered injuries during seizures - some serious (Lancet, 1997; 350: 1751-2).

In addition to removing a vital warning signal, lamotrigine also aggravates seizures. Although some children with myoclonic epilepsy appear to benefit from this treatment, the drug may fail to control, or may even aggravate, this type of seizure (Neurology, 2000; 55: 1758). Lamotrigine can also worsen seizures when used as an add-on drug (Epilepsia, 1995; 36 [Suppl 3]: S65).

For babies with severe myoclonic epilepsy of infancy, in which spasms may occur in one or more groups of muscles separately or synchronously about the body, lamotrigine has simply served to make their epilepsy worse. In a study of 21 infants with this rare disorder, the vast majority showed worsening of at least one type of seizure after add-on therapy with lamotrigine. Eventually, all but two of these infants discontinued the drug - and all but one of them improved on stopping the drug. Because lamotrigine’s dose needs to be increased in steps, deterioration of the patient’s condition developed insidiously, so much so that, in some cases, it was not recognised by the treating physicians (Epilepsia, 1998; 39: 508-12).