Epilepsy that medication does not control

There is one figure that puts this whole subject in order, and almost nobody tells patients about it.

In a study that followed 525 people from diagnosis, 47 per cent became free of seizures on the first drug. Another 14 per cent got there with a second or a third. A further 3 per cent with two drugs combined. In all, around 63 per cent were controlled.

And this is the figure that changes the conversation: among those in whom the first drug failed for lack of effect, only 11 per cent were controlled afterwards. Carrying on trying drugs, beyond a certain point, has a low and measured yield.

What "drug-resistant" means

An epilepsy is regarded as drug-resistant when two appropriate medicines, well chosen and used at adequate doses, have failed to control the seizures. It does not take ten years or eight medicines: two, properly tried.

That definition matters because it has a practical consequence. From that point on, the right course is not to carry on trying indefinitely: it is to investigate whether the case is a surgical one. Investigating it does not mean having an operation. It means finding out where the seizures come from and whether that place can be treated without damaging anything important.

The numbers on surgery

There are two randomised trials, and they are worth knowing along with their limits.

The first, from 2001, took 80 patients with temporal lobe epilepsy and allocated them at random to surgery or to continued medical treatment. At one year, 58 per cent of the operated group were free of seizures that impair awareness, against 8 per cent of the medical group. Quality of life was significantly better as well. Four patients — 10 per cent — had adverse effects from the surgery. One patient in the medical group died.

The second, from 2012, set out to answer something different: whether it is better to operate early rather than wait. The plan was for 200 participants and only 38 could be recruited, because people would not agree to be randomised, so the trial was stopped. Among those 38: free of seizures in the second year, 11 of 15 in the surgical group and 0 of 23 in the medical group. But the improvement in quality of life did not reach statistical significance on an intention-to-treat basis. And there was memory decline in 36 per cent of those operated on.

The authors themselves ask that it be read with caution, and it should be put exactly that way: seizure control was emphatic; the rest, less conclusive because of the size.

There is one finding from that same study that stands on its own. It documented that referral to an epilepsy surgery centre arrives, on average, after some twenty years of seizures. That delay is the central problem of this condition, and it has nothing to do with surgical technique.

In long-term follow-up, the proportion of patients free of seizures more than ten years after surgery holds at around 61 per cent.

And if surgery is not an option

Not every drug-resistant epilepsy can be operated on. Sometimes the seizures come from more than one place, sometimes from the area of language or of movement, sometimes the origin cannot be located at all.

In those cases there are options that involve removing nothing, and instead modulate the electrical activity of the brain with a device. In the seven-year follow-up of one of those systems, seizure frequency fell by a median of 75 per cent, and the seizures that generalise fell by 71 per cent. It is not a cure: it is a reduction, and for many people the difference between twenty seizures a month and five changes their whole life.

There is also the ketogenic diet in specific indications, above all paediatric ones, and image-guided laser treatment for small, well-localised lesions.

What you will hear that is not true

"You have to put something in their mouth so they do not swallow their tongue." It is the most widespread myth about epilepsy and it is false and dangerous: swallowing the tongue is not anatomically possible, and putting objects in breaks teeth, injures the mouth and can cause choking.

"If two drugs did not work, you have to keep trying drugs." After the first drug fails for lack of effect, only 11 per cent are controlled by the ones that follow. At that point the right course is to assess whether the case is a surgical one, not to move on to the tenth medicine.

"Epilepsy surgery is a last resort." The 2001 trial showed 58 per cent against 8 per cent in favour of surgery in drug-resistant temporal lobe epilepsy. Treating it as a last resort is what produces those two decades of delay.

"Surgery changes your personality or wipes your memory." The cognitive risk is real and it has been measured: 10 per cent adverse effects in one trial, 36 per cent verbal memory decline in the other, which was small. It is a risk to be weighed case by case, above all when the side of the dominant hemisphere is being operated on. Denying it would be dishonest; turning it into an absolute prohibition would be dishonest too.

"Cannabis cures epilepsy." The landmark trial was done in a specific, severe childhood syndrome, and it showed a reduction in convulsive seizures from 12.4 to 5.9 a month. That is a real improvement in a particular indication. It is not a cure, and it does not carry over automatically to every epilepsy.

"Epilepsy always shortens life." The World Health Organization estimates a risk of premature death up to three times higher, and at the same time points out that up to 70 per cent of people with epilepsy could live free of seizures with the right diagnosis and treatment. Both things are true, and the second depends on access.

What will happen during the assessment

If the surgical option is raised, what follows is a staged assessment, and it is worth knowing that it takes time.

First, a very detailed history of what the seizures are like: what you feel beforehand, what you do during them, what you remember afterwards. The videos your family recorded on a phone are worth a great deal; take them with you.

Then an MRI with a protocol specific to epilepsy — which is not the ordinary MRI — and an electroencephalogram. In many cases, an admission of several days with simultaneous video and electroencephalogram, to record the seizures as they happen.

And a neuropsychological assessment, to establish which functions are affected and which have to be protected.

All of that is then discussed by a committee bringing together neurology, neurosurgery, neurophysiology, neuropsychology and neuroimaging. That such a committee exists is a good sign in a centre; that it does not, a reasonable ground for looking for another.

Questions worth writing down before you go

• Does my epilepsy meet the criteria for drug resistance?
• How many drugs have I tried, at what doses, and why was each one stopped?
• Should I be referred to an epilepsy surgery centre for assessment?
• If the origin is localised, what function is at risk in that area?
• What is the specific chance of becoming free of seizures in my case?
• If I am not a candidate for resection, am I a candidate for a neurostimulation device?

The questions people ask most