Page 358 - ILAE_Lectures_2015

those in remission, 20% continued on AEDs while 50% had successfully discontinued
medication and remained seizure-free for ≥5 years. In a cohort of children with active
epilepsy followed up for 12 years 64% were in terminal remission (defined as ≥3 years seizure
free) after 12 years14.

In a study of children followed up for an average 37 years, 67% were in terminal remission,
on or off medication. Early remission, defined as remission occurring within the first year of
treatment, was achieved by 31%, and the remission continued to terminal remission in half
of these. Remission without relapse occurred in 50% with a mean delay of nine years. A total
of 14% entered remission but subsequently relapsed with further remission, indicating a
relapse-remitting pattern, while 19% continued with seizures from the onset17. Of children
followed up for a median of 40 years, 93% had one or more periods of remission (one year),
emphasising the overall excellent prognosis of childhood epilepsy19.

For those with chronic epilepsy, up to one-third will have a relapsing remitting pattern with
at least one period of significant seizure freedom20.

Prognostic factors

Many studies have looked at possible predictors of seizure prognosis, including age of onset,
gender, aetiology, seizure type, EEG patterns, number of seizures prior to treatment and early
response to treatment21. In patients presenting with a first-ever seizure, the presence of
multiple discrete seizures within 24 hours is not associated with a worse prognosis than those
with a single seizure22. Remote symptomatic epilepsy, the presence of a neurological birth
deficit and learning disability are consistently shown to be associated with a poorer prognosis.
In one study the three-year remission rate was 89% for those with idiopathic epilepsy and
normal examination compared to only 49% in those with a neurological deficit or learning
disability13. The number of seizures in the first six months after onset has been found to be a
strong determinant of the probability of subsequent remission, with 95% of those with two
seizures in the first six months achieving a five-year remission compared with only 24% of
those with more than ten seizures23.

Seizure type has been an inconsistent prognostic factor with some studies indicating that those
with partial seizures have a poorer prognosis12 while other studies have demonstrated a poorer
prognosis for those with generalised onset seizures24. People with multiple seizure types, as
is typical in the childhood encephalopathies, appear to have a poorer prognosis25. A
significant reduction or complete cessation of seizures within three months of initiating
treatment has been shown to be a strong predictor of subsequent remission26. The probability
of seizure remission decreases significantly with each successive treatment failure. Only 11%
of patients who discontinued the first appropriate AED due to lack of efficacy became seizure
free on a second AED and only 4% on a third medication or on polypharmacy27.

Children who experience clusters of seizures during treatment are much more likely to have
refractory epilepsy than children without clusters and are less likely to achieve five-year
terminal remission28. Children who continued to have weekly seizures during the first year of
treatment had an eight-fold increase in the risk of developing intractable epilepsy and a two-
fold increase in the risk of never achieving one-year terminal remission19.

The impact of aetiology on prognosis

When comparing prognosis by aetiology, patients with idiopathic generalised epilepsy appear
to have a better prognosis than patients with symptomatic or cryptogenic partial epilepsy. In
one study 82% of people with idiopathic generalised seizures achieved one-year seizure