Immediate versus deferred antiepileptic drug treatment for early epilepsy and single seizures: A randomized controlled trial
Marson A, Yacoby A, Johnson A, Kim L, Gamble C and Chadwick D; on behalf of the Medical Research Council MESS Study Group;
Commented by , 26 Jul 2005
Background
There has been a long-standing debate on whether early treatment reduces the probability of developing drug-refractory epilepsy (ref. 1; ref. 2).
Aim
To compare the impact of immediate versus deferred antiepileptic drug (AED treatment in seizure patients in whom the indication for starting treatment is uncertain.
Methods
- Non-blind, randomized, multicentre comparison of immediate versus deferred AED therapy in patients with single seizures and early epilepsy in whom risk/benefit ratio of treatment was deemed to be uncertain.
- Primary outcomes measures were time to first, second and fifth seizure, time to 2-year remission, and seizure freedom between years 1 and 3 and between years 3 and 5 after randomization
Results
1847 patients were invited to join the trial and 1443 consented to randomization. The two groups had comparable features. 56% had had only one seizure at randomization. In treated patients, carbamazepine or valproate were used in over 90% of cases.
In the deferred treatment group, 47% of patients never had a seizure recurrence.
Immediate treatment reduced time to first seizure, second seizure, and first tonic-clonic seizure as well as time to achieve 2-year remission. However, there were no differences between immediate treatment and deferred treatment for seizure freedom rates between years 1 and 3 (74% vs 71% respectively) and years 3 and 5 (76% vs 77% respectively).
Quality of life measures and serious complications did not differ between the two groups. However, more patients given immediate treatment had drug-related adverse events.
Professor Perucca's comments
This is the largest randomized controlled trial (RCT) ever completed in seizure patients!!
The authors planned to recruit 3000 patients because their initial intention was to enable model testing and validation with a split-half approach. Although "only" 1443 patients were recruited, the study retained huge power to compare the primary outcome measures.
The emerging message is that early treatment reduces risk of recurrence but does not affect long-term response to AEDs. However, 14 patients need to be treated after their first seizure to prevent one additional patient having a seizure recurrence, which questions the risk/benefit ratio of such early treattment.
As expected, the impact of treatment becomes greater in patients who already had two or more seizures, simply because the risk of recurrence without therapy increases with increasing number of previous seizures (ref. 3).
I would take objection to the authors' interpretation that their findings represent "reliable evidence" that early treatment “does not interfere with the process of long-term epileptogenesis”.
Though this interpretation was endorsed by an accompanying editorial (ref. 4), what the study simply shows is that for the specific population meeeting the inclusion criteria early treatment has no impact on responsiveness to AEDs.
Showing that early treatment did not interfere with epileptogenesis would have required discontinuing treatment in “responsive” patients and showing similar relapse rates in the two groups.
These findings of this study are not, regrettably, new. Earlier, much smaller RCTs had already concluded that early treatment prevents recurrence rates (ref. 5; ref. 6; ref. 7; ref. 8) without impacting on long-term outcome (ref. 9; ref. 10).
Overall, the contribution of this article to new knowledge appears to be modest and probably disproportionate to the efforts involved with conducting such a massive trial.
The article gives no information concerning factors (other than number of previous seizures) which influence risk/benefit ratio of early treatment in individual patients. It is also impossible to tell to what extent findings reported for the pooled population can apply to patients with specific syndromes, etiologies and other risk factors.
The hope is that such information will emerge from further analysis of the data. The authors state that "to provide estimates of risk seizure recurrence for individual patients, data from our study will be used to generate prognostic models that will further inform decision making."
These prognostic models are eagerly awaited, even though their validity could remain uncertain given that the part of the study intended for model testing could not be completed.
References
1. Reynolds EH. Early treatment and prognosis of epilepsy. Epilepsia 1987; 28; 97-106
2. Berg AT, Shinnar S. Do seizures beget seizures? An assessment of the clinical evidence in humans. J Clin Neurophysiol 1997; 14; 102-10
3. Shinnar S, Berg AT, O’Dell C, Newstein D, Moshe SL, Hauser WA. Predictors of multiple seizures in a cohort of children prospectively followed from the time of their first unproked seizure. Ann Neurol 2000; 48; 140-7
4. McIntosh A, Berkovic SF. Treatment of new-onset epilepsy: Seizures beget discussion. Lancet 2005; 365; 1985-6
5. Camfield P, Camfield C, Dooley J, Smith E, Garner B., et al. Randomized study of carbamazepine versus no medication after a first seizure in childhood. Ann Neurol 1989; 39; 851-2
6. Chandra B. First seizures in adults: To treat or not to treat. Clini Neurol Neurosurg 1992; 94; S613
7. First Seizure trial Group (FIRST). Randomized clinical trial on the efficacy of antiepileptic drugs in reducing the risk of relapse a firest unprovoked tonic-clonic seizure. Neurology 1993; 43; 478-83
8. Gilad R, Lampl Y, Gabbay U, Eshel Y, Sarova-Pinhas I. Early treatment of a single generalized tonic-clonic seizure to prevent recurrence. Arch Neurol 1996; 53; 1149-52
9. Camfield P, Camfield C, Smith S, Dooley J, Smith E. Long-term outcome is unchanged by antiepileptic drug treatment after a first seizure: A 15-year follow-up from a randomized trial in childhood. Epilepsia 2002; 43; 662-3
10. Musicco M, Beghi E, Solari A, Viani F. Treatment of first tonic-clonic seizure does not improve the prognosis of epilepsy. Neurology 1997; 49; 991-8