Prediction of risk of seizure recurrence after a single seizure and early epilepsy: further results from the MESS trial
Kim LG, Johnson TL, Marson AG, Chadwick DW; MRC MESS Study group;
Commented by , 25 Jul 2006
Background
In patients who present with a single epileptic seizure or rare seizures, the decision to treat or not to treat requires careful consideration of risk/benefit ratio. This is primarily dependent on the probability of seizure recurrence in the absence of treatment, and the impact of treatment on such probability (ref. 1).
Aim
To assess predictors of seizure recurrence after a single seizure and early epilepsy in patients randomized to treatment or no treatment.
Methods
- the MESS study is a non-blinded randomised comparison of immediate versus deferred therapy with antiepileptic drugs (AEDs, usually carbamazepine or valproate) in 1443 patients in whom the risk/benefit ratio of treatment was uncertain (ref. 2). Methods and results are discussed in an earlier commented article (CNSforum, 26.07.2005)
- a prognostic model investigated individual patients' factors to identify groups at low, medium and high risk of recurrence
- a split-sample approach was used, i.e. the model was developed in a subgroup of the population and validated on the remainder of the sample.
Results
- three factors predicted the risk of seizure recurrence: (i) number of seizures at presentation; (ii) presence of a neurological disorder; (iii) an abnormal electroncephalogram (EEG)
- patients with 2 or 3 seizures at presentation, a neurological disorder or an abnormal EEG were at medium risk for recurrence. Patients with two of these features and those with >3 seizures were at high risk. Patients with a single seizure were at low risk
- Estimated risks of seizure recurrence at 3 years were
- low risk group: 35% with immediate treatment, 28% with deferred treatment (no treatment benefit)
- medium risk group: 35% with immediate treatment, 50% with deferred treatment (some treatment benefit)
- high risk group: 46% with immediate treatment, 67% with deferred treatment (some treatment benefit)
Professor Perucca's comments
The MESS study is the largest published randomised trial in persons with seizures (ref. 2). Its main findings, discussed earlier in CNSforum (26.07.2005), confirmed earlier evidence that early treatment is efficacious in reducing seizure recurrence (ref. 3, ref. 4, ref. 5, ref. 6), but it does not affect long-term prognosis in terms of ultimate seizure control (ref. 7, ref. 8).
Assessing the risk/benefit ratio of treatment requires an understanding not only of the efficacy of treatment, but also of the risk of recurrence if treatment is not given (ref. 9). This study identified three important risk factors, i.e. the number of seizures at presentation, the presence of a neurological disorder (i.e. a presumed etiology for the seizures) and an abnormal EEG (not just epileptiform abnormalities, but also slow-wave disturbance without spikes or sharp waves).
Although none of the findings of this study can be considered novel (ref. 9, ref. 10), its powerful database and randomised methodology allowed a better quantitation of the role of each of the three risk factors identied, as well as estimates of treatment effects in different risk groups. These estimates, however, apply only to a population meeting the inclusion criteria of the MESS trial, i.e., patients with often minor or rare seizures.
Specifically, the risk assessment model might not apply to patients with a first severe seizure, patients with seizures occurring at short intervals, and patients seen quickly after their seizure. Moreover, few patients in Kim’s sample had potential associated epileptogenic conditions, thereby preventing assessment of the comparative role of each.
In practice, Kim’s results support what is already standard policy, i.e. to consider treatment to be generally indicated when at least two seizures have occurred. In patients with a single unprovoked seizure, treatment is generally not indicated unless additional risk factors for recurrence are present.
In a 1995 article, Beghi and I (ref. 1) proposed a treatment algorithm whereby a first unprovoked seizure should be generally treated when "EEG epileptifom abnormalities and/or an associated epileptogenic condition" was identified. This is basically in line with the latest analysis.
As pointed out by the authors, treatment decisions in people presenting with one or few seizures cannot – and should not – be dictated by a mathematical model alone. Additional factors need to be factored in, including the nature of the any underlying neurological disorder, the nature of imaging or EEG findings, the type of epilepsy (for example, rolandic epilepsy usually does not require treatment), the patients’s age, profession, and attitude towards the risk of seizures and the risk of treatment (ref. 10).
References
1. Beghi E, Perucca E. The management of epilepsy in the 1990s. Acquisitions, uncertainties and priorities for future research. Drugs 1995; 49 (5); 680-94
2. Marson A, Jacoby A, Johnson A, Kim L, Gamble C, Chadwick D; et al. Immediate versus deferred antiepileptic drug treatment for early epilepsy and single seizures: a randomised controlled trial. Lancet 2005; 365(9476); 2007-2013
3. Camfield P, Camfield C, Dooley J, Smith E, Garner B. A randomized study of carbamazepine versus no medication after a first unprovoked seizure in childhood. Neurology 1989; 39 (6);851-852
4. Chandra B. First seizure in adults: to treat or not to treat. Clinical Neurology and Neurosurgery 1992; 94 Suppl; 61-63
5. First Seizure Trial Group (FIR.S.T. Group). Randomized clinical trial on the efficacy of antiepileptic drugs in reducing the risk of relapse after a first unprovoked tonic-clonic seizure. First Seizure Trial Group (FIR.S.T. Group). Neurology 1993; 43; 478-483
6. Gilad R, Lampl Y, Gabbay U, Eshel Y, Sarova-Pinhas I. Early treatment of a single generalized tonic-clonic seizure to prevent recurrence. Archives of Neurology 1996; 53 (11); 1149-1152
7. 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 (6); 662-3
8. Musicco M, Beghi E, Solari A, Viani F. Treatment of first tonic-clonic seizure does not improve the prognosis of epilepsy. Neurology 1997; 49; 991-998
9. 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 unprovoked seizure. Annals of Neurology 2000; 48; 140-147)
10. Perucca E, Beghi E, Dulac O, Shorvon S, Tomson T. Assessing risk to benefit ratio in antiepileptic drug therapy. Epilepsy Research 2000; 41 (2); 107-139