• number of patients free from post stroke recurrent crisis [ Time Frame: one year ] [ Designated as safety issue: No ]
  

• To compare retention time of LEV vs CBZ since first intake throughout treatment period [ Time Frame: one year ] [ Designated as safety issue: No ]
  
• To compare time to second seizure in both treatments. [ Time Frame: one year ] [ Designated as safety issue: No ]
  
• To evaluate differences in cognitive function and in quality of life in levetiracetam and carbamazepine patients having post-stroke seizures at the end of treatment period [ Time Frame: one year ] [ Designated as safety issue: No ]
  
• evaluate EEG changes as compared with baseline with that obtained at the end of treatment period [ Time Frame: one year ] [ Designated as safety issue: No ]
  
• To compare seizure frequency in levetiracetam and carbamazepine groups throughout treatment period [ Time Frame: one year ] [ Designated as safety issue: No ]
  
• To evaluate the safety of levetiracetam versus carbamazepine throughout the treatment period [ Time Frame: one year ] [ Designated as safety issue: Yes ]
  

Stroke is the most common cause of seizures in the elderly and seizures are among the most common sequelae of stroke.

About 10% of patients experience seizures since stroke onset up to several years (Silverman 2002). Arbitrarly a cut point of 2 weeks divide early seizures from late seizures (Honey 2000, Olofson 2000, Berger 1989).Late occurrence of late seizure appears to carry a high risk for epilepsy (Wilmore 1990).

The risk of epilepsy in some patients with a single stroke-related seizure is high enough to justify starting an anticonvulsant therapy before a second seizure occurs(Labovitz 2003).

Levetiracetam (S-α-ethil-2-oxo-pyrrolidine acetamide) is S-enantiomer of a pyrrolidine derivative and is unrelated to any other AED and has a unique preclinical and clinical profile (Gower et al 1992).

Levetiracetam (LEV) binds with a stereospecific binding site in the CNS that is saturable and reversible (Noyer 1995). This site actually known as LBS Levetiracetam Binding Site) is unique and do not correspond to any known receptor or channel that might be involved in neuroexcitability (Gillard 2003).

LEV selectively inhibits N-type Ca2 channels of CA1 pyramidal hyppocampal neurons (Lukyanetz et a 2002) and, despite of not having any activity on GABA-gated currents, it shows a potent ability to reverse the inhibitory effects of the negative allosteric modulators zinc and β-carbolines on both GABAA and glycine receptor mediated responses(Rigo et al. 2002).

LEV has no effects on normal neurons (Birnstiel et al.1997) LEV as other AEDs has effect in decreasing repetitive neuronal firing, but only LEV reduces the number of cells firing synchronously (amplitude) of the evoked PS(Margineau and Klitgaard 2000).

The efficacy profile of the drug has been established through three pivotal randomized double blind, placebo controlled, parallel studies on 904 patients suffering from partial seizures secondarily or not generalised that were not well controlled by previous treatment (Shorvon et al. 2000; Ben-Menachem et al. 2000; Cereghino et al. 2000).In these three studies LEV showed a significant reduction of seizure frequency. A pooled analysis of the results from these three studies supports a dose-response effect for levetiracetam: responder rates were 28.5, 34.3 and 41.3 % for patients treated with levetiracetam 1000, 2000, 3000 mg/day respectively, as compared with 13.1% for placebo group. The respective values for complete seizure freedom were 4.7, 6.3, 8.6 and 0.8%(Privitera 2002, Boon et al, 2002).

In a review of data for 1422 patients treated with levetiracetam, 38.6% of patients experienced a ≥ 50% reduction in seizure frequency and 20% experienced a reduction of ≥ 75%. The present study is not blinded because one of the purposes with the study has been to mimic daily clinical practice as close as possible.