Failure of seizure termination mechanisms

Epileptic seizures are typically brief (1–2 minutes) and self-limiting, even in the absence of anticonvulsant medication, and the brain remains refractory to further seizure attacks for a period of time after the seizure (the postictal refractory period). GABAergic mechanisms are likely to be involved in seizure termination mechanisms and there is some support for this contention. Stringer and Lothman (113) have characterized a stereotyped form of paroxysmal event in the dentate gyrus that can be used as a marker for reverberatory seizure activity in the hippocampal formation called maximal dentate activation (MDA). MDA has the propensity to lengthen when repeatedly elicited and has quantifiable parameters. The time to onset of MDA is used as an indicator of the ease with which seizure activity can be initiated, while the duration of MDA is an indicator of the ability of the system to terminate seizures (113). The GABA_A receptor antagonists bicuculline and picrotoxin lengthen MDA, and conversely, muscimol, a GABA_A agonist, and baclofen, a GABA_B agonist, shorten MDA but have little effect on the time to onset of MDA (114). This suggests that GABAergic mechanisms are more important in seizure termination than seizure initiation mechanisms. However, if GABAergic mechanisms played a significant role in seizure termination, then administration of GABA receptor antagonists should prolong the duration of brief seizures. Although there is some evidence to support this conjecture, the majority of studies show that subconvulsant doses of GABA antagonists have no effect on kindled seizures, suggesting that endogenous GABA is not involved in seizure termination. For example, the GABA_A antagonists bicuculline and picrotoxin do enhance secondary generalization after amygdaloid kindling (87), yet norharmane, a benzodiazepine inverse agonist, failed to promote secondary generalization. Similarly, other studies have found that picrotoxin does not affect the duration of amygdala-kindled seizures (68). In addition, GABA_A antagonists also do not augment pilocarpine-induced seizures to cause SE (120), nor does bicuculline alter either the ictal or interictal phases of penicillin-induced SE (47), suggesting that activation of GABA_A receptors is probably not involved in seizure arrest mechanisms (28).

A more favourable hypothesis has been proposed (26, 28). It has been postulated that the seizure-induced liberation of endogenous anticonvulsant substances is involved in seizure termination and postictal protection from further seizure episodes, as shown in Figure 24.1 (26, 28). These early studies in our laboratory led us to ask (132), if there is a failure in the ability of the brain to terminate brief epileptic seizures, then do seizures continue unabated, and does SE develop? The following discussion reviews some of the processes proposed in the spontaneous arrest of brief seizures.

Figure 24.1.  

Endogenous anticonvulsant substances. Brief seizures lead to the production of seizure terminators, which are responsible for arresting brief seizures. Levels of adenosine (128) and perhaps nitric oxide (NO) (12, 124) rise rapidly during the brief seizure, may reach some threshold and then terminate the seizure. The potent inhibitory effects of adenosine, mediated by A₁-adenosine receptors, on the release of excitatory amino acid neurotransmitters, as well as its direct hyperpolarizing action, may mediate its anticonvulsant effects (38, 39). Postseizure anticonvulsants are produced in the brain and these account for the postseizure refractory period. Adenosine levels return to baseline slowly after seizure arrest (44), and therefore may also provide postseizure anticonvulsant effects (126). However, the main substances that appear to exert postseizure anticonvulsant effects are opioids and prostaglandins, because interfering with their production or effects reduces postseizure refractoriness (28).