Complexities in the association of human blood brain barrier disruption with seizures: importance of patient population and method of disruption.

We recently reported that epileptic seizures might be facilitated through the action of blood–proteins that leak into the CNS during the chronic epileptic phase in rats (Van Vliet et al., 2007). This was further supported by our mannitol experiments in epileptic rats which showed increased seizure progression in a subgroup of rats. Mannitol can be used to open the blood–brain barrier (BBB) by producing osmotic shrinking of the endothelial cells and mechanical separation of the tight junctions that form the BBB. Therefore mannitol can be used to deliver various drugs directly to the brain. Previous studies by Neuwelt et al. (Neuwelt et al., 1983, 1986; RomanGoldstein et al., 1994) pointed to the fact that seizures could be one of the complicating factors when patients with brain tumours were treated with mannitol. Since these patients were also treated with chemotherapeutic and contrast agents, seizures might have occurred through the pro-epileptogenic action of these agents. On the other hand, blood proteins that leak into the CNS might also facilitate seizure activity especially when potentially epileptogenic tumours are present. Many brain tumours are highly epileptogenic: epilepsy has been reported in 480% of low-grade gliomas (Vertosick et al., 1991), in 30–60% of high-grade gliomas (Scott and Gibberd, 1980), in 30% of meningiomas (Lieu and Howng, 2000) and in 20% of primary CNS lymphomas (Hochberg and Miller, 1988). Marchi and colleagues recently reported that leakage of blood proteins into the CNS via mannitol treatment might induce seizures. This treatment was performed in brain tumour patients (with CNS lymphomas), but they substantiated their conclusion by findings in two normal pigs in which seizures occurred after mannitol injection that was accompanied by BBB leakage. It is tempting to speculate that these acute seizures were caused by blood-borne substances that have entered the brain after the BBB is compromised (Marchi et al., 2007). We need to be cautious however not to overestimate the role of a leaking BBB in acute seizure activity. After all, (1) we did not observe acute seizures after mannitol treatment in control rats and (2) an epileptogenic focus only develops several days after cortical BBB disruption in control rats (Seiffert et al., 2004; Ivens et al., 2007; Tomkins et al., 2007). Thus the pro-epileptogenic action of mannitol in normal pig brain could reflect methodological differences or species specificity. Moreover it is very likely that other mechanisms play a role in the manifestation of the acute seizures versus the development of an epileptogenic focus or the increased excitability on an already epileptogenic ‘background’. There are anecdotal data that suggest that mechanical BBB disruption by electrode placement in the brain, decreases seizure activity, which would argue against the pro-epileptogenic role of BBB leakage. The fact that these patients receive AEDs and other perioperative drugs (such as corticosteroids) makes it rather difficult to estimate the contribution of the BBB disruption itself. Controlled animal studies might resolve this issue. Brain (2007), 130, e78