LONG-TERM SUBDURAL SURFACE cortical cerebral blood flow (CBF) and elec
trocorticographic monitoring was performed in 12 patients with complex
partial seizures. A total of 40 seizures were analyzed. Baseline CBF
values from nonepileptic and epileptic temporal lobe (mean +/- standar
d error) were 60.0 +/- 1.0 and 50.2 +/- 1.8 ml/100 g per minute, respe
ctively (P < 0.05). In general, clinical seizure onset was preceded by
a 20-minute preictal CBF increase from baseline in the epileptic temp
oral lobe. Peak early postictal CBF values of nonepileptic and epilept
ic temporal lobes were 57.7 +/- 13.3 and 89.0 +/- 21.7 ml/100 g per mi
nute (P > 0.05) at 5.2 +/- 2.2 and 2.4 +/- 1.0 minutes (P > 0.05) afte
r clinical seizure onset, respectively. Statistically significant diff
erences between nonepileptic and epileptic temporal lobe CBF were dete
cted at 50 minutes (74.0 +/- 14.2 and 37.5 +/- 9.2 ml/100 g per minute
, respectively; P < 0.05) and 60 minutes (75.6 +/- 13.6 and 36.1 +/- 8
.5 ml/100 g per minute, respectively; P < 0.05) postictal. The data su
ggest that the optimal times for CBF analysis to differentiate epilept
ic from nonepileptic temporal lobe are 1) during the interictal period
and 2) late (50 to 60 minutes) postictal. The results of this study s
hould improve the understanding of the dynamic cerebral perfusion patt
erns in the epileptic human brain.