Abstracts

RATIONALE:_ Temporal lobe dysfunction is associated with altered sleep structure and quality. Experimental amygdala kindling induces disturbed sleep patterns favoring waking and light sleep (Van Sweden, 1990). Temporal lobe epilepsy patients have frequent stage shifts and numerous awakenings (Baldy-Moulinier, 1982). METHODS: We reviewed nocturnal EEG samples taken during long-term video-EEG monitoring in 41 patients evaluated for epilepsy surgery. Twenty-five patients had temporal lobe epilepsy (TLE) and 16 had extratemporal epilepsy (XTLE). Analyzed EEG segments included time samples taken every 30 minutes between the hours of 2300 and 0700 (total of 17 possible samples), in addition to segments detected by a computer program as possible spikes or seizures. In most cases, the first night of recording, before medications were tapered significantly, was analyzed. Patients were excluded if they did not have at least one night with no seizures and at least 11 time samples between 2300 and 0700. Amount of nocturnal sleep was estimated by counting the number of time samples showing only waking EEG, correcting for missing samples. Delayed sleep was defined as only waking samples up to and including 0100. Fisher's Exact Test was used to compare the proportion of TLE vs. XTLE patients with specified numbers of waking samples, and with delayed sleep as defined above. RESULTS: There was no significant difference between patients with TLE and those with XTLE with respect to total number of waking samples between 2300 and 0700. However, all 6 patients who remained awake until at least 0100 had TLE (p=.065) CONCLUSIONS:Epilepsy surgery candidates with TLE may have later sleep onset than those with XTLE. This could reflect impaired mechanisms of sleep initiation in the presence of a dysfunctional temporal lobe.