Abstracts

Rationale: Seizure threshold may be decreased during the early morning hours in selected patients with epilepsy. Furthermore, duration of sleep and sleep deprivation may influence epileptogenicity. Objectives were to determine the frequency of interictal epileptiform discharges (IEDs) on routine EEGs, to analyze the relationship between IEDs and time of day of the EEG, and to investigate the effect of decreased sleep duration on the occurrence of IEDs.Methods: 30,479 consecutive routine EEGs between 1996 and 2005 were retrospectively reviewed for generalized and focal sharp waves. Patients with emergency EEGs and portable EEGs outside the EEG laboratory were excluded. Sleep duration during the night prior to the EEG and the time of day of the EEG was recorded.Results: IEDs were seen in 5,317 out of 30,527 EEGs (17.4 %), more often in the pediatric population, (2,578 out of 9,855 EEGs or 26.2%) than in adults (2,739 out of 20,672 EEGs or 13.2%). 1,530 (15.1%) of children had focal, 717 (7.1%) had generalized, and 331 (3.3%) had both generalized and focal IEDs. 1,927 (9.3%) of adults had focal, 660 (3.2%) had generalized, and 152 (0.7%) had generalized and focal IEDs. Patients with sharp waves were significantly younger (on average 26 years) as compared to patients without sharp waves (on average 37 years) (p<0.001). Out of 17,688 EEGs done before noon, IEDs occurred in 3,091 (17.4%), while the 12,839 afternoon EEGs showed IEDs in 2,226 EEGs (17.3%). 55.9% of all EEGs with focal IEDs were recorded before noon (as compared to 44.1% after noon). 63.2% of all EEGs with generalized IEDs were recorded before noon (as compared to 36.8% after noon). Before noon EEGs with focal IEDs were less frequent (55.9%) than EEGs without IEDs (57.9%) (p<0.0001). Before noon EEGs with generalized IEDs (63.2%), or EEGs with both focal and generalized IEDs (59%) were significantly more frequent than EEGs without IEDs (57.9%) (p<0.0001). The sleep duration did not differentiate between patients with and without IEDs (median sleep duration 7 hours). Morning EEGs followed a shorter sleep (median 6 hours) than afternoon EEGs (median 8 hours) (p<0.001) in both adult and pediatric patients. Interestingly, sleep deprived EEGs (5 hours or less) were seen most frequently among those without IEDs (20.1%), followed by EEGs with generalized IEDs (19.2%), EEGs with focal IEDs (15.9%) and EEGs with focal and generalized IEDs (12.84%) (p<0.0001).Conclusions: IEDs were seen more frequently during pediatric routine EEGs as compared to adult EEGs. Overall percentage of IEDs recorded before and after noon was equal. However, generalized IEDs were more frequent before noon than focal IEDs. Decreased sleep duration was seen more frequently in morning EEGs and was associated with a higher frequency of generalized IEDs on EEG as compared to focal IEDs. Although referral and selection bias at our tertiary epilepsy center cannot be excluded, large numbers may have reduced this bias. Morning EEGs after decreased sleep duration may be particularly useful to demonstrate generalized IEDs.