Contribution of sleep deprived EEG on diagnosis in clinical practice

Purpose; The place of sleep deprived electroencephalography (SD-EEG) in the diagnosis of epileptic patients' remains unclear despite the fact that it is being used for a long time. In the present study, where the majority of data was analyzed retrospectively and some were analyzed prospectively, we aimed to investigate how much this activation method contributes to diagnosis and in which situations it would be more efficient. Material and method; A total of 1400 SD-EEG, which had been recorded between 2004 and 2015 in the KTU Faculty of Medicine, EEG laboratory, were reviewed. Patients who had normal/ nonspecific routine EEG (R-EEG) or an EEG showing other activity without interictal epileptiform abnormalities, who could be obtained clinical data, who stayed awake for 20 hours and over before the UD-EEG for adults and stayed awake for longer periods than they used to for children, were included. A total of 566 patients, who was diagnosed as epilepsy or non epileptic paroksismal event based on clinical findings, cranial MRI/CT and EEG findings, were included in the study. The patients were divided into 4 groups. The first group (235 patients) whose final diagnosis was epilepsy and had recordings during both wakefulness and sleep consisted. The second group (33 patients) whose final diagnosis was epilepsy and who didn't slept during recording consisted. The third group (263 patients) consisted, whose final diagnosis was non-epileptic paroxysmal event and who had recordings during both wakefulness and sleep. The fourth group (35 patients) whose final diagnosis was non-epileptic paroxysmal event and who didn't sleep during recording consisted. 24 hours sleep deprivation prior to SD-EEG was considered as total SD for adults and children. 20 to 24 hours of sleep deprivation was considered as partial SD for adults. 24 hours sleep deprivation was considered as total SD and shorter than 24 hours was considered as partial SD in the patients with shorter sleep durations than they used to for pediatric group. EEG activation was used for detection interictal epileptiform discharges (IIED) on SD-EEG. Results; Activation rate was 41,7% in Group 1,24% in Group 2 and 4,6% in Group 3 and %8,6 in group 4. Whilst the sensitivity of SD-EEG was determined to be 41,7% for Group 1, specificity of SD-EEG was found to be 95,4% for Group 3 (p=<0,001). Nonspecific findings in R-EEG and SD time performed before SD-EEG in Group 1, did not statistically significant change in SD-EEG activation. In Group 1, comparing the patients according to whether they are adult or young/child, they have family history of epilepsy or personal history of febrile convulsion, type of epilepsy, use antiepileptic, kind of antiepileptic and number of antiepileptic drug, whether there are epilepsy risk factors for focal epilepsy and whether there is pathological finding on neuroimaging, did not statistically significant role in SD-EEG activation. We determined that only the type of epilepsy plays statistically significant role in SD-EEG activation. In Group 1, the rate of determining IIEDs was statistically significantly higher for generalized epilepsy as compared to focal epilepsy and/or epilepsies in which focal-generalized discrimination could not be made (56%, 39,7% and 27,6%, respectively) (p=0,033). Analyzing the distribution of IEDs in Group 1 patients among sleep, wakefulness and both sleep and wakefulness (combined), it was observed to be the least on wakefulness traces and the most frequent during combined sleep and wakefulness (34% during sleep alone, 25% during wakefulness alone and 41% during combined sleep and wakefulness), although no statistically significant difference was determined. Conclusion; Sleep deprivation is a beneficial activation method with high specificity, which can be easily applied in case of suspicion for the diagnosis of epilepsy. We are in the opinion that this protocol would contribute more to diagnosis particularly in the presence of suspicion of generalized epilepsy. Moreover, findings that are helpful in diagnosis can be determined even the patient does not sleep during SD-EEG recording. However, extending recording period and trying to catch a sleep period, if possible, might provide additional contribution in activation. Keywords: Epilepsy, activation method, nonepileptic paroxysmal event, sleep deprivation, sleep deprived electroencephalography