The diagnostic yield of the routine EEG at a tertiary care center in Saudi Arabia: A retrospective study.

BACKGROUND: The Routine Electroencephalography (REEG) records cerebral electrical activity to aid in the diagnosis and classification of Epilepsy. Indiscriminate use of the REEG may lower its clinical yield. The pretest clinical variables contributing to the yield of the REEG outcome have not been well-established in the context of developing healthcare systems where REEG utilization may differ from well-established centers. The aim of this study is to determine the yield of the REEG and the pretest clinical variables predicting the yield of the REEG at a single center in the context of the developing healthcare system in Saudi Arabia. MATERIALS AND METHODS: We reviewed REEG reports at a single center in Jeddah, Saudi Arabia between 2015 and 2018. We collected demographic and clinical data from the patients' electronic files. Patients of age ≥18-year-old were included. We collected age, gender, nationality, the indication for the REEG, co-morbidities, antiseizure medicines (ASMs), and details related to the yield of the REEG (normal or abnormal, epileptiform or non-epileptiform, focality, and the presence of rhythmic or periodic patterns or seizures). RESULTS: We included 500 records. Fifty-nine percent were females. The mean age was 39 ±â€¯17 years. Of the recorded REEGs, 42.4% were abnormal, 14.6% of them showing definite epileptiform discharges and 85.4% showing only slowing. Half of the REEGs of individuals previously diagnosed with epilepsy revealed abnormal findings. ASM use was associated with slowing in the REEG (p < .05). Using logistic regression, history of a previous stroke and use of carbamazepine or lamotrigine were predictors of the presence of abnormalities in the REEG, while use of ≥2 ASMs predicted the presence of epileptiform discharges. Stroke also predicted abnormal slowing alongside increased age. CONCLUSION: The clinical yield of the REEG in this sample of patients from a single institution in the context of a developing healthcare system was relatively low. History of stroke and being on more than one ASM may predict that the REEG will show relevant abnormalities.