John Libbey Eurotext

Effect of rapid EEG on anti-seizure medication usage Article à paraître


  • Figure 1
  • Figure 2
  • Figure 3


1 Department of Neurology & Neurological Sciences, Stanford University School of Medicine, Stanford, CA, USA
2 Department of Neurology, California Pacific Medical Center, Sutter Health, San Francisco, CA, USA
3 Department of Emergency Medicine, Stanford University, Stanford, CA USA
Deepika Kurup
291 Campus Drive, Stanford, CA 94305, USA
Kimford Meador
213 Quarry Road, MC 5979, Palo Alto, CA 94304-5979, USA


To study how early diagnoses from rapid EEG (rEEG) during the initial evaluation of patients with suspected non-convulsive seizures correlates with changes in anti-seizure medication (ASM) use.


We performed a retrospective chart review of 100 consecutive adult patients at an academic medical center who underwent rEEG monitoring for suspected non-convulsive seizures. We collected information on the timing of ASM administration and categorized EEG diagnoses as seizures (SZ), highly epileptiform patterns (HEP), or normal or slow activity (NL/SL). We used a χ2 test to determine whether the use of ASMs was significantly different between SZ/HEP and NL/SL cases.


Of 100 patients, SZ were found in 5%, HEP in 14%, and no epileptiform/ictal activity in 81%. Forty-six percent of patients had received ASM(s) before rEEG. While 84% of HEP/SZ cases were started or continued on ASMs, only 51% of NL/SL cases were started or continued on ASMs after rEEG (χ2 [1, n=100] = 7.09, p=0.008). Thirty-seven patients had received sedation (i.e., propofol or dexmedetomidine) prior to rEEG. In 15 patients (13/30 NL/SL, 2/7 HEP/SZ), sedation was discontinued following rEEG.


Our study demonstrates that seizures were rapidly ruled out with rEEG in 81% of patients while 19% of patients were rapidly identified as having seizures or being at higher risk for seizures. The rapid evaluation of patients correlated with a significant reduction in ASM treatment in NL/SL cases compared to HEP/SZ cases. Thus, early access to EEG information may lead to more informed and targeted management of patients suspected to have nonconvulsive seizures.