Accuracy and interobserver reliability of scalp ictal EEG

Neurology. 1992 Dec;42(12):2279-85. doi: 10.1212/wnl.42.12.2279.

Abstract

We studied the accuracy and reliability of scalp ictal EEG in 137 complex partial seizures (119 temporal and 18 extratemporal) in 35 patients in whom we knew the correct site of seizure origin because all patients had been seizure-free for more than 2 years after seizure surgery. Three electroencephalographers independently determined side of seizure origin based on activity at onset of electrographic seizure (ASO), rhythmic theta and alpha (RTA), postictal findings (PIF), and the electrographic seizure as a whole. When all seizures were analyzed, including those with generalized features or obscured by artifact, we determined side of seizure onset correctly in 76% to 83% of temporal seizures and 47% to 65% of extratemporal seizures. In most of the remainder, a lateralization judgment was impossible. When analysis was confined to those seizures in which lateralization was possible, we correctly lateralized 93% to 99% of temporal seizures and 89% to 100% of extratemporal seizures. Interobserver reliability was excellent. RTA and PIF were more accurate than ASO. RTA was significantly more common in temporal seizures. Our data indicate that lateralization by scalp EEG is highly accurate and reliable.

MeSH terms

  • Electroencephalography / methods*
  • Epilepsy, Complex Partial / physiopathology*
  • Functional Laterality
  • Humans
  • Observer Variation
  • Reproducibility of Results
  • Scalp / physiopathology*