Hypsarrhythmia: Frequency of variant patterns and correlation with etiology and outcome

U. Kramer, W. C. Sue, M. A. Mikati*

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review


To determine the frequency and significance of the EEG features of hypsarrhythmia, we analyzed the pre-ACTH records of 53 consecutive patients with infantile spasms for the severity of the following abnormalities: disorganization of background, slowing, high amplitude, spike activity, and for the presence or absence of each of the following patterns and variants: electrodecremental discharges, absence of normal sleep activity, relative normalization, hemihypsarrhythmia, burst suppression (BS), occipital hypsarrhythmia, interhemispheric asymmetry, and interhemispheric synchronization. We calculated a total score indicating the severity of the hypsarrhythmia for each record. The hypsarrhythmia variant patterns occurred frequently in up to 69% of the records. Patients with cerebral dysgenesis were more likely to have hemihypsarrhythmia or BS pattern persistent throughout the EEG. Patients with history of perinatal hypoxia-ischemia were more likely to have absence of normal sleep activity. The occurrence of each of the other variant patterns did not correlate with etiology. Favorable outcome did not correlate with the occurrence, or absence, of any of the variant patterns but was associated with faster background activity (<75% delta), a lower total hypsarrhythmia score (≤10), and with absence of electrodecremental discharges on the pre-ACTH EEG. We conclude that variant patterns of hypsarrhythmia are frequent, generally do not correlate with prognosis, and thus are best included within the definition of hypsarrhythmia. The severity of the hypsarrhythmia, however, does have significant prognostic implications.

Original languageEnglish
Pages (from-to)197-203
Number of pages7
Issue number1
StatePublished - Jan 1997
Externally publishedYes


Dive into the research topics of 'Hypsarrhythmia: Frequency of variant patterns and correlation with etiology and outcome'. Together they form a unique fingerprint.

Cite this