OP19 – 2435: Prognostic significance of EEG in infantile spasms

Abstract

Objective To assess the prognostic value of initial characteristics of EEG for primary and long-term outcome in infantile spasms, and correlation between hypsarrhythmia duration/resolution/normalization during treatment and long-term outcome. Methods In the cohort of 180 infants with infantile spasms treated in the Institute with vigabatrin from the beginning of 1996 to the end of 2010. the analysis of the impact of EEG features on the first EEG to primary outcome (cessation of spasms) was done. Also the impact of initial EEG characteristics and after 6 weeks of treatment on long-term outcome was investigated after follow up period of 2.4 to 18.9 years (M=10.64; SD=4.40). The duration of hypsarrhythmia was assessed with standard statistical methods and logistic regression analysis. Results In the initial EEG hypsarithmia occurred more frequently in non-responders (χ 2 (1)=3.789, p=0.052), whereas modified hypsarithmia occurred more frequently in responders (χ 2 (1)=4.282, p=0.039). Odds for favorable long-term outcome in patient with initial hypsarrhythmia was 0.45 and with initial modified hypsarrhythmia was 1.05. After 6 weeks of treatment, odds for favorable long-term outcome in patient with hypsarrhythmia was 0.17 and with normal EEG was 1.45 (χ 2 (3)=15.261, p=0.002). Duration of hypsarrhythmia did not statistically influence the long-term outcome assessed as divided (χ 2 (2)=3.647, p=0.161) or continuous variable. Conclusion Features of hypsarrhythmia on initial EEG did not significantly determine the primary or long-term outcome in our patients. The duration of hypsarrhythmia did not correlate with the long-term outcome. But the normalization of EEG was associated with cessation of spasms, and reactivity to therapy with good outcome. That points to the complex interplay between duration of hypsarrhythmia and basic etiologic condition in the prognosis of IS.