Abstracts

Rationale: The course of pediatric epilepsy over many years is complex and cannot be summarized in a single, terminal outcome. We sought to characterize the course of epilepsy over a period lasting up to 20 years and identify common meaningful patterns that may have implications for anticipatory guidance and management. Methods: Children (0-15 years) were prospectively recruited and followed from the time of initial diagnosis of epilepsy (1993-1997) in Connecticut, US. Frequent phone contact and medical record review allowed detailed recording of periods of remission and relapse, and use of anti-seizure medications (ASM). Seizure outcomes included multiple 1, 2, 3, and 5, year remission periods, pharmacoresistance (failure of 2 appropriately used ASMs), and complete remission at last contact (CR-LC, 5-years both seizure and ASM-free) at the end of follow-up. Early seizure outcome (2-yr after diagnosis) was characterized as in ≥1 yr remission, pharmacoresistant or neither. Epilepsies were grouped: focal seizures-self-limited (F-SL), encephalopathic (ENC), generalized genetic (GGE), nonsyndromic (NSE), and other uncharacterized (UNC). Uncomplicated epilepsy presentation was defined as the absence of obvious neurologic condition, insult or disability. Statistical analyses were performed with Χ² tests. Analyses were limited to children followed ≥10 years. Results: Of 613 children originally enrolled, 13 (2%) died before 10 years and 516 (84%) were successfully followed ≥10 years for an average of 17.2yr (range 10-21). A 1-yr remission occurred in 490 (95%) children; 256 (52%) relapsed. Comparable figures for 2, 3, and 5 year remission were, 92%, 89%, and 81%. Relapses following 2, 3, and 5 year remission occurred in 47%, 29%, and 15%. Repeat remissions and further relapses were common. 116 (22%) children experienced pharmacoresistance. At two years after diagnosis, 288 (89%) were in early remission, 58 (11%) had failed two ASMs, but for 172 (34%) the outcome was unclear. Composite seizure outcomes were created with eight categories ranging from (a) smooth sailing (Camfields 2002, early, lasting remission resulting in CR); (b) later but sustained remission resulting in CR-LC and with subsequently less ideal categories ending with (h) never achieved 1-year remission (Table 1). Substantial differences (p<0.0001) in the distribution of this composite outcome variable were observed across different forms of epilepsy and in those with and without complicated epilepsy presentations (Figure 1). Conclusions: The course of epilepsy is more complex than can be summarized by the remission status at one point in time. A substantial proportion of children experience a smooth-sailing course over many years. Only 5% never experience a prolonged (>1yr) period of remission. The majority have a range of difficulties in seizure control but, over the course of time, experience substantial, if not sustained, periods seizure-free. These patterns vary substantially by clinical features of the epilepsy and may help explain some social outcomes in later adulthood despite apparently good seizure control. Funded by NINDS-R37-31146