THE UNITED KINGDOM INFANTILE SPASMS STUDY: NEURODEVELOPMENTAL OUTCOME AT 4.2 YEARS OF AGE

Objectives: Infantile spasms are a severe infantile seizure disorder with a high risk of poor outcome. We undertook a randomised controlled trial of 107 infants, comparing hormonal treatments (prednisolone or tetracosactide) with vigabatrin. We have reported our early outcome measure, absence of spasms on days 13 and 14: vigabatrin was inferior to hormonal treatment; hormonal treatment 40/55 (73%), vigabatrin 28/52 (54%), difference 19% (95% CI 1% to 36%, p=0.043). Developmental outcome at 14 months showed no difference in mean Vineland adaptive behaviour composite scores (VABS) between treatment groups but a significant interaction with underlying aetiology. In the subgroup with no underlying aetiology, the mean score was 88.2 (17.3) on hormonal treatment and 78.9 (14.3) on vigabatrin; difference 9.3%, 95% CI +1.2 to +17.3; p=0.025. Our objective was to determine if the difference in VABS between treatment groups for those with no underlying aetiology is greater as the children get older and more complex development is assessed. Methods: The VABS were administered by one researcher (KD). Composite scores for each treatment group (and by aetiology) were analysed by intention to treat. Results: Of the original 107 infants, 9 have died and 77 have had a VABS assessment at a mean age of 4.2 years (SD 0.8). Median (IQR) VABS was 60(42–97) for the 39 allocated hormonal treatments and 50(36–67) for the 38 allocated vigabatrin. For those infants with no underlying aetiology, median (IQR) VABS was 96(54–102) on hormonal treatment, n=21, and 63(38–91) on vigabatrin, n=16. Wilcoxon Rank Sum Test p=0.033. Conclusions: In the subgroup with no underlying aetiology, better initial control of spasms for those allocated hormonal treatment was associated with significantly better neurodevelopment at 4.2 years. However, there is a significant fall in VABS score for the whole group between the two assessments, the reason for which is not clear.