NO EXCESS OF NEUROIMAGING ABNORMALITIES ASSOCIATED WITH ROLANDIC EPILEPSY

Objectives: Rolandic epilepsy (RE) is an idiopathic epilepsy syndrome. A number of recent uncontrolled case reports have suggested an excess of neuroimaging abnormalities, both epilepsy specific as well as incidental. However, association with structural abnormalities may be influenced by observer bias. In order to test the hypothesis of excess imaging abnormalities, we compared the prevalence of specific and non-specific abnormalities in Rolandic epilepsy patients and controls.

Methods: The neuroimaging of 16 RE patients and 16 pediatric recurrent headache controls was evaluated independently by two neuroradiologists. The evaluators were blind both to the disease status of the subjects, and to the study hypothesis. They evaluated the presence of hippocampal and perihippocampal abnormalities, congenital malformations, subcortical and periventricular parenchymal hyperintensities, and dilated perivascular spaces; they also scored imaging on criteria of quality, sequences, and number of studies. We calculated odds ratios of association.

Results: The mean age of cases and controls differed slightly (cases 8.0 years, controls 12.2). The quality and quantity of imaging information did not differ between groups. 30 subjects had MRIs, 2 had CTs only. There were no significant overall differences in the frequency of abnormalities between groups (OR 1.1, 95% CI: 0.5–2.5). In the specific categories, there was (1 case, 0 control) with hippocampal abnormality; (4,2) congenital malformations; (1,3) subcortical parenchymal hyperintensities; (3,4) periventricular parenchymal hyperintensities; (4,3) dilated perivascular spaces. Inter-rater agreement was excellent, ranging from kappa of 0.34 to 1.0.

Conclusion: We found no evidence for an excess of neuroimaging abnormalities associated with RE. This is as expected for an idiopathic epilepsy syndrome. The previous reports of excess pathological or incidental findings likely reflects observer bias.