Rose Dawn Bharath
A majority of the resting-state functional MRI literature on epilepsy describes connectivity differences in patients with epilepsy compared with healthy controls. Some studies reported diffuse decreased connectivity, while others found regions with increased connectivity, and still others found both. The diversity of these results is beautifully brought out in the commentary “Functional Connectivity in Mesial Temporal Lobe Epilepsy: A Dynamic Concept.”1 This diversity is what drew us to investigate the neurobiological aspects of epilepsy. Based on Gowers’2 hypothesis that “seizure begets seizures” and the phenomenon of seizure-induced kindling in animal models,3 we hypothesized resting connectivity could be used to study the effect of higher seizure frequency in humans. To control for other disease-related confounders such as type of seizure, duration of epilepsy, age of onset, family history of epilepsy, presence of secondary generalization, and use of antiepileptic drugs, we selected two groups of patients with hot water epilepsy. These two groups were similar in demographic- and disease-related variables except for seizure frequency. Our work revealed that patients with frequent seizures had widespread network abnormalities involving the default mode network, with increased connectivity of the medial temporal lobe structures akin to the phenomenon of “kindling.”
Prior to our study, when we discussed epilepsy imaging in our journal clubs, we paid little attention to the several disease-related confounders that could alter connectivity. Now, we look at the mention of these variables in articles dealing with epilepsy. This has also changed the quality of our participation in epilepsy meetings at our institute.
The in vitro potential of resting-state fMRI in understanding disease neurobiology has incited interest in our neurology and neurosurgery colleagues, and they seem keen to use this technique in their research.
As a follow-up to this study, we used graph theory measures to work on another disease-related variable—the family history of epilepsy—and found that there are connectivity changes proportional to the familial burden of the disease.