Hippocampal abnormality and response to vagus nerve stimulation in epilepsy

Vagus nerve stimulation (VNS) reduces seizure frequency and severity in some, but not all, individuals with epilepsy. The hippocampus has been implicated in VNS response, but is yet to be studied structurally using T1-weighted (T1w) magnetic resonance imaging (MRI). In this study we hypothesized greater hippocampal abnormality in VNS non-responders. Using hippocampal morphometrics, we extracted the volumes of four hippocampal regions from T1w MRI across three groups; VNS responders ( <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:semantics><mml:mrow><mml:mi>n</mml:mi> <mml:mo>=</mml:mo> <mml:mn>42</mml:mn></mml:mrow> </mml:semantics> </mml:math> ), non-responders ( <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:semantics><mml:mrow><mml:mi>n</mml:mi> <mml:mo>=</mml:mo> <mml:mn>50</mml:mn></mml:mrow> </mml:semantics> </mml:math> ), and healthy controls ( <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:semantics><mml:mrow><mml:mi>n</mml:mi> <mml:mo>=</mml:mo> <mml:mn>100</mml:mn></mml:mrow> </mml:semantics> </mml:math> ). We first calculated the multivariate Mahalanobis distance using z-scores from all four hippocampal regions to measure abnormality relative to controls. We then compared traditional univariate measures to the Mahalanobis distance. Response to VNS was defined as having a <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:semantics><mml:mrow><mml:mo>≥</mml:mo> <mml:mn>50</mml:mn> <mml:mo>%</mml:mo></mml:mrow> </mml:semantics> </mml:math> seizure reduction 2 years post-implantation. Hippocampal morphometrics were significantly more abnormal in non-responders than responders ( <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:semantics><mml:mrow><mml:mi>p</mml:mi> <mml:mo>=</mml:mo> <mml:mn>.005</mml:mn> <mml:mo>,</mml:mo> <mml:mtext>biserial</mml:mtext> <mml:mspace></mml:mspace> <mml:mi>r</mml:mi> <mml:mo>=</mml:mo> <mml:mn>.32</mml:mn></mml:mrow> </mml:semantics> </mml:math> ) using the multivariate Mahalanobis distance. Univariate approaches did not differ significantly between responders and non-responders ( <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:semantics><mml:mrow><mml:mi>p</mml:mi> <mml:mo>></mml:mo> <mml:mn>.05</mml:mn></mml:mrow> </mml:semantics> </mml:math> ). At the group level, non-responders to VNS had greater structural hippocampal abnormality when using the multivariate approach. Conversely, this effect was lost with the univariate analysis. This suggests that abnormality is likely present in different parts of the hippocampus in different individuals. Future studies should incorporate multivariate, and potentially multi-modal, information to better characterize the mechanisms of VNS response.