Degree of hippocampal atrophy is related to side of seizure onset in temporal lobe epilepsy

Abstract

Background and purpose: Temporal lobe epilepsy (TLE) is associated with pathologic changes in hippocampal physiology and morphology. Our aim was to quantify volume reduction of the right and left hippocampus in patients with TLE and to investigate whether the degree of hippocampal atrophy is related to the side of seizure onset.

Methods: The volume of the right and left hippocampus was estimated for 50 controls and 101 patients with TLE, by applying the unbiased Cavalieri method on MR images.

Results: Pairwise comparisons, within a multivariate analysis of variance and adjusted by using the Bonferroni correction, revealed that both right and left hippocampal volumes were, on average, significantly smaller in patients with right-sided seizure onset (R-patients) relative to those of controls (P < .001 and P = .04, respectively). Furthermore, left hippocampal volume was significantly smaller in patients with left-sided seizure onset (L-patients) compared with controls (P < .001), but the right-sided hippocampal volume was not significantly smaller (P = .71). Moreover, a correlation analysis revealed that the strong linear association between the right and left hippocampal volumes existing in the control population (r = 0.73) is partially lost in patients with TLE (r < or = 0.48), and this loss in correlation appears to be more pronounced in L-patients than in R-patients.

Conclusion: Our MR imaging results suggest that although the major damage in patients with TLE is located in the hippocampus ipsilateral to the side of seizure onset, R-patients are more likely to have bilateral hippocampal volume reduction. These findings support the hypothesis that cerebral hemispheres may not only differ in their functionality organization but also in their vulnerability to a neurologic insult.

Fig 1.

Illustration of the application of the Cavalieri method to estimate hippocampal volume.

Fig 2.

Illustration of the point-counting technique applied to estimate hippocampal volume from MR images of a control (C, top row), patient with left-sided seizure onset (LP, second row), and patient with right-sided seizure onset (RP, bottom row).

Fig 3.

Illustration of the boundaries of the hippocampus. The splitting of the lateral ventricles form the posterior border (108, Split LV), and the alveus forms the anterior border (150).

Fig 4.

Empirical probability distributions and Q-Q plots for the right and left volume and asymmetry index of the hippocampus of the control population.

Fig 5.

Left and right hippocampal volume estimate (left panel) and hippocampal volume asymmetry index (right panel) for controls and patients with TLE. The dashed lines represent the 99% prediction lower bounds for the left and right hippocampal volume estimate (left panel) and the 99% prediction interval for the hippocampal volume asymmetry index (right panel). The proportion of patients showing abnormal hippocampal volume is indicated in parentheses.

Fig 6.

Left hippocampal volume versus right hippocampal volume in R-patients (top left panel), L-patients (top right panel), and controls (bottom panel). The dashed lines represent the 99% lower bounds for the right and left hippocampal volume obtained from the control data.