In vivo human hippocampal cingulate connectivity: A corticocortical evoked potentials (CCEPs) study

Abstract

Objective

To investigate the human limbic system using cortico-cortical evoked potential (CCEP), which reveals the brain networks.

Methods

Five patients with nonlesional medically intractable focal epilepsy with ictal onset outside the limbic system were enrolled. All patients underwent stereoelectroencephalogram electrode implantation in order to delineate the epileptogenic zone. Alternating 1 Hz electrical stimuli were delivered to the hippocampus and posterior cingulate gyrus. A total of sixty stimuli were averaged in each trial to obtain CCEP responses.

Results

Hippocampal stimulation elicited prominent CCEP responses in the posterior cingulate gyrus. The latencies of early (N1) and late (N2) negative peak ranged 20–60 ms and 102–175 ms respectively. In addition, CCEP responses were observed in the posterior parahippocampal gyrus, medial superior frontal gyrus (SFG) and orbitofrontal cortex. Stimulation of posterior cingulate contacts induced CCEPs in the hippocampus with N1 and N2 latencies of 25–43 ms and 90–234 ms respectively in all five patients.

Conclusion

This finding supports the assertion that the hippocampus is connected with the posterior cingulate gyrus, posterior parahippocampal gyrus, medial SFG and orbitofrontal cortex. The hippocampus and posterior cingulate gyrus have a bidirectional network through the cingulum.

Significance

The present study provides new insight into the human limbic network.

Introduction

In the 1930s, Penfield et al. in the field of epilepsy surgery performed craniotomies and stimulated the cortex with electricity in awake patients, which clarified the cerebral functional localization such as hearing, experience, and memory, as shown in the famous homunculus (Penfield and Boldrev, 1937, Penfield and Flanigin, 1950, Penfield and Jasper, 1954). Fifty years later, Lüders and his colleagues more safely and accurately mapped cerebral functions by chronic intracranial electrodes in patients with epilepsy (Lüders et al., 1985, Lüders et al., 1988, Lesser et al., 1986). This intracranial electrode implantation is now employed routinely for preoperative evaluation to identify epileptic focus. Functional mapping performed in the course of the surgical treatment of epilepsy has not only achieved safe preoperative evaluation but also had a marked impact on the history of human brain science to clarify normal human brain function. However, only the cerebral surface was stimulated in most studies, and deep regions of the limbic system including the hippocampus, amygdaloid body, and cingulate gyrus have not been well evaluated. Papez (1937) proposed that the limbic system involved in memory and emotion is composed of the hippocampus, mammillary body, thalamus, and cingulate gyrus on the basis of animal experiments. Lacruz et al. (2010) reported that bilateral hippocampal stimulation induced memory deficits that was not observed in unilateral hippocampal stimulation using single pulse stimulation technique, and they suggests that recognition memory can be processed independently by the hippocampus on either hemisphere. However, besides these reports, little information is available regarding the limbic network and in vivo human limbic network has not been established. Understanding this network would be useful not only for understanding human higher brain functions, such as memory, recognition, and emotion, but also for elucidating pathophysiology of neurological disorders, such as Alzheimer’s disease and mesial temporal lobe epilepsy.

After 2000, the understanding of cortical functions has been advanced further and human brain network study has recently begun. The electrical stimulation was recently introduced in in vivo human to track the various brain networks (Wilson et al., 1990, Wilson et al., 1991, Greenlee et al., 2007, Lacruz et al., 2007, Keller et al., 2011) and evaluate the cortical excitability (Valentin et al., 2002, Valentin et al., 2005a, Valentin et al., 2005b, Flanagan et al., 2009). Matsumoto et al. (2004b) reported in vivo connection between two speech areas: Broca and Wernicke areas, employing Cortico-Cortical Evoked Potential (CCEP), which facilitated neurophysiological confirmation of neuroconnections that was previously unclear. Subsequently, other CCEP studies have been able to demonstrate primary motor network (Matsumoto et al., 2007), bilateral sensorimotor connections (Terada et al., 2008, Terada et al., 2012), bilateral basal temporal connections (Umeoka et al., 2009), basal temporal language network (Koubeissi et al., 2012), supplementary motor area (Kikuchi et al., 2012), parieto-frontal network (Matsumoto et al., 2012), ictal propagation pathway (Enatsu et al., 2012b), reorganized language network (Enatsu et al., 2013) and cortical excitability (Iwasaki et al., 2010, Enatsu et al., 2012a). Furthermore, CCEP responses are well correlated with diffusion tensor imaging (DTI) tractography (Conner et al., 2011) and resting state functional MRI (Keller et al., 2011). Rosenberg et al. (2009) recorded evoked responses to electrical thalamic stimulation from stereoelectroencephalogram (SEEG) in the similar method in order to reveal thalamocortical connectivity.

In this study, we attempted to identify the functional connectivity between hippocampus and cingulate gyrus, which constitute part of the limbic system, using CCEP methods with accurate SEEG electrodes.