Electroclinical characteristics of seizures arising from the precuneus based on stereoelectroencephalography (SEEG)

The precuneus is a discrete area located in the posterior region of the medial parietal cortex, neighbored anteriorly by the marginal branch of the cingulate cortex, posteriorly by the parietal-occipital fissure, and inferiorly by the subparietal sulcus [1]. Anatomical and connectivity studies have indicated that the precuneus belongs to a widespread network of cortical and subcortical structures [2]. Primary connections of the precuneus include the posterior cingulate and retrosplenial cortices, other regions of the parietal cortex, the frontal cortex, and the temporal-parietal-occipital area [1]. Among these connections, we focus on the frontoparietal circuits, as they are regarded as the main elements of the cortical motor system [3]. Each motor area receives afferents from a specific set of parietal territories. Studies have shown that area 5ci [4], which is in the anterior medial region of the precuneus, is essentially the secondary somatosensory area due to its close association with the supplementary motor area (SMA), and that areas 5 L and 5 M are connected to the superior parietal lobule (SPL), medial primary somatosensory area (SI), and motor cortex. The central region of the precuneus exhibits overlap with the mesial 7A and is connected to the dorsolateral and dorsomedial prefrontal cortices [5, 6].

Seizures arising from the precuneus are rare, and few studies have aimed to characterize the clinical presentation of such seizures within the anatomic context of the frontoparietal circuits. Previous studies are limited in that their analyses focused on parietal lobe epilepsy as a general entity [7]. These studies showed that an important feature of parietal lobe epilepsy is the polymorphism of ictal manifestation [8]. Parietal lobe epilepsy is usually associated with various auras, including somatosensory impairments; disturbances of body image; vertiginous sensations; and visual, auditory, or aphasic auras [9, 10]. Focal motor clonic activity, tonic posturing, and oral-gestural automatisms occur in 57, 28, and 17% of patients, respectively [11]. More recent studies have investigated seizures arising from subregions of the parietal lobe, including the precuneus. Seizures arising from the precuneus are more frequently associated with body movement sensations, visual auras, eye movements, vestibular manifestations, asymmetric tonic posturing, and hypermotor activity [1]. Precuneal epilepsy is difficult to differentiate from other types of epilepsy—particularly frontal lobe epilepsy—because seizure onset occurs at an anatomically deep and semiologically silent area [12]. As such seizures may spread to the frontal cortex, misinterpretation of the clinical manifestations can lead to incorrect localization. False localization has been reported in up to 16% of cases [13]. Hence, it is important to improve our understanding of the clinical and neurophysiological features of precuneal epilepsy to improve pre-operative evaluation and surgical strategies.

Stereoelectroencephalography (SEEG) offers distinct advantages over conventional noninvasive approaches, which may be insufficient due to the deep origin of precuneal epilepsy. First introduced by Talairach and Bancaud in the early 1960’s [14], SEEG is an invasive approach that can be used to reliably identify deeply buried anatomic structures, allowing one to construct a dynamic, three-dimensional (3D) spatiotemporal picture of epileptic activity [15]. Indeed, previous studies have utilized SEEG to precisely investigate the electrical activity of different brain structures involved in seizure generation and propagation [10, 16] .

In focal epilepsy, the symptoms are determined by the dynamic evolution of epileptic discharge from the initial seizure onset zone to the adjacent/distant cortical or subcortical areas. Therefore, characterizing the temporo-spatial evolution of seizure propagation and symptoms is crucial for improving our understanding of different clinical phenotypes [17, 18]. In the present study, we investigated the electroclinical characteristics of epilepsy originating from the precuneus, as well as the correlation between clinical phenotypes and the ictal pattern of seizure propagation.

In the present study, we analyzed semiological characteristics and EEG/SEEG data in order to identify anatomic–electrical–clinical relationships in 10 patients with precuneal epilepsy. Previous studies have revealed that precuneal epilepsy is associated with several types of premonitory auras [1]. In one such study, visual auras were reported in three patients (50%), vestibular responses in two patients (33%), and sensations of falling or movement in one patient [10]. Previous studies have demonstrated that stimulation of the anterior portion of the precuneus results in body image disturbance, while stimulation of the posterior portion results in visuomotor illusions [21, 22]. Additional studies have reported that stimulation of the precuneus evokes vestibular responses [23]. Our findings are mostly consistent with these previous studies, although body image disturbance occurred more frequently in our patient group, perhaps because most patients exhibited seizure activity in the anterior precuneus.

We observed two major patterns of activity in patients of the present study: BATS and HMS. BATS is often associated with seizure activity in the mesial parietal region [24], similar to patterns observed in patients with mesial frontal lobe epilepsy. However, seizures originating from the precuneus are typically briefer than those with an SMA onset, and patients with precuneal seizures are typically quicker to regain awareness [25]. These features were also observed in our patients who experienced BATS.

Some studies have reported that seizures originating from the lateral and mesial parietal regions may manifest as hypermotor activity [26]. Such hypermotor activity was also observed in some patients of the present study. However, HMS typically has a frontal lobe onset and occur in 15–27% of patients with frontal lobe epilepsy, in which seizures primarily rise from the orbitofrontal or mesial frontal cortex [27‐29]. In such cases, it is difficult to distinguish seizures arising from the parietal lobe from those arising from the frontal lobe. However, some studies have indicated that the type of aura and the delay between seizure onset and hypermotor activity can be used to make this distinction [26]. As previously mentioned, precuneal HMS is often preceded by unique auras (e.g., body image disturbance and vestibular responses). In addition, patients with frontal lobe seizures almost always experience hypermotor activity at seizure onset or during the first third of the seizure [30]. However, in patients with precuneal seizures, there is usually a substantial delay between seizure onset and the appearance of hypermotor activity, which may represent the time required for the seizure to propagate from the precuneus to the frontal lobe or to subcortical structures [31, 32].

In precuneal epilepsy, the interictal EEG provides little clue to the lateralization and localization of seizure activity [9], as discharge is prone to spread to other cerebral lobes and induce an error in localization. The widespread connectivity between the precuneus and other cerebral regions may represent the underlying cause of errors in localization. Hence, it is difficult to distinguish precuneal from non-precuneal seizures via scalp EEG, particularly those arising from the mesial frontal lobe. SEEG represents a unique method for identifying and characterizing the underlying physiologic mechanisms of precuneal seizures.

In SEEG methodology, the EZ is defined as the site of origin for epileptic seizures [33]. Therefore, we focused our SEEG analyses mainly on the identification of the brain regions in which specific ictal patterns developed during seizures [34, 35]. The region associated with the primary organization of ictal discharge and subsequent propagation is correlated with the anatomical positions of the electrodes that display maximal abnormal activity, and with the evolution of clinical seizure signs. At present, however, the identification of the EZ is based on the ability of experienced clinical neurophysiologists to identify the relevant anatomic–electrical–clinical correlations and SEEG patterns.

In the present study, we analyzed the ictal SEEG of five patients in whom the relevant electrodes involved the frontoparietal network. In three cases, BATS was observed only once the ictal activity had spread to the SMA, PCL, PrCG, or PoCG. Our results indicated that abnormal ictal activity is easily propagated from the precuneus to these regions, suggesting that the precuneus is functionally connected to the SMA and other central areas. Some previous studies have reported that the medial aspect of the posterior parietal lobe is connected to the SMA and premotor areas [4, 36]. Therefore, seizures occurring within the medial side of the parietal lobe may spread to the SMA, causing bilateral asymmetric tonic manifestations [24].

In two patients of the present study, HMS was observed only when ictal activity had spread to the PCC and MCC. Some studies have reported that the middle cingulate gyrus receives afferent connections from the medial surface of the parietal lobe, particularly the precuneus. Such connections provide information regarding noxious stimulation in the body, directing the motor area of the cingulate gyrus to initiate evasive actions/behaviors. Thus, seizure activity originating from the precuneus may induce hypermotor movements [37]. In one case series, two patients with electrodes implanted in the motor aspect of the MCC exhibited fast discharge in this region during hypermotor movements [26]. Thus, hypermotor activity associated with precuneal seizures may be due to the ease with which abnormal ictal activity propagates from the precuneus to the PCC and MCC. Taken together, the accumulated evidence suggests that the precuneus is functionally connected with the PCC and MCC.

The present study possesses several limitations of note. First, the study was limited by the number and coverage of the implanted electrodes. The spatial limitations of recording may have caused difficulties in identifying all possible pathways of seizure propagation. The study was also limited by the small number of included cases and its retrospective design. Nonetheless, our study provides valuable evidence regarding the electroclinical characteristics of precuneal epilepsy, as well as the possible mechanisms underlying variations in semiology based on SEEG findings within frontoparietal circuits. Such findings may represent clinical indicators of precuneal epilepsy. Future studies should investigate further subdivisions of the precuneus using a greater number of electrodes in order to provide a more complete characterization of precuneal epilepsy.

The clinical characteristics of precuneal epilepsy are complex and various. Aura type (e.g., body image disturbance and vestibular response), BATS, and HMS are the main indicators of precuneal epilepsy. Scalp EEG is of little use when attempting to localize precuneal seizures. Analysis of high-quality SEEG data can help to identify the relevant anatomic-electrical-clinical correlations. Our findings indicate that the clinical characteristics of precuneal epilepsy are different among patients, and that the final electro–clinical phenotype depends on the pattern of seizure spread.