Neuroimaging of memory in frontal lobe epilepsy

Highlights

• We summarize the existing neuroimaging literature of how fMRI has contributed to our understanding of memory function in FLE.

• Early onset of FLE alters brain network development that may be associated with cognitive and behavioral deficits.

• Brain network development in FLE may be associated with compensation-related mechanisms.

• Understanding FLE as a network disorder can shed light on heterogeneous symptoms and deficits beyond the ictal onset zone.

• Appropriate consideration of clinical variables is critical in understanding brain network organization in FLE.

Abstract

In a large percentage of epilepsies, seizures have focal onset. These epilepsies are associated with a wide range of behavioral and cognitive deficits sometimes limited to the functions encompassed within the ictal onset zone but, more frequently, expanding beyond it. The presence of impairments associated with neuroanatomical areas outside of the ictal onset zone suggests distal propagation of epileptic activity via brain networks and interconnected whole-brain neural circuitry. In patients with frontal lobe epilepsy (FLE), using functional magnetic resonance imaging (fMRI) to identify deficits in working, semantic, and episodic memory may provide a lens through which to understand typical and atypical network organization. A network approach to focal epilepsy is relevant in these patients because of the frequently noted early age of seizure onset. Early seizure-related disruption in healthy brain development may result in a significant brain reorganization, development of compensation-related mechanisms of dealing with function abnormalities and disruptions, and the propagation of epileptic activity from the focus to widespread brain areas (functional deficit zones). Benefits of a network approach in the study of focal epilepsy are discussed along with considerations for future neuroimaging studies of patients with FLE.

Introduction

The term focal onset or localization-related epilepsy refers to seizure generation limited to a specific area [[1], [2], [3]]. Focal epilepsies account for 60–70% of all epilepsies [4]. In addition to seizures coupled with behavioral symptoms of, e.g., sensory disturbance, loss of consciousness, and convulsions, focal epilepsy is associated with transient or lasting/permanent cognitive and behavioral symptoms that result in substantial disability and negative effects on societal and personal outcomes. Early approaches to identifying epileptic tissue for resection focused on establishing the ictal onset zone. However, over time, this concept has evolved and expanded to include identification of many different, sometimes overlapping, zones that either contribute to or are affected by ictal activity [5]. Most recently, the discussion has focused on the “zone” vs. “network” approach and their contributions to seizure, cognitive, and other outcomes. Focal epilepsies often result in a complex interregional interactions across multiple brain regions that include a distributed set of cortical and subcortical regions beyond the ictal onset zone such as areas that produce ictal symptoms, areas of cortical tissue generating interictal spikes, and remote functional deficit zones that produce transient or permanent neurological and psychological effects during interictal periods [6].

In patients with focal epilepsies, abnormal synchronization within and between brain regions affects system level dynamics [[7], [8], [9]]. This often results in a dysfunction of local and remote brain areas that reflects the location of the ictal onset zone and the ictal spread [[7], [8], [9]]. Temporal lobe epilepsy (TLE), the most common type of focal epilepsy, accounts for ~ 50% of focal epilepsies [4], while FLE is the second most common [10]. Patients with TLE report, in addition to the deficits in the executive functions including attention and problem solving, impairments in verbal and nonverbal memory and language that allow lateralization of the ictal onset zone to the dominant or nondominant hemisphere [[11], [12], [13], [14]]. In contrast to TLE, the wide range of impairments observed in FLE, including reduced or variable performance on measures of executive functioning, social cognition, motor planning, memory, and associative learning [[15], [16], [17]], renders precise localization, lateralization, and attribution of deficits to specific brain regions more difficult [18]. The ambiguity in localization and lateralization of deficits in FLE is likely the result of the highly interconnected nature of the frontal lobe, which allows for a quick and widespread propagation of epileptic activity from the frontal lobe to the other brain regions and which affects a range of local and remote functional deficit zones [19].

As TLE is more prevalent and associated with a more robust and consistent set of behavioral and cognitive symptoms than FLE [14,17,18], neuroimaging studies of patients with focal epilepsy tend to focus on deficits and network structure associated with TLE. Typically, the focus of studies on TLE is on two main underlying drivers—cognitive functions that are more accessible for testing (e.g., language or memory encoding/retrieval) and presurgical evaluation. Especially, the second focus is of importance as a possible contributor to surgical outcome compared with the wider/more diverse representation of executive function, attention, and working memory in addition to affective functions [20]. As such, neuroimaging studies of patients with FLE are less common, and considering the overall relatively low prevalence of clinically significant memory deficits in FLE, studies of memory using neuroimaging tools are scarce [21,22]. A network perspective expands a traditional lesion-based approach in order to characterize widespread effects of a focal disruption [8,23,24]. Understanding deficits associated with focal epilepsies through a network perspective allows for discussion regarding underlying whole-brain network structure, healthy and atypical development of brain networks, and cognitive and behavioral deficits associated with areas outside of the ictal onset zone.

One of the most frequent complaints in patients with epilepsy relates to memory problems/difficulties. Memory impairment may pose a significant barrier to completing education and gainful employment, thus negatively impacting quality of life and ability to become a productive member of the society [[11], [12], [13]]. As such, examining relationships between memory and network organization in patients with TLE and FLE can serve as a conduit for understanding the cognitive deficits related to the presence of epileptogenic zone and remote, nonfocal areas while the gained knowledge can be translated into interventions facilitating treatment of such deficits (e.g., various verbal and nonverbal memory complaints) [25,26].

Human memory is typically divided into the following three major stages: sensory memory, short-term (or working) memory, and long-term memory [[27], [28], [29]], with long-term memory broadly categorized into declarative and nondeclarative. Declarative memory is most understood as the following two memory systems: an episodic memory system that deals with autobiographical events specific to time and place and a semantic memory system that encompasses discrete facts and general and personal knowledge about the world [27,29,30]. As these types of memory contribute to various aspects of daily life, the mechanisms and networks underlying these types of memory also differ.

It is important to differentiate memory systems when evaluating memory impairments in patients with focal epilepsy: the brain regions underlying the memory processes may be overlapping with the area(s) of focal seizure onset or with the remote functional deficit zone(s). For example, major areas involved in episodic memory encoding and retrieval encompass medial temporal lobe, medial prefrontal cortex, hippocampus and other subcortical areas, sensorimotor cortex, and cerebellum [31]. While impairments in episodic memory are expected in patients with TLE, the presence of such deficits in patients with FLE would suggest additional mechanisms potentially related to seizure propagation. Thus, the questions posed by this targeted review are as follows:

• How has fMRI contributed to our understanding of working, semantic, and episodic memory functioning in FLE?

• How has fMRI of memory contributed to our understanding of focal epilepsy as a network disorder?

• How may accounting for clinical variables impact the current methods of studying memory in patients with FLE?

The goal of this targeted review was to summarize existing literature addressing these questions and pose considerations for future neuroimaging studies of patients with FLE that address the identified gaps in our knowledge.