The different patterns of seizure-induced aphasia in temporal lobe epilepsies

Highlights

• Assessment of language function in the early ictal state provides valuable information on seizure localization.

• Impaired speech comprehension was associated with posterior lateral involvement of Left TL.

• Anomia and reduced fluency without comprehension deficit was associated with left anterior and medio-basal TL involvement.

• The language production deficits cannot be explained by an involvement of Broca’s area.

• Jargonaphasia was associated with left basal temporal involvement.

Abstract

Objectives

Ictal language disturbances may occur in dominant hemisphere temporal lobe epilepsy (TLE), but little is known about the precise anatomoelectroclinical correlations. This study investigated the different facets of ictal aphasia in intracerebrally recorded TLE.

Methods

Video-stereoelectroencephalography (SEEG) recordings of 37 seizures in 17 right-handed patients with drug-resistant TLE were analyzed; SEEG electroclinical correlations between language disturbance and involvement of temporal lobe structures were assessed. In the clinical analysis, we separated speech disturbance from loss of consciousness.

Results

According to the region involved, different patterns of ictal aphasia in TLE were identified. Impaired speech comprehension was associated with posterior lateral involvement, anomia and reduced verbal fluency with anterior mediobasal structures, and jargonaphasia with basal temporal involvement. The language production deficits, such as anomia and low fluency, cannot be simply explained by an involvement of Broca's area, since this region was not affected by seizure discharge.

Significance

Assessment of language function in the early ictal state can be successfully performed and provides valuable information on seizure localization within the temporal lobe as well as potentially useful information for guiding surgery.

Introduction

It has long been known that seizures arising from the dominant temporal lobe of the language-dominant hemisphere may elicit speech disturbances. Paroxysmal dysphasia was associated with left hemisphere lesions by Jackson [1]. In the premonitoring era, this phenomenon was described in a patient with left temporal lobe epilepsy who became seizure-free after temporal lobectomy [2]. Later, based on video-EEG recordings, seizures arising in left temporal lobe were shown to be strongly associated with postictal language disturbance [3], [4], but ictal aphasia or dysphasia was not precisely characterized.

In addition, the overlap between loss of consciousness and language disturbance is a difficult problem, which is not easy to resolve, and may explain why in some studies left TLE have been more frequently associated with loss of consciousness than right TLEs [5].

In the last ten years, detailed evaluation of consciousness using multiple criteria scales have been proposed to evaluate loss of consciousness and distinguish this from language impairment [6], [7], [8]. The different proposed consciousness rating scales included (1) nonverbal response (orientation, gestural interaction), (2) verbal response (e.g., ability to follow commands requiring verbal answers, naming), and (3) the ability to recall events that occurred during the seizure.

Although aphasia may be the most prominent or sole manifestation of seizures [9], [10], the description of its semiology remains vague, and even if changes in semiological terminology have been made, recommending that partial seizures should be now described according to their “dyscognitive” features [11], ictal aphasia has been included in the automatism section. “Dysphasic seizure” is, thus, described as “impaired communication involving language without dysfunction of relevant primary or sensory pathways manifested as impaired comprehension, anomia, paraphasic errors, or a combination of these” [12]. No distinction has yet been proposed between the different patterns of ictal aphasia, as it has been described in stroke [13].

In a review of ictal aphasia, Benatar pointed out that the literature on this topic is relatively scant and most of the papers included in the review dealt with status epilepticus [9]. On the basis of several well-studied reported cases, the author concluded that potentially any type of aphasia might occur as an ictal sign. In addition, seizures do not seem to predominantly produce any single type of aphasia. Finally, it is difficult to establish a straightforward correlation between the type of aphasia and location of the epileptogenic zone.

The lack of data about ictal aphasia could be explained by the difficulty of testing and evaluating language during the brief and transient state of a seizure.

The aims of the present study were (1) to describe the different patterns of ictal aphasia in temporal lobe seizures and (2) to identify precise anatomoelectroclinical correlations using intracerebral recordings (stereoelectroencephalography, SEEG) defining the contribution of different regions (including temporal lobe regions but also frontal and insular regions) involved in temporal lobe seizures with ictal aphasia. The SEEG approach allows sampling of multiple distant areas included in the network language with high spatial and temporal resolution. To differentiate loss of consciousness from speech disturbance, we selected seizures with early clinical testing and with preserved nonverbal interaction (that is, excluding seizures with significant loss of consciousness using the operant definition). The aim was to establish specific electroclinical patterns.