If seizures left speechless: CA-P-S C-A-R-E, a proposal of a new ictal language evaluation protocol

Aphasia is a disturbance produced by the alteration of cortical areas involved in language skill elaboration, usually secondary to vascular, tumoral, or inflammatory lesions that disrupt a cortical network encompassing dominant fronto-temporal-parietal regions and non-dominant temporal-parietal areas [1, 2]. An epileptic discharge involving primary language areas may induce transient and reversible aphasia, which may show similar aphasic manifestations as other aetiologies [3, 4]. The characterization of language deficits during ictal testing should be performed soon after ictal onset, in order to differentiate aphasia from other language/cognitive disturbances, obtaining key localizing and lateralizing information [5]. Indeed, some disturbances such as impaired awareness may be related to seizure propagation and should not be used as localizing signs. These may have a significant impact on diagnostic management and treatment options, especially epilepsy surgery [3], affecting patient outcomes. The characteristics of ictal aphasia are not well elucidated, likely due to the intrinsic difficulty of assessing the various language aspects in the limited time frame related to seizure duration, as well as the use of differing methodologies in many previous reports. To date, only a few studies reported a specific and well-described neuropsychological evaluation of language during epileptic seizures but were limited to non-epileptic convulsive status or were focused on post-ictal language evaluation [6‐8]. The latter is the Cincinnati method which consists of presenting, during video-EEG monitoring, a simple sentence through visual channel as soon as a seizure is detected, asking the patient to read the sentence continuously until it is read correctly, in order to detect post-ictal paraphasic errors [6‐8]. Nevertheless, to date, there is no standardized protocol specific for ictal language testing. Recently, an ILAE task force developed an ictal testing battery that could allow standardization among different centers [9]. However, the high number of items used to test multiple ictal symptoms and the absence of open-ended questions may limit the efficacy of this battery, especially with very brief seizures and when dysphasia is present [9]. Furthermore, the inter-observer reliability of ILAE protocol has not been systematically evaluated. Based on these premises, we aimed to create a standardized protocol for language examination in patients with suspected ictal aphasia during video-EEG (VEEG) recording and assessed its efficacy in clinical practice.

We developed an original protocol for ictal language evaluation summarized in the acronym CA-P-S C-A-R-E that means Closed Answers, Pro-speak question, Simple orders, Common object denomination, Audio repetition, Reading, Evoke. The protocol was the result of a multi-step process that benefitted from a specific logopedic analysis of VEEG recordings of seizures with ictal aphasia.

Our study confirmed that the recognition of ictal aphasia and its differentiation from impaired awareness is difficult even for expert clinicians, as with 60% of patients studied with logopedic analysis, it was not possible to fully characterize the ictal language deficit. Beyond the intrinsic limitations of ictal assessment, such as the very brief duration or the overlap of multiple symptoms, heterogeneity in test choice, their sequence of administration, and the lack of formation in language evaluation could significantly affect the efficacy of ictal language examination, leading to a loss of useful semiology information and resulting in a substantial inter-operator/unit variability. To address these critical issues, we elaborated a standardized protocol, effective at testing language in a short period of time. This is particularly important in the pre-surgical evaluation for epilepsy surgery, as an early ictal language impairment may suggest an overlap between the language symptomatogenic zone and the epileptogenic zone. The use of the developed protocol in tested patients allowed the distinction between impaired awareness and ictal aphasia in all seizures. In one case, it was even possible to detect an early language deficit before ictal propagation and consequent impaired awareness. The protocol effectiveness likely benefitted from the laboratory VEEG setting, in which there is a close interaction between patient and examiner. This setting facilitated the recognition of aphasic features that may be easily overlooked if the seizures would be tested late. It was decided to test the protocol interobserver reliability for differentiating between ictal aphasia and awareness impairment, as the full language deficit characterization was challenging even for the expert multidisciplinary team. There was substantial agreement found among a heterogeneous population of health professionals with various degrees of experience, suggesting that the employment of our standardized battery might also reduce the inter-observer variability in differentiating ictal aphasia and awareness impairment in examiners without specific training in language evaluation.

A few studies aimed to standardize testing for ictal language examination. Among them, Loesch et al. proposed a battery of five tasks (remember the word, tell me your name, raise both arms, denominate an object, repeat questions 2–4 until full recovery) [14], while Trebuchon and colleagues demonstrated the usefulness of an experience-based protocol for ictal testing, which was well-structured and easy to use [6].

Even if these protocols are quite complete, we believe that our protocol allows a more comprehensive language evaluation. In order to collect an adequate language sample and facilitate recognition of early dysphasic features, open-ended questions should be administered first. Moreover, both previous protocols do not include simple orders by imitation, which are fundamental to discriminate a comprehension deficit from impaired awareness. The latter is a primary cause of reduced quality of life in epileptic patients [15] and is also considered one of the main criteria used to classify focal seizures [16]. For this reason, different authors attempted to develop specific scales and ictal tests to characterize the level and the content of consciousness during the seizure and in the post-ictal period [17‐19]. Overall, these approaches improve the characterization of ictal semiology and seizure classification but are still not widely implemented and mostly did not take language into account.

The limitations of our study were the small number of ictal video-EEG recordings included in the analysis, the small number of ictal aphasia recordings tested with the developed protocol, the need of a fast and close interaction between patient and examiner to ensure the evaluation efficacy, and choice of the tests and protocol sequence based on single-center experience.

We proposed a protocol made by simple and easy to memorize tests to evaluate language and multiple cognitive functions (visual, auditory, praxia, attention, awareness, memory) during a seizure in the shortest possible time, thus helping the examiners to characterize transient language deficits and differentiate them from impaired awareness. We suggest that it should be used to standardize ictal examination especially in epilepsy monitoring units (EMUs) in order to minimize inter-examiner variability. The proposed ictal testing battery should be validated prospectively in different epilepsy centers.