Functional MRI and neuropsychological evidence for language plasticity before and after surgery in one patient with left temporal lobe epilepsy

Disentangling neuroplasticity mechanisms in post-stroke language recovery

A major objective in post-stroke aphasia research is to gain a deeper understanding of neuroplastic mechanisms that drive language recovery, with the ultimate goal of enhancing treatment outcomes. Subsequent to recent advances in neuroimaging techniques, we now have the ability to examine more closely how neural activity patterns change after a stroke. However, the way these neural activity changes relate to language impairments and language recovery is still debated. The aim of this review is to provide a theoretical framework to better investigate and interpret neuroplasticity mechanisms underlying language recovery in post-stroke aphasia. We detail two sets of neuroplasticity mechanisms observed at the synaptic level that may explain functional neuroimaging findings in post-stroke aphasia recovery at the network level: feedback-based homeostatic plasticity and associative Hebbian plasticity. In conjunction with these plasticity mechanisms, higher-order cognitive control processes dynamically modulate neural activity in other regions to meet communication demands, despite reduced neural resources. This work provides a network-level neurobiological framework for understanding neural changes observed in post-stroke aphasia and can be used to define guidelines for personalized treatment development.

The effect of different brain lesions on the reorganization of language functions within the dominant hemisphere assessed with task-based BOLD-fMRI

BACKGROUND AND PURPOSE

Language reorganization has been described in brain lesions with respect to their location and timing, but little is known with respect to their etiology. We used fMRI to investigate the effects of different types of left hemisphere lesions (GL = gliomas, TLE = temporal lobe epilepsy and CA = cavernous angioma) on the topographic intra-hemispheric language plasticity, also considering their location.

METHODS

Forty-seven right-handed patients with 3 different left hemisphere lesions (16 GL, 15 TLE and 16 CA) and 17 healthy controls underwent BOLD fMRI with a verb-generation task. Euclidean distance was used to measure activation peak shifts among groups with respect to reference Tailarach coordinates of Inferior Frontal Gyrus, Superior Temporal Sulcus and Temporo-Parietal Junction. Mixed-model ANOVAs were used to test for differences in activation peak shifts.

RESULTS

Significant activation peak shifts were found in GL patients with respect both to HC and other groups (TLA and CA). In addition, in the same group of patients a significant effect of tumor location (anterior or posterior) was detected.

CONCLUSIONS

We demonstrated that intra-hemispheric language plasticity is influenced by the type of lesion affecting the left hemisphere and that fMRI is especially valuable in the preoperative assessment of such reorganization in glioma patients.

Interactive mapping of language and memory with the GE2REC protocol

Previous studies have highlighted the importance of considering cognitive functions from a dynamic and interactive perspective and multiple evidence was brought for a language and memory interaction. In this study performed with healthy participants, we present a new protocol entitled GE2REC that interactively accesses the neural representation of language-and-memory network. This protocol consists of three runs related to each other, providing a link between tasks, in order to assure an interactive measure of linguistic and episodic memory processes. GE2REC consists of a sentence generation (GE) in the auditory modality and two recollecting (2REC) memory tasks, one recognition performed in the visual modality, and another one recall performed in the auditory modality. Its efficiency was evaluated in 20 healthy volunteers using a 3T MR imager. Our results corroborate the ability of GE2REC to robustly activate fronto-temporo-parietal language network as well as temporal mesial, prefrontal and parietal cortices in encoding during sentence generation and recognition. GE2REC is useful because it: (a) requires simultaneous and interactive language-and-memory processes and jointly maps their neural basis; (b) explores encoding and retrieval, managing to elicit activation of mesial temporal structures; (c) is easy to perform, hence being suitable for more restrictive settings, and (d) has an ecological dimension of tasks and stimuli. GE2REC may be useful for studying neuroplasticity of cognitive functions, especially in patients with temporal lobe epilepsy who show reorganization of both language and memory networks. Overall, GE2REC can provide valuable information in terms of the practical foundation of exploration language and memory interconnection.

Tracking longitudinal language network reorganisation using functional MRI connectivity fingerprints

•

FMRI connectivity fingerprints represent patient-unique language networks.

•

Fingerprints can be statistically tested to detect reorganisation in individuals.

•

Connectivity fingerprints track surgery-related adaptations in individual patients.

•

Network-level changes appear related to presence of language symptoms.

Large individual differences in how brain networks respond to treatment hinder efforts to personalise treatment in neurological conditions. We used a brain network fingerprinting approach to longitudinally track re-organisation of complementary phonological and semantic language networks in 19 patients before and after brain-tumour surgery. Patient task fingerprints were individually compared to normal networks established in 17 healthy controls. Additionally, pre- and post-operative patient fingerprints were directly compared to assess longitudinal network adaptations. We found that task networks remained stable over time in healthy controls, whereas treatment induced reorganisation in 47.4% of patient fluency networks and 15.8% of semantic networks. How networks adapted after surgery was highly unique; a subset of patients (10%) showed ‘normalisation’ while others (21%) developed newly atypical networks after treatment. The strongest predictor of adaptation of the fluency network was the presence of clinically reported language symptoms. Our findings indicate a tight coupling between processes disrupting performance and neural network adaptation, the patterns of which appear to be both task- and individually-unique. We propose that connectivity fingerprinting offers potential as a clinical marker to track adaptation of specific functional networks across treatment interventions over time.

Effects of Epilepsy on Language Functions: Scoping Review and Data Mining Findings

PURPOSE

This study involved a scoping review to identify possible gaps in the empirical description of language functioning in epilepsy in adults. With access to social network data, data mining was used to determine if individuals with epilepsy are expressing language-related concerns.

METHOD

For the scoping review, scientific databases were explored to identify pertinent articles. Findings regarding the nature of epilepsy etiologies, patient characteristics, tested language modalities, and language measures were compiled. Data mining focused on social network databases to obtain a set of relevant language-related posts.

RESULTS

The search yielded 66 articles. Epilepsy etiologies except temporal lobe epilepsy and older adults were underrepresented. Most studies utilized aphasia tests and primarily assessed single-word productions; few studies included healthy control groups. Data mining revealed several posts regarding epilepsy-related language problems, including word retrieval, reading, writing, verbal memory difficulties, and negative effects of epilepsy treatment on language.

CONCLUSION

Our findings underscore the need for future specification of the integrity of language in epilepsy, particularly with respect to discourse and high-level language abilities. Increased awareness of epilepsy-related language issues and understanding the patients' perspectives about their language concerns will allow researchers and speech-language pathologists to utilize appropriate assessments and improve quality of care.

Assessment of grammar optimizes language tasks for the intracarotid amobarbital procedure

PURPOSE

A previous study showed that assessment of language laterality could be improved by adding grammar tests to the recovery phase of the intracarotid amobarbital procedure (IAP) (Połczyńska et al. 2014). The aim of this study was to further investigate the extent to which grammar tests lateralize language function during the recovery phase of the IAP in a larger patient sample.

METHODS

Forty patients with drug-resistant epilepsy (14 females, thirty-two right-handed, mean age 38.5years, SD=10.6) participated in this study. On EEG, 24 patients had seizures originating in the left hemisphere (LH), 13 in the right hemisphere (RH), and 4 demonstrated mixed seizure origin. Thirty participants (75%) had bilateral injections, and ten (25%) had unilateral injections (five RH and five LH). Based on results from the encoding phase, we segregated our study participants to a LH language dominant and a mixed dominance group. In the recovery phase of the IAP, the participants were administered a new grammar test (the CYCLE-N) and a standard language test. We analyzed the laterality index measure and effect sizes in the two tests.

KEY FINDINGS

In the LH-dominant group, the CYCLE-N generated more profound language deficits in the recovery phase than the standard after injection to either hemisphere (p<0.001). At the same time, the laterality index for the grammar tasks was still higher than for the standard tests. Critically, the CYCLE-N administered in the recovery phase was nearly as effective as the standard tests given during the encoding phase.

SIGNIFICANCE

The results may be significant for individuals with epilepsy undergoing IAP. The grammar tests may be a highly efficient measure for lateralizing language function in the recovery phase.

Machine learning-XGBoost analysis of language networks to classify patients with epilepsy

Our goal was to apply a statistical approach to allow the identification of atypical language patterns and to differentiate patients with epilepsy from healthy subjects, based on their cerebral activity, as assessed by functional MRI (fMRI). Patients with focal epilepsy show reorganization or plasticity of brain networks involved in cognitive functions, inducing ‘atypical’ (compared to ‘typical’ in healthy people) brain profiles. Moreover, some of these patients suffer from drug-resistant epilepsy, and they undergo surgery to stop seizures. The neurosurgeon should only remove the zone generating seizures and must preserve cognitive functions to avoid deficits. To preserve functions, one should know how they are represented in the patient’s brain, which is in general different from that of healthy subjects. For this purpose, in the pre-surgical stage, robust and efficient methods are required to identify atypical from typical representations. Given the frequent location of regions generating seizures in the vicinity of language networks, one important function to be considered is language. The risk of language impairment after surgery is determined pre-surgically by mapping language networks. In clinical settings, cognitive mapping is classically performed with fMRI. The fMRI analyses allowing the identification of atypical patterns of language networks in patients are not sufficiently robust and require additional statistic approaches. In this study, we report the use of a statistical nonlinear machine learning classification, the Extreme Gradient Boosting (XGBoost) algorithm, to identify atypical patterns and classify 55 participants as healthy subjects or patients with epilepsy. XGBoost analyses were based on neurophysiological features in five language regions (three frontal and two temporal) in both hemispheres and activated with fMRI for a phonological (PHONO) and a semantic (SEM) language task. These features were combined into 135 cognitively plausible subsets and further submitted to selection and binary classification. Classification performance was scored with the Area Under the receiver operating characteristic curve (AUC). Our results showed that the subset SEM_LH BA_47-21 (left fronto-temporal activation induced by the SEM task) provided the best discrimination between the two groups (AUC of 91 ± 5%). The results are discussed in the framework of the current debates of language reorganization in focal epilepsy.

Post-Surgical Language Reorganization Occurs in Tumors of the Dominant and Non-Dominant Hemisphere

PURPOSE

Surgical resection of brain tumors may shift the location of cortical language areas. Studies of language reorganization primarily investigated left-hemispheric tumors irrespective of hemispheric language dominance. We used functional magnetic resonance imaging (fMRI) to investigate how tumors influence post-surgical language reorganization in relation to the dominant language areas.

METHODS

A total of, 17 patients with brain tumors (16 gliomas, one metastasis) in the frontotemporal and lower parietal lobes planned for awake surgery underwent pre-surgical and post-surgical language fMRI. Language activation post-to-pre surgery was evaluated visually and quantitatively on the statistically thresholded images on patient-by-patient basis. Results were qualitatively compared between three patient groups: temporal, with tumors in the dominant temporal lobe, frontal, with tumors in the dominant frontal lobe and remote, with tumors in the non-dominant hemisphere.

RESULTS

Post-to-pre-surgical distributions of activated voxels changed in all except the one patient with metastasis. Changes were more pronounced in the dominant hemisphere for all three groups, showing increased number of activated voxels and also new activation areas. Tumor resection in the dominant hemisphere (frontal and temporal) shifted the activation from frontal towards temporal, whereas tumor resection in the non-dominant hemisphere shifted the activation from temporal towards frontal dominant areas.

CONCLUSION

Resection of gliomas in the dominant and in the non-dominant hemisphere induces postsurgical shifts and increase in language activation, indicating that infiltrating gliomas have a widespread influence on the language network. The dominant hemisphere gained most of the language activation irrespective of tumor localization, possibly reflecting recovery of pre-surgical tumor-induced suppression of these activations.

NEREC, an effective brain mapping protocol for combined language and long-term memory functions

Temporal lobe epilepsy can induce functional plasticity in temporoparietal networks involved in language and long-term memory processing. Previous studies in healthy subjects have revealed the relative difficulty for this network to respond effectively across different experimental designs, as compared to more reactive regions such as frontal lobes. For a protocol to be optimal for clinical use, it has to first show robust effects in a healthy cohort. In this study, we developed a novel experimental paradigm entitled NEREC, which is able to reveal the robust participation of temporoparietal networks in a uniquely combined language and memory task, validated in an fMRI study with healthy subjects. Concretely, NEREC is composed of two runs: (a) an intermixed language-memory task (confrontation naming associated with encoding in nonverbal items, NE) to map language (i.e., word retrieval and lexico-semantic processes) combined with simultaneous long-term verbal memory encoding (NE items named but also explicitly memorized) and (b) a memory retrieval task of items encoded during NE (word recognition, REC) intermixed with new items. Word recognition is based on both perceptual-semantic familiarity (feeling of 'know') and accessing stored memory representations (remembering). In order to maximize the remembering and recruitment of medial temporal lobe structures, we increased REC difficulty by changing the modality of stimulus presentation (from nonverbal during NE to verbal during REC). We report that (a) temporoparietal activation during NE was attributable to both lexico-semantic (language) and memory (episodic encoding and semantic retrieval) processes; that (b) encoding activated the left hippocampus, bilateral fusiform, and bilateral inferior temporal gyri; and that (c) task recognition (recollection) activated the right hippocampus and bilateral but predominant left fusiform gyrus. The novelty of this protocol consists of (a) combining two tasks in one (language and long-term memory encoding/recall) instead of applying isolated tasks to map temporoparietal regions, (b) analyzing NE data based on performances recorded during REC, (c) double-mapping networks involved in naming and in long-term memory encoding and retrieval, (d) focusing on remembering with hippocampal activation and familiarity judgment with lateral temporal cortices activation, and (e) short duration of examination and feasibility. These aspects are of particular interest in patients with TLE, who frequently show impairment of these cognitive functions. Here, we show that the novel protocol is suited for this clinical evaluation.

What do patients with epilepsy tell us about language dynamics? A review of fMRI studies

The objective of this review is to resume major neuroimaging findings on language organization and plasticity in patients with focal and refractory epilepsy, to discuss the effect of modulatory variables that should be considered alongside patterns of reorganization, and to propose possible models of language reorganization. The focal and refractory epilepsy provides a real opportunity to investigate various types of language reorganization in different conditions. The 'chronic' condition (induced by the epileptogenic zone or EZ) is associated with either recruitment of homologous regions of the opposite hemisphere or recruitment of intrahemispheric, nonlinguistic regions. In the 'acute' condition (neurosurgery and EZ resection), the initial interhemispheric shift (induced by the chronic EZ) could follow a reverse direction, back to the initial hemisphere. These different patterns depend on several modulatory factors and are associated with various levels of language performance. As a neuroimaging tool, functional magnetic resonance imaging enables the detailed investigation of both hemispheres simultaneously and allows for comparison with healthy controls, potentially creating a more comprehensive and more realistic picture of brain-language relations. Importantly, functional neuroimaging approaches demonstrate a good degree of concordance on a theoretical level, but also a considerable degree of individual variability, attesting to the clinical importance with these methods to establish, empirically, language localization in individual patients. Overall, the unique features of epilepsy, combined with ongoing advances in technology, promise further improvement in understanding of language substrate.

Inter-hemispheric language functional reorganization in low-grade glioma patients after tumour surgery

Despite many claims of functional reorganization following tumour surgery, empirical studies that investigate changes in functional activation patterns are rare. This study investigates whether functional recovery following surgical treatment in patients with a low-grade glioma in the left hemisphere is linked to inter-hemispheric reorganization. Based on literature, we hypothesized that reorganization would induce changes in the spatial pattern of activation specifically in tumour homologue brain areas in the healthy right hemisphere. An experimental group (EG) of 14 patients with a glioma in the left hemisphere near language related brain areas, and a control group of 6 patients with a glioma in the right, non-language dominant hemisphere were scanned before and after resection. In addition, an age and gender matched second control group of 18 healthy volunteers was scanned twice. A verb generation task was used to map language related areas and a novel technique was used for data analysis. Contrary to our hypothesis, we found that functional recovery following surgery of low-grade gliomas cannot be linked to functional reorganization in language homologue brain areas in the healthy, right hemisphere. Although elevated changes in the activation pattern were found in patients after surgery, these were largest in brain areas in proximity to the surgical resection, and were very similar to the spatial pattern of the brain shift following surgery. This suggests that the apparent perilesional functional reorganization is mostly caused by the brain shift as a consequence of surgery. Perilesional functional reorganization can however not be excluded. The study suggests that language recovery after transient post-surgical language deficits involves recovery of functioning of the presurgical language system.

Rehabilitation of verbal memory by means of preserved nonverbal memory abilities after epilepsy surgery

We present a patient with epilepsy who underwent left anterior temporal cortex resection, sparing the hippocampus, to stop drug-refractory seizures. Given that one year after surgery the patient showed verbal memory difficulties, we proposed a short (twelve weeks) and intensive (two times a week) training based on visual imagery strategies as the nonverbal memory abilities were preserved. Neuropsychological and fMRI assessments were performed before and after rehabilitation to evaluate the cognitive progress and cerebral modifications induced by this rehabilitation program. Our results showed that the rehabilitation program improved both scores for verbal memory and the everyday quality of life. Changes in cerebral activity highlighted by fMRI suggest that the program might have facilitated the development of compensatory strategies, as reflected by the shift of activation from the anterior to the posterior cerebral network during a verbal memory task. One year after the rehabilitation program, the patient reported using mental imagery in everyday life for routine and professional activities. Although supplementary evidence is necessary to increase the robustness of these findings, this case report suggests that an efficient rehabilitation program is feasible and (a) should be based on the individual cognitive profile and on the preserved cognitive abilities, (b) can be short but intensive, (c) can be applied even months after the lesion occurrence, and (d) can induce a positive effect which may be sustainable over time.

Functional language shift to the right hemisphere in patients with language-eloquent brain tumors

Objectives

Language function is mainly located within the left hemisphere of the brain, especially in right-handed subjects. However, functional MRI (fMRI) has demonstrated changes of language organization in patients with left-sided perisylvian lesions to the right hemisphere. Because intracerebral lesions can impair fMRI, this study was designed to investigate human language plasticity with a virtual lesion model using repetitive navigated transcranial magnetic stimulation (rTMS).

Experimental design

Fifteen patients with lesions of left-sided language-eloquent brain areas and 50 healthy and purely right-handed participants underwent bilateral rTMS language mapping via an object-naming task. All patients were proven to have left-sided language function during awake surgery. The rTMS-induced language errors were categorized into 6 different error types. The error ratio (induced errors/number of stimulations) was determined for each brain region on both hemispheres. A hemispheric dominance ratio was then defined for each region as the quotient of the error ratio (left/right) of the corresponding area of both hemispheres (ratio >1  =  left dominant; ratio <1  =  right dominant).

Results

Patients with language-eloquent lesions showed a statistically significantly lower ratio than healthy participants concerning “all errors” and “all errors without hesitations”, which indicates a higher participation of the right hemisphere in language function. Yet, there was no cortical region with pronounced difference in language dominance compared to the whole hemisphere.

Conclusions

This is the first study that shows by means of an anatomically accurate virtual lesion model that a shift of language function to the non-dominant hemisphere can occur.