Elsevier

World Neurosurgery

Volume 97, January 2017, Pages 123-131
World Neurosurgery

Original Article
Clinical Impact and Implication of Real-Time Oscillation Analysis for Language Mapping

https://doi.org/10.1016/j.wneu.2016.09.071Get rights and content

Background

We developed a functional brain analysis system that enabled us to perform real-time task-related electrocorticography (ECoG) and evaluated its potential in clinical practice. We hypothesized that high gamma activity (HGA) mapping would provide better spatial and temporal resolution with high signal-to-noise ratios.

Methods

Seven awake craniotomy patients were evaluated. ECoG was recorded during language tasks using subdural grids, and HGA (60–170 Hz) maps were obtained in real time. The patients also underwent electrocortical stimulation (ECS) mapping to validate the suspected functional locations on HGA mapping. The results were compared and calculated to assess the sensitivity and specificity of HGA mapping. For reference, bedside HGA–ECS mapping was performed in 5 epilepsy patients.

Results

HGA mapping demonstrated functional brain areas in real time and was comparable with ECS mapping. Sensitivity and specificity for the language area were 90.1% ± 11.2% and 90.0% ± 4.2%, respectively. Most HGA-positive areas were consistent with ECS-positive regions in both groups, and there were no statistical between-group differences.

Conclusions

Although this study included a small number of subjects, it showed real-time HGA mapping with the same setting and tasks under different conditions. This study demonstrates the clinical feasibility of real-time HGA mapping. Real-time HGA mapping enabled simple and rapid detection of language functional areas in awake craniotomy. The mapping results were highly accurate, although the mapping environment was noisy. Further studies of HGA mapping may provide the potential to elaborate complex brain functions and networks.

Introduction

When brain lesions are located in eloquent functional areas, particularly those close to language-related structures, it is necessary to perform detailed functional mapping to preserve brain functions while ensuring maximal lesion resection.1, 2, 3

The combined use of different electrophysiologic monitoring modalities provides reliable mapping to support surgical decision making. One of the clinical gold standards is electrocortical stimulation (ECS). Wilder Penfield revealed that stimulating the precentral gyrus elicited responses contralaterally, which was first correlated with brain anatomy in 1937.4 ECS to language areas for inhibition of various language capabilities is more complicated than motor mapping, making language mapping more complex. Technical difficulties remain, such as evaluating induced symptoms, appropriate tasks, and patient cooperation, as well as identification of optimal stimulation sites. In addition to technical difficulties, Pouratian et al.5 noted conditional variability of ECS mapping, along with seizure risks within individuals and across institutes in their review. However, ECS is still a reliable and indispensable procedure all over the world.

Previous reports, including ours, described a higher sensitivity but lower specificity (55%–80%) of language functional magnetic resonance imaging (fMRI) than those of ECS mapping.6, 7, 8 Accordingly, fMRI is not yet sufficient to localize the associated brain functions.

Power changes of oscillatory neuronal activities over various frequency ranges have recently received strong attention as physiologic correlates of blood oxygenation level−dependent responses.9, 10, 11 Among these oscillatory changes, high gamma activity (HGA) augmentation ranging from approximately 60–140 Hz is assumed to reflect localized cortical processing.12, 13 Sinai et al.13 reported a detailed comparison between ECS and HGA analysis in patients with subdural grids implantation at the bedside. Their HGA recording procedure was off-line and showed HGA dynamics over tasks. The researchers concluded that off-line HGA analysis could not supplant stimulation mapping because they did not refer to the temporal changes of HGA in detail.

We consider development of real-time HGA analysis not only at the bedside, but also in the operation room, to be important. As a result, we recently reported a user interface for “real-time” HGA mapping in 4 cases with awake craniotomy by combining various language tasks.14 The real-time mapping technique enables us to understand the underlying electrophysiologic temporal dynamics in the left hemisphere.15 In this report, we would like to stress the clinical impact of real-time HGA mapping in patients with awake craniotomy and compare the process with bedside mapping.

Section snippets

Patients

On a practical level, it is difficult to enroll many patients who could perform tasks perfectly in awake craniotomy. Seven patients undergoing awake craniotomy participated in this study. The patients had brain tumor lesions harboring language-related cortices in the dominant hemisphere. To preserve language functions, we recorded electrocorticography (ECoG) and applied ECS via subdural grids during awake craniotomy. For reference, we previously performed bedside language mapping in 5 epilepsy

Results

During awake craniotomy, all patients were sufficiently awake to follow the tasks. HGA mapping was successfully performed in all 7 patients. After HGA mapping, we obtained consistent validation using ECS in all 7 patients (Figure 2).

The sensitivity and specificity of picture naming for awake craniotomy were 90.1% ± 11.2% and 90.1% ± 4.2%, respectively. On the other hand, those of word reading were 88.5% ± 11.3% and 89.3% ± 4.1% respectively. Because of the higher sensitivity of the picture

Discussion

We have applied real-time HGA mapping to 7 brain tumor cases during awake craniotomy and 5 epilepsy cases with subdural grids on the left hemisphere at the bedside. Real-time HGA mapping enabled rapid identification of language-related areas. This procedure has a number of advantages. First, it is a nonstimulation mapping technique, which reduces the risk of seizure. Second, it enables significant shortening of the time required for cortical mapping. Third, it detects functional representations

Conclusion

This study demonstrated the feasibility of real-time HGA mapping in clinical use. HGA mapping allows us to rapidly and precisely map areas without the risk of evoking seizure. In particular, HGA mapping is a powerful mapping procedure in patients undergoing awake craniotomy because of its low invasiveness and rapid mapping. Real-time HGA mapping has the potential to be an alternative to ECS mapping. In this study, although we measured mainly language functions, HGA mapping can also possibly be

Acknowledgments

The authors wish to thank the staff of Leica in Kyoto City, Japan for their technical support.

References (22)

  • N. Kunii et al.

    A detailed analysis of functional magnetic resonance imaging in the frontal language area: a comparative study with extraoperative electrocortical stimulation

    Neurosurgery

    (2011)
  • Cited by (16)

    • A distributed network supports spatiotemporal cerebral dynamics of visual naming

      2021, Clinical Neurophysiology
      Citation Excerpt :

      However, the cerebral network dynamics supporting visual naming remains poorly characterized. High-gamma modulation (HGM) during visual naming has been shown to be a specific and accurate biomarker for speech/language sites ascertained by electrical stimulation mapping (ESM) with both subdural electrodes and stereo-electroencephalography (SEEG) (Arya et al., 2017, Babajani-Feremi et al., 2016, Ervin et al., 2020, Ogawa et al., 2017, Sinai et al., 2005), and has been harnessed to study spatiotemporal dynamics of cortical activation to define functional networks. Different components of cortical language networks were activated with different temporal envelopes, having cascading spatial–temporal patterns, in a study including 7 patients (Wang et al., 2016).

    • High-gamma modulation language mapping with stereo-EEG: A novel analytic approach and diagnostic validation

      2020, Clinical Neurophysiology
      Citation Excerpt :

      With subdural grids, electrical cortical stimulation mapping (ESM) is considered the gold-standard for such functional brain mapping (Jayakar et al., 2014). However, neurophysiologic and patient-safety concerns associated with ESM fostered development of an alternative methodology for functional mapping based on task-related power changes in EEG spectra, which has shown good accuracy for classification of ESM speech/language sites in multiple studies (Arya et al., 2018a, Arya et al., 2017, Ogawa et al., 2017, Sinai et al., 2005, Towle et al., 2008, Wang et al., 2016, Zea Vera et al., 2017). Nearly all the work on this high-gamma modulation (HGM) language mapping has been performed with subdural electrodes, with a paucity of studies on this modality with SEEG.

    • Spatiotemporal dynamics of auditory and picture naming-related high-gamma modulations: A study of Japanese-speaking patients

      2019, Clinical Neurophysiology
      Citation Excerpt :

      Picture naming is the task most commonly employed in high-gamma language mapping, partly because of its simplicity of the task design (Arya et al., 2018). Based on previous iEEG studies, the spatiotemporal dynamics of high-gamma augmentation during picture naming are qualitatively similar across patients speaking distinct native languages, including Japanese (Tanji et al., 2005; Kunii et al., 2013; Ogawa et al., 2017), Chinese (Lin et al., 2015; Wen et al., 2017), English (Sinai et al., 2005; Edwards et al., 2010; Cervenka et al., 2013; Arya et al., 2017; Babajani-Feremi et al., 2018; Forseth et al., 2018; Nakai et al., 2019), French (Lachaux et al., 2007; Llorens et al., 2011), and Dutch (Bauer et al., 2013). Picture naming-related high-gamma augmentation was reported to take place commonly in the bilateral occipital and fusiform regions immediately following stimulus presentation, subsequently involve left inferior frontal gyrus (IFG), and finally involve the bilateral inferior pre- (iPreCG) and postcentral gyri (iPostCG) during overt responses (Forseth et al., 2018; Nakai et al., 2019).

    • ECoG high-gamma modulation versus electrical stimulation for presurgical language mapping

      2018, Epilepsy and Behavior
      Citation Excerpt :

      All analyses were performed using the “MADA” library in R [14]. Fifteen studies were included, having 1 to 17 patients, with mean age varying from 10.3 to 53.6 years (Table 1) [15–29]. Six of the studies included native speakers of languages other than English.

    View all citing articles on Scopus

    Conflict of interest statement: This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. It was supported in part by a Grant-in-Aid for Young Scientists (B) No. 26870024 from 2014–2015, a Grant-in-Aid for Scientific Research (B) No. 24390337 from 2012–2015, a Grant-in-Aid for Exploratory Research No. 26670633 from 2014–2016 from the Ministry of Education, Culture, Sports, Science and Technology, a Grant-in-Aid for Scientific Research on Innovative Areas, No. 26870024, the European Union FP7 Integrated Project VERE No. 257695, and the EU project “High Profile” No. 269356. We are grateful for this support.

    View full text