Prolonged Transcortical Mixed Aphasia Following Awake Glioma Resection: Functional Language Network Disruption Associated with Persistent Epileptiform Activity

Abstract

Aphasia occurring after tumor resection in the language-dominant hemisphere is commonly interpreted as a consequence of irreversible structural injury. In awake craniotomy, intraoperative language mapping enables real-time functional monitoring and is expected to reduce the risk of postoperative aphasia. However, postoperative language impairment is not always attributable to structural damage alone, and functional network disturbances may also contribute.

We report a case of prolonged aphasia following awake glioma resection in the dominant hemisphere, in which epileptiform activity was suspected to underlie a prolonged functional disruption of the language network. Intraoperative language mapping confirmed preserved language function throughout tumor resection. However, during the late intraoperative phase, stimulation-induced after-discharges were observed on electroencephalography, followed by the emergence of unstable language output. Postoperatively, severe aphasia persisted despite the absence of new ischemic or hemorrhagic lesions on neuroimaging. Serial electroencephalography demonstrated sustained epileptiform activity within the interictal–ictal continuum, raising concern for possible aphasic status epilepticus.

Detailed language assessment revealed severe impairment of spontaneous speech and auditory comprehension, with relative preservation of repetition, reading aloud, and writing. This pattern was consistent with transcortical mixed aphasia, classically described as isolation of the speech area, suggesting functional disconnection of the language network despite anatomical preservation of core language cortices. Although electroencephalographic abnormalities gradually attenuated with antiseizure treatment, language impairment remained prolonged.

Prolonged aphasia after awake glioma resection may reflect potentially reversible language network disconnection associated with persistent epileptiform activity rather than structural injury alone, supported by SLTA-confirmed recovery at 6 months.

Introduction

Aphasia following brain tumor resection is often interpreted as an irreversible consequence of direct surgical injury, particularly when lesions are located near language-related cortices in the dominant hemisphere. Awake craniotomy with intraoperative language mapping is an established technique that allows real-time functional assessment during tumor resection and is widely employed to minimize the risk of postoperative aphasia.¹⁾

However, not all aphasic symptoms emerging after awake surgery are attributable to structural cortical damage. Nonconvulsive status epilepticus (NCSE) is clinically underrecognized because it lacks overt convulsive activity, yet it represents a major clinical subcategory of status epilepticus.²⁾ Its clinical presentation is heterogeneous. Among these, aphasic status epilepticus (ASE), in which aphasia constitutes the sole clinical manifestation, is relatively rare. Although its precise incidence remains unknown, ASE is considered to represent only a small subset of NCSE. In the postoperative setting, ASE may be overlooked or mistaken for surgically induced aphasia, particularly when obscured by impaired consciousness or changes in general condition.³⁾

ASE is generally regarded as a reversible condition; however, prolonged language impairment may occur despite timely recognition and treatment, particularly in complex postoperative settings where ongoing epileptic activity can be difficult to fully control. In some patients, sustained language disturbance manifests as a clinical profile classically described as isolation of the speech area, corresponding to transcortical mixed aphasia.⁶⁾ In this condition, Broca's and Wernicke's areas are anatomically preserved, yet functional connectivity with surrounding association cortices and language-related white matter pathways is disrupted. This syndrome is therefore better understood not as a simple focal lesion but as a manifestation of language network disconnection. Recent reports suggest that ictal or peri-ictal activity in NCSE or ASE can extend beyond focal pathology, leading to widespread network dysfunction and potentially resulting in reversible or prolonged higher-order cognitive deficits.³⁾ However, prolonged clinical presentation of an isolation-of-the-speech-area-like syndrome following awake glioma resection remains uncommon in the neurosurgical literature.

Here, we report a case of prolonged transcortical mixed aphasia following awake glioma resection in the language-dominant hemisphere, in which NCSE was considered a possible underlying mechanism. Intraoperative language monitoring confirmed preserved language function during tumor resection, whereas aphasic symptoms emerged and persisted in close temporal association with electroencephalographic findings suggestive of epileptic activity from the late intraoperative phase through the postoperative period. Based on the clinical course and electroencephalographic findings, the patient's aphasia was considered unlikely to result from direct surgical injury and was instead strongly suggestive of a functional isolation of the language network secondary to epileptic activity.

Case Report

The patient was a right-handed woman in her early 60s. Written informed consent was obtained from the patient for publication of this case report and any accompanying images. Language dominance had been confirmed to be left hemispheric by a preoperative Wada test. She had previously undergone craniotomy for resection of a left frontal glioma (glioblastoma, IDH-wild type, WHO Grade 4) four years prior to the current presentation, followed by concomitant temozolomide-based chemoradiotherapy and completion of 12 cycles of adjuvant temozolomide therapy. After the initial treatment, she exhibited no apparent neurological deficits. Instrumental Activities of Daily Living (IADL) were fully independent, with a score of 8/8. She maintained a high level of social participation, including attending multiple dance classes as a hobby and traveling abroad with her spouse.

In October of year X, follow-up magnetic resonance imaging revealed the appearance of a nodular contrast-enhancing lesion posterior to the previous resection cavity (Figure 1a, 1b). Additional imaging studies were also suggestive of tumor recurrence. Tractography showed that the recurrent lesion was located in close proximity to the arcuate fasciculus (AF), frontal aslant tract (FAT) (Figure 1c), and cingulum bundle. Given the anatomical relationship to language-related white matter pathways, awake craniotomy with intraoperative language mapping was planned in order to preserve language function.

Figure 1

Operative and imaging findings.

(a, b) Preoperative gadolinium-enhanced T1-weighted MRI showing the lesion in axial (a) and coronal (b) views. (c) Three-dimensional tractography overlaid on the preoperative MRI. The frontal aslant tract (FAT) is shown in light blue, and the arcuate fasciculus is shown in orange. (d) Intraoperative findings illustrated using a three-dimensional volume-rendered model based on preoperative imaging data. White circles with black borders indicate subdural electrodes placed on the cortical surface for continuous electrocorticographic monitoring. The numbers correspond to the electrode contacts shown in panel (f). Lightning symbols indicate the site of bipolar subcortical stimulation shown in panel (e). Subcortical stimulation at 8 mA at this site induced language disturbance characterized by unstable language output, followed immediately by the emergence of evolving after-discharges, as demonstrated in panel (f). (g, h) Postoperative gadolinium-enhanced T1-weighted MRI demonstrating gross total resection in axial (g) and coronal (h) views.

Preoperative neuropsychological assessment

Preoperative language assessment using the Standard Language Test of Aphasia (SLTA), a comprehensive and standardized aphasia battery widely used in Japan, yielded full scores, with no evidence of aphasia. Communicative behavior in practical conversation, including pragmatics, social appropriateness, and politeness, was well preserved. Intellectual functioning, as assessed by the Wechsler Adult Intelligence Scale–Fourth Edition (WAIS-IV), showed a Full-Scale Intelligence Quotient (FSIQ) of 103 (Verbal Comprehension Index [VCI] 98, Perceptual Reasoning Index [PRI] 109, Working Memory Index [WMI] 100, and Processing Speed Index [PSI] 96), which fell within the average range. Memory function evaluated by the Wechsler Memory Scale–Revised (WMS-R) revealed an overall memory index of 104 (verbal memory 106, visual memory 99, attention/concentration 107, and delayed recall 97), also within the average range. No significant discrepancy was observed between verbal and performance-based indices. Furthermore, no impairment was detected in frontal lobe–related functions, including executive control and inhibitory processes (Table 1).

Table 1

Neuropsychological and language assessments before and after surgery

Domain	Assessment	Subtest/Measure	Preoperative	POD 50	POM 6
SLTA, Standard Language Test of Aphasia (Japanese version); WAIS-IV, Wechsler Adult Intelligence Scale–Fourth Edition; FSIQ, Full Scale Intelligence Quotient; VCI, Verbal Comprehension Index; PRI, Perceptual Reasoning Index; WMI, Working Memory Index; PSI, Processing Speed Index; FAB, Frontal Assessment Battery; CAS, Cognitive Assessment System (questionnaire-based); WCST, Wisconsin Card Sorting Test; CA, Categories Achieved; PEN, Perseverative Errors; DMS, Difficulty Maintaining Set; WFT, Word Fluency Test; TMT-J, Trail Making Test–Japanese version; BADS, Behavioural Assessment of the Dysexecutive Syndrome; WMS-R, Wechsler Memory Scale–Revised POD, Postoperative Day POM, Postoperative Month TCMA, Transcortical Mixed Aphasia Postoperative scores should be interpreted with caution, as executive dysfunction may have affected task performance. At 6 months postoperatively, only a partial neuropsychological assessment was performed due to time constraints.
Language function	SLTA		No aphasia	TCMA	No aphasia
Intellectual function	WAIS-IV	FSIQ	103	-	-
		VCI	98	-	-
		PRI	109	<50	87
		WMI	100	-	-
		PSI	96	50	71
Frontal lobe function	FAB		17	-	-
	CAS	Questionnaire method	23	-	-
	WCST
	First trial	CA	2	-	-
		PEN (PEN%)	25 (73.5)	-	-
		DMS	0	-	-
	Second trial	CA	5	-	-
		PEN (PEN%)	3 (23)	-	-
		DMS	0	-	-
	WFT	Animal naming	19 words	-	-
		Phonemic fluency (“ka”)	13 words	-	-
	Stroop test	D	12″41	-	-
		W	12″80	-	-
		C	12″34	-	-
		AC	0	-	-
		E	0	-	-
	Digit span	Forward	6 digits	-	-
		Backward	4 digits	-	-
	TMT-J	(A)	Borderline	-	-
		(B)	Borderline	-	-
	BADS	Overall classification	Average	-	-
Memory function	WMS-R	Verbal memory	106	-	-
		Visual memory	99	<50	79
		General memory	104	-	-
		Attention/concentration	107	-	-
		Delayed recall	97	-	-

Assessment during awake surgery

Intraoperative cortical and subcortical mapping was performed under awake conditions to assess language and higher cognitive functions. The patient remained alert and cooperative throughout mapping and tumor resection. Task selection was determined based on lesion anatomy: the lateral aspect of the nodular lesion was located in the lateral frontal region, adjacent to the frontal aslant tract (FAT) and arcuate fasciculus, while the medial aspect was close to the cingulate gyrus and corpus callosum. Accordingly, frontal executive function tasks were administered in addition to language assessments.

Language tasks, including spontaneous speech, visual naming, auditory naming, and repetition, were intermittently evaluated during mapping and resection, and no language impairment was observed during the initial phase of tumor removal. Executive function tasks, including the 2-back task and the Stroop test, were also performed. During the Stroop test (four-color Kanji version), transient task-related instability was observed during bipolar stimulation at an intensity of 8 mA around the lateral inferior frontal gyrus. Subdural electrode contacts No. 1-4 were placed over the same region for continuous electrocorticographic monitoring (Figure 1d).

Subcortical electrical stimulation was performed in accordance with the guidelines of the Japan Awake Surgery Conference, beginning at 4 mA and incrementally increased. No language disturbance was observed at 4 mA or 6 mA. At 8 mA, unstable language output was elicited (Figure 1e), which was interpreted as the functional threshold. Tumor resection then started and subsequently proceeded toward the lateral margin of the enhancing mass, a region presumed to be adjacent to the FAT. Immediately thereafter, after-discharges were detected on intraoperative electroencephalography (Figure 1f).

Cold saline was applied to the cortical surface, and the surgical team paused until the after-discharges subsided. Although the electroencephalographic after-discharges resolved, language disturbance persisted, predominantly consisting of perseverative and semantic paraphasic errors. No generalized convulsions or overt focal motor seizures were observed. During language testing, correct responses were intermingled with paraphasic errors. Because such fluctuations in performance are commonly encountered during awake craniotomy, particularly in the context of intraoperative fatigue, fatigue-related effects were also considered. Nevertheless, given the temporal sequence—stimulation followed by after-discharges and subsequent persistence of language disturbance—it was considered likely that epileptiform activity contributed to the observed language dysfunction. At this point, the resection was judged to have reached its functional limit, and general anesthesia was reintroduced.

Postoperative course

Immediately after surgery, the patient exhibited impaired consciousness corresponding to a Glasgow Coma Scale (GCS) score of E3V3M4, which persisted until postoperative day (POD) 8. During this period, detailed language assessment was not feasible. Regarding antiseizure medications (ASMs), the patient had been receiving perampanel (PER) at a dose of 2 mg/day preoperatively. However, given the occurrence of aphasic symptoms associated with intraoperative after-discharges and the presence of impaired consciousness immediately after surgery, ASMs were intensified on the day of surgery to intravenous levetiracetam (LEV) 1000 mg/day in combination with PER 2 mg/day.

Postoperative MRI performed on POD 1 confirmed gross total resection of the lesion, with no abnormal findings in the resection cavity (Figure 1g, h). Because impaired consciousness persisted thereafter, the first postoperative electroencephalographic (EEG) examination was conducted on POD 2. The EEG demonstrated sustained focal epileptiform discharges characterized by persistent sharp waves with associated slow waves over the left frontocentral region (Figure 2a). These abnormalities were interpreted as reflecting sustained focal epileptiform activity, possibly situated within the interictal–ictal continuum (IIC) spectrum.⁴⁾ In light of the intraoperative stimulation-induced after-discharges observed over the same region, and the temporal association between the postoperative electroencephalographic abnormalities and the clinical symptoms, possible nonconvulsive status epilepticus was suspected.

Figure 2

Postoperative electroencephalographic findings.

(a) Electroencephalography (EEG) on postoperative day (POD) 2. The background rhythm is poorly sustained, with persistent sharp waves accompanied by slow waves predominantly over the left frontocentral region (F3–C3), consistent with sustained focal epileptiform activity.

(b) EEG on POD 22. The previously sustained sharp waves over the left frontal region are markedly attenuated, and overall background continuity is restored.

Accordingly, ASM therapy was further escalated on POD 2 to LEV 2000 mg/day, PER 4 mg/day, and lacosamide (LCM) 100 mg/day. Because the clinical condition remained unchanged, fosphenytoin was additionally administered at a dose of 22.5 mg/kg on POD 3. Following these therapeutic interventions, EEG recorded on POD 9 demonstrated partial reduction and fragmentation of the previously sustained left frontal epileptiform activity. At the same time, the level of consciousness improved to a GCS score of E4V3M6; however, severe language disturbance persisted.

Ultimately, ASM therapy was intensified to LEV 2000 mg/day, PER 4 mg/day, LCM 200 mg/day, and brivaracetam (BRV) 50 mg/day. From POD 22 onward, EEG findings showed marked attenuation of focal sharp waves over the left frontal region, along with restoration of background continuity (Figure 2b). These findings suggested a prolonged state of sustained epileptiform activity that gradually improved with intensive antiseizure therapy.

Postoperative language assessment

By postoperative day (POD) 9, the level of consciousness had improved to a GCS score of E4V3M6; however, the patient remained drowsy with a persistent tendency to stay in bed, and the level of arousal was still unstable. Although vocal intensity was markedly reduced, verbal output gradually became detectable at the level of single words and simple phrases. Eye contact and visual tracking were preserved, and the patient was able to fixate on medical staff and respond with a smile, suggesting intact interpersonal recognition.

Bedside language screening demonstrated correct responses in 3 of 5 items on high-frequency word naming items. Reading aloud of both kana (Japanese phonetic script) and kanji (Japanese logographic script) words was possible, as was repetition at the word and sentence levels. In contrast, echolalia, completion behavior, and perseverative paraphasic errors were prominent, and the language disturbance persisted over time.

A comprehensive language assessment was performed on POD 50. Spontaneous speech production and auditory comprehension were severely impaired, whereas repetition, reading aloud, and writing—modalities that do not primarily rely on semantic processing—were relatively preserved. This pattern was consistent with transcortical mixed aphasia (Figure 3). No apparent motor deficits were observed.

Figure 3

SLTA performance profile before and after surgery.

Line graph showing percentage correct scores for each subtest of the Standard Language Test of Aphasia (SLTA) at four time points: preoperative (filled circle, solid line), postoperative day 50 (open square, dashed line), 3-month postoperative (filled triangle, dotted line), and 6-month postoperative (filled diamond, dash-dot line). Postoperatively, marked declines are observed across most language domains at day 50. Partial recovery is evident at 3 months, and improvement to a non-aphasic level is confirmed at 6 months postoperatively.

At 3 months postoperatively, the SLTA was administered at the rehabilitation hospital and revealed persistent deficits across multiple language domains, indicating that recovery was still in progress. At 6 months postoperatively, repeat SLTA at our institution confirmed improvement to a non-aphasic level (Table 1, Figure 3). This clinical trajectory is consistent with a functionally reversible disruption of the language network secondary to prolonged epileptiform activity, rather than irreversible anatomical disconnection.

Serial neuroimaging studies obtained during the clinical course did not reveal any new ischemic or hemorrhagic lesions that could account for the severity or persistence of language impairment.

Discussion

This case is novel in that prolonged language dysfunction consistent with isolation of the speech area (transcortical mixed aphasia) developed in the context of possible nonconvulsive status epilepticus (NCSE), despite preservation of language function during awake tumor resection in the language-dominant hemisphere.

In general, postoperative aphasia following tumor resection in the language-dominant hemisphere is often interpreted as an irreversible deficit resulting from direct cortical injury or disruption of language-related white matter pathways. However, awake craniotomy enables real-time assessment of language function through cortical and subcortical mapping, allowing immediate detection of language disturbances associated with surgical manipulation.¹⁾ When language function is preserved intraoperatively, postoperative aphasia is less likely to reflect irreversible surgical injury. In the present case, because language functions including spontaneous speech, auditory comprehension, and naming were preserved throughout tumor resection, irreversible damage to the language cortex caused by resection was unlikely from an intraoperative assessment perspective.

In contrast, the clear temporal association between the emergence of sustained epileptiform activity toward the end of the awake phase and the subsequent worsening of language symptoms, including increased paraphasic errors, suggests that an epileptic mechanism likely contributed to the observed language dysfunction. Postoperatively, the coexistence of impaired consciousness and aphasia with persistent electroencephalographic abnormalities, together with improvement of EEG findings in response to antiseizure medication, was compatible with prolonged interictal–ictal continuum activity with possible nonconvulsive status epilepticus.^4,5)

The aphasia observed in this patient was characterized by severe impairment of spontaneous speech and auditory comprehension, with relative preservation of repetition, reading aloud, and writing—modalities that do not primarily rely on semantic processing. This pattern is consistent with transcortical mixed aphasia, classically described as isolation of the speech area. In this condition, Broca's and Wernicke's areas are anatomically preserved, while functional connectivity with surrounding association cortices and frontal executive control networks is disrupted.⁶⁾

Furthermore, the presence of echolalia and completion behavior in this case suggests that the aphasia was not the result of focal structural damage to the language cortices themselves but rather reflected impaired functional connectivity between language regions and higher-order control networks. While the integration of semantic processing within the language network was disrupted, relatively automatic language pathways underlying repetition and reading remained preserved. This dissociation supports the notion that a state of isolation of the speech area emerged as a manifestation of language network disconnection.⁶⁾

Notably, although the resection did not involve the classical Broca's area, the lesion was located adjacent to the dorsal inferior frontal region, including areas recently implicated in motor aspects of speech initiation (e.g., area 55b)⁷⁾. This region was not explicitly targeted during preoperative planning; however, its anatomical proximity may have increased the vulnerability of the language network to sustained epileptic activity in the postoperative period.

Aphasic status epilepticus is generally regarded as a reversible condition.³⁾ However, electroencephalographic abnormalities in this case required approximately 22 days to attenuate despite intensive antiseizure treatment, and language impairment persisted thereafter. This clinical course suggests that prolonged epileptiform activity may exert sustained effects on the language network, even when antiseizure treatment is administered.³⁾

This report has several limitations. First, it describes a single case, and therefore the generalizability of the findings is limited. Second, although the clinical course and electroencephalographic findings strongly suggested a network-level mechanism, direct functional imaging or invasive electrophysiological data to confirm disruption of specific language networks were not available. Further studies are needed to clarify the mechanisms underlying prolonged aphasia associated with NCSE in the postoperative setting. Another limitation is that, although electroencephalographic monitoring was performed and antiseizure medications were administered, the optimal intensity and duration of treatment for aphasic status epilepticus remain uncertain.^3,8) It is possible that earlier or more aggressive antiseizure strategies might have altered the clinical course; however, no established treatment protocol currently exists for such cases.

Although preoperative tractography confirmed close proximity of the FAT to the recurrent tumor (Figure 1c), and partial FAT involvement cannot be entirely excluded based on postoperative MRI findings, the improvement to a non-aphasic level on SLTA at 6 months postoperatively argues against permanent FAT disconnection as the dominant pathomechanism. Complete FAT disruption is known to cause persistent deficits in speech initiation,⁹⁾ and the favorable outcome in our patient is more consistent with a functionally reversible disruption of the language network secondary to prolonged epileptiform activity.

The recurrent nature of the tumor in this case represents an important confounding factor. Prior aggressive resection and postoperative radiotherapy may have induced subclinical disruption of the language network, including possible partial injury to the inferior fronto-occipital fasciculus (IFOF), which mediates semantic processing. Such preexisting vulnerabilities may have lowered the threshold for postoperative language network failure in the setting of superimposed epileptiform activity. It should be noted that the semantic paraphasic errors observed during intraoperative language mapping occurred following after-discharges induced by subcortical stimulation at 8 mA, and are interpreted as stimulation-induced language disturbance rather than a manifestation of preexisting network reorganization. This case underscores the importance of careful preoperative tractography and functional mapping in patients undergoing recurrent glioma resection involving regions adjacent to major language fiber tracts.

A notable limitation of this report is the absence of postoperative diffusion tensor imaging tractography, which would have allowed direct assessment of the structural integrity of the frontal aslant tract and arcuate fasciculus. However, the improvement to a non-aphasic level on SLTA at 6 months postoperatively suggests that permanent tract disconnection is unlikely to have been the dominant mechanism. Although partial white matter tract injury cannot be entirely excluded, the clinical course is more consistent with functional rather than anatomical disruption. Future cases should include postoperative tractography to better characterize the relative contributions of structural and functional mechanisms.

Several measures may help reduce the risk of similar complications in future cases. First, in patients with a history of prior surgery and radiotherapy, alterations in cortical and subcortical excitability and reorganization of the language network may have occurred; a more conservative stimulation protocol beginning at lower intensities, along with preoperative assessment of functional network integrity using resting-state fMRI, is therefore advisable. Second, if language disturbance persists following after-discharge detection during awake surgery, this should be recognized not merely as fatigue-related variability but as a potential sign of postictal functional impairment, with pre-defined criteria for suspending resection or transitioning to general anesthesia. Third, in cases where intraoperative after-discharges are observed, continuous postoperative EEG monitoring should be initiated promptly to detect early transition to nonconvulsive status epilepticus (NCSE). In the present case, NCSE was suspected based on intraoperative findings and antiseizure medication was intensified immediately postoperatively; prophylactic escalation at the time of intraoperative after-discharge detection may also warrant consideration. Finally, in recurrent glioma cases with tumors in close proximity to major language fiber tracts, thorough preoperative informed consent regarding the risk of prolonged postoperative aphasia, including that of NCSE origin, is recommended.

Conclusion

Prolonged aphasia following awake glioma resection may arise not only from direct surgical injury but also from sustained epileptiform activity leading to functional isolation of the language network. Careful postoperative electroencephalographic evaluation and timely therapeutic intervention should therefore be considered in patients with unexplained or prolonged language dysfunction.

Disclaimer

Author Koji Yoshimoto is one of the Editorial Board members of the Journal.

This author was not involved in the peer-review or decision-making process for this paper.

Acknowledgments

The authors thank Kazuhiro Samura, MD, PhD, Department of Neurosurgery, Saiseikai Karatsu Hospital, for his clinical care of the patient and for providing the SLTA data obtained at 3 months postoperatively.

Conflict of Interest Disclosure

All authors have no conflict of interest.

References

• 1)  De Witt Hamer PC, Gil Robles S, Zwinderman AH, et al. Impact of intraoperative stimulation brain mapping on glioma surgery outcome: a meta-analysis. J Clin Oncol. 2012;30 (20):2559-65. doi: 10.1200/JCO.2011.38.4818
• 2)  Trinka E, Cock H, Hesdorffer D, et al. A definition and classification of status epilepticus—Report of the ILAE Task Force on Classification of Status Epilepticus. Epilepsia. 2015;56 (10):1515-23. doi: 10.1111/epi.13121
• 3)  Kamogawa M, Ikegaya N, Miyake Y, et al. Verbal and memory deficits caused by aphasic status epilepticus after resection of a left temporal lobe glioma. Surg Neurol Int. 2021;12:614. doi: 10.25259/SNI_1120_2021
• 4)  Hirsch LJ, Fong MWK, Leitinger M, et al. American Clinical Neurophysiology Society's Standardized Critical Care EEG Terminology: 2021 Version. J Clin Neurophysiol. 2021;38 (1):1-29. doi: 10.1097/WNP.0000000000000806
• 5)  Leitinger M, Trinka E, Gardella E, et al. Diagnostic accuracy of the Salzburg EEG criteria for non-convulsive status epilepticus: a retrospective study. Lancet Neurol. 2016;15 (10):1054-62. doi: 10.1016/S1474-4422(16)30137-5
• 6)  Berthier ML, Starkstein SE, Leiguarda R, et al. Transcortical aphasia: importance of the nonspeech dominant hemisphere in language repetition. Brain. 1991;114 (3):1409-27. doi: 10.1093/brain/114.3.1409
• 7)  Hazem SR, Awan M, Lavrador JP, et al. Middle frontal gyrus and area 55b: perioperative mapping and language outcomes. Front Neurol. 2021;12:646075. doi: 10.3389/fneur.2021.646075
• 8)  Kubota Y, Nakamoto H, Kawamata T. Nonconvulsive status epilepticus in the neurosurgical setting. Neurol Med Chir (Tokyo). 2016;56 (10):626-31. doi: 10.2176/nmc.ra.2016-0118
• 9)  Kinoshita M, Menjot de Champfleur N, Deverdun J, et al. Role of fronto-striatal tract and frontal aslant tract in movement and speech: an axonal mapping study. Brain Struct Funct. 2015;220 (6):3399-3412. doi: 10.1007/s00429-014-0863-0