Secondary Logo

Journal Logo

SEIZURE AND DEVELOPMENTAL DISORDERS: Edited by Philippe Ryvlin

Insular seizures and epilepsies: Ictal semiology and minimal invasive surgery

Ryvlin, Philippea; Nguyen, Dang Khoab

Author Information
Current Opinion in Neurology: April 2021 - Volume 34 - Issue 2 - p 153-165
doi: 10.1097/WCO.0000000000000907
  • Free

Abstract

INTRODUCTION

It has been more than 20 years that the first patient with nonlesional epilepsy was diagnosed as suffering from insular ictal onset using stereo-electroencephalography (SEEG) [1], opening a new facet to the investigation and treatment of patients considered for epilepsy surgery. The recognition of this new entity, beyond the previously known cases of epileptogenic insular lesion detected on neuroimaging, participated to and benefited from the international development of SEEG [2], which in turn led to the identification of more cases of insular epilepsies.

Yet, all series published prior to 2017 included 10 patients or less, often single case reports, with a majority being biased toward a specific semiology such as sleep-related or nocturnal hypermotor seizures [3–7], painful seizures [8], reflex seizures [9–11], gelastic seizures [12], ictal bradycardia or AV block [13–15] and choking or gagging spells [16–19], ecstatic or orgasmic sensations [20,21] (see Table 1).

Table 1 - Characteristics of series of patients with insular or insulo-opercular seizures
Publication Number of patients % adults Pure insular ictal onset zone Normal MRI FCD (MRI or pathology) SEEG performed Insular surgery Engel class I outcome Primary findings and caveats
Series > 10 cases
 Wang et al.[26▪▪] 37 54% 22% 59% 73% 100% 100% 77% Hierarchical clustering delineated 4 anatomically-related clinical patterns. Includes 8 cases from Wang et al. 2019
 Singh et al.[24▪▪] 12 75% 75% 75% 50% 100% 58% 86% Hierarchical clustering delineated 2–3 anatomically-related clinical patterns
 Peltola et al.[25▪▪] 11 36% 100% 45% 45% 100% 64% 100% Distinction of anterior versus posterior insular seizures
 Wang et al.[23▪] 18a 55% 11% 67% 61% 100% 94% 88% Detailed semiology of individual patients. Caveat: 4 out of 22 patients with pure opercular ictal onset were excluded
 Freri et al.[22] 13a 0% 0 15% 54% 54% 100% 62% Detailed semiology of individual patients. Caveat: 3 out of 16 patients with concomitant mesial temporal ictal onset were excluded
 TOTAL series > 10 cases 91 47% 33% 55% 62% 93% 89% 80% Eight patients counted twice in Wang et al. 2019 and 2020
Series ≤ 10 cases
 Aldosari et al.[28] 1 100% 0 100% UNK 100% 100% 100% Eating-triggered seizures. Caveat: epileptogenic zone and surgery involved the mesial temporal region
 Kharytonov et al.[31] 2 0% UNK 0% 0 0 0 NA Asymetrical spasms with insular lesions. Caveat: No SEEG nor surgery, and lesions larger than the insula
 Wiwchar et al.[32] 1 0% 0 0% 100% 100% 100% 100% Ictal pouting. Caveat: FCD type II primarily fronto-opercular rather than insular
 Kuruumbi et al.[50] 1 100% 100% 100% 0 100% 100% 100% Caveat: surgery was fronto-temporo-insular
 Xiao et al.[11] 3 100% 33% 67% UNK 100% 100% 100% Reflex insular seizures (2 somatosensory, 1 auditory), all from the posterior insula (2 also involved the opercula)
 Gras-combes et al.[49] 6b 100% 86% 83% 83% 100% 100% 83% Detailed semiology of individual patients. All patients included in the larger series from Singh et al.[24▪▪]
 Thompson et al.[10] 1a 100% 0 100% 0 100% 100% 100% Reflex auditory aura with anterior insular and fronto-opercular seizure. Caveat: 1 pure fronto-opercular case excluded
 Montavont et al.[8] 4a 80% 25% 50% UNK 100% 0 NA Ictal pain arising from insula (mostly posterior) and SII. Caveat: 1 patient cured by resection of frontal FCD excluded
 Dylgjeri et al.[30] 10 0% 50% 20% 80% 100% 100% 70% Pediatric series of consecutive insular cases. Caveat: Lack semiologic details
 Geevasinga et al.[19] 2 100% 0 0% 50% 50% 50% 100% Chocking and asphyxiation during insular seizures. Only one of the 2 cases underwent SEEG and surgery
 Tran et al.[12] 4 100% 0 100% 50% 100% 100% 100% Gelastic seizures. Caveat: 1 case with an independent obitofrontal FCD
 Tayah et al.[15] 1 100% 100% 0% UNK 0 0 NA Ictal bradycardia associated with a small posterior insular lesion. Caveat: No SEEG nor surgery
 Nesbitt et al.[18] 1 100% UNK 100% UNK 0 0 NA Chocking seizures during sleep. Caveat: lack of SEEG or surgery. Insular origin based on semiology and FDG-PET only
 Kriegel et al.[48] 2 100% 100% 100% UNK 100% 0 NA
 Proserpio et al.[7] 8 UNK 0 50% 50% 100% 75% 75% Sleep-related insular seizures
 Landtblom et al.[21] 1 100% UNK 100% UNK 0 0 NA Ecstatic seizure. Caveat: Insular onset only suggested by SPECT data. No SEEG nor surgery performed
 Dioniso et al.[17] 1 100% 100% 0% 0 0 0 NA Gagging seizures misdiagnosed as gastro-oesophageal reflux with intrainsular cavernoma. Caveat: No SEEG nor surgery
 Blauwblomme et al.[9] 1b 100% 100% 0% 0 100% 100% 100% Eating triggered seizures from the middle short gyrus. The patient is included in the larger series from Singh et al.[24▪▪]
 Anzellotti et al.[20] 1 100% UNK 100% UNK 0 0 NA Ictal genital arousal and orgasm. Caveat: lack of SEEG or surgery. Insular origin based on MEG only
 Surges et al.[14] 1 100% UNK 0% UNK 0 0 NA Ictal AV block. Caveat: lack of ictal EEG, SEEG or surgery. Insular origin based on MRI lesion of uncertain significance
 Davis et al.[16] 1 100% UNK 100% 0 0 0 NA Chocking seizures during sleep. Caveat: lack of SEEG or surgery. Insular origin based on ictal SPECT only
 Zhang et al.[6] 1 0% 100% 0% UNK 100% 100% 100% Hypermotor seizures of insular origin associated with an orbitofrontal lesion
 Dobeserger et al.[5] 1 100% 100% 100% 0 100% 100% 100% Hypermotor seizures originating from the anterior insula
 Kaido et al.[3] 2 50% UNK 0% 50% 0 100% 100% Hypermotor seizures originating from the posterior ventral insula. Caveat: intracranial EEG without insular electrodes
 Ryvlin et al.[4] 3b 33% 67% 100% UNK 100% 0 NA Hypermotor seizures originating from the anterior insula. All patients included in the larger series from Singh et al.[24▪▪]
 Ryvlin [1] 2b 100% 50% 0% UNK 100% 0 NA First temporal-like and frontal-like insular seizure cases from Lyon. Included in Isnard et al. 2004 and Ryvlin et al. 2006
 Rossetti et al.[47] 1 100% 100% 0% 0 100% 100% 100% Ictal dysgueusia.
 Isnard et al.[33] 4a 100% 100% 75% UNK 100% 50% 100% First series of SEEG-defined insular seizures. Caveat: 2 out of 6 patients also had mesial temporal ictal onset and were excluded
 Seeck et al.[13] 1 0% UNK 0% 100% 0 0 NA Ictal bradycardia. Caveat: No SEEG nor surgery
 TOTAL series ≤ 10 cases 55 64% 44% 49% 45% 80% 65% 89% Likely bias toward selecting surgically-cured patients in small series
 TOTAL all series 146 52% 37% 53% 58% 88% 80% 86%
FCD, focal cortical dysplasia; NA, Not applicable; SEEG, stereo-electroencephalography; UNK, unknown.
aSome patients were excluded from these series due to either pure extra-insular seizure onset or involvement of the mesial temporal regions at ictal onset (see column “Primary findings and caveats” for details).
bAll patients from these series were also included in other larger series and are thus not counted in the totals (see column “primary findings and caveats” ).

In fact, all series collating more than 10 patients with insular epilepsy were published since 2017 [22,23▪,24–26▪▪], and all but one since 2019, offering an opportunity to update our knowledge on ictal insular semiology.

In parallel to progress in identifying insular epilepsies and their electroclinical correlations, minimal invasive surgical therapies have been tested in this type of epilepsy in particular Magnetic Resonance Imaging (MRI)-guided laser ablation and responsive neurostimulation (RNS). MRI-guided laser ablation is particularly well suited for deep located small-volume epileptogenic zones, such as the mesial temporal structures, periventricular nodular heterotopia, hypothalamic hamartoma, or the insula. Indeed, insular resection using standard open-skull surgery is associated with a significant risk of neurological deficit. Accordingly, a number of small series and case reports of MRI-guided laser ablation have accumulated over the last few years.

In this review, we will thus concentrate on two chapters devoted to ictal semiology and minimal invasive surgery in patients with insular seizures and epilepsies, 

FB1
Box 1:
no caption available

ICTAL SEMIOLOGY

In line with its integrative role in multiple brain functions and networks, including interoception, cognition, emotion and behavior, the insula has proved to be a highly eloquent cortex, giving rise to a large variety of subjective and objective ictal signs (see Table 2). The latter might reflect intra-insular, peri-insular (e.g., opercular), or more distant propagation of the ictal discharge, as well as involvement of cortico-subcortical network [24▪▪], thus masquerading seizure types encountered in temporal or frontal lobe epilepsies [1]. Insular epileptogenic onset zones also often encompass part of the nearby suprasylvian or infrasylvian opercula, making it difficult to distinguish the role of each of these structures in generating ictal signs, and leading to consider insulo-opercular rather than pure insular epilepsy in the majority of patients (see Table 1). Yet, pure insular epilepsies are also well documented [24▪▪,25▪▪], with most subjective signs reported in such patients being reproduced by direct electrical stimulation of the insula [24▪▪,27], thus suggesting their direct emergence from insular ictal discharges. Elementary orofacial motor signs are likely to reflect discharges from the suprasylvian opercula, whereas in contrast, other ictal motor signs observed in insular and insulo-opercular seizures are likely to reflect propagations to other cortical and subcortical brain regions, in particular the frontal premotor and central regions.

Table 2 - Subjective ictal signs in patients with insular or insulo-opercular seizures
Subjective signs during insular seizures
Publication Number of patients Nocturnal seizures Reflex seizures Aura Non painful somatosensory Laryngeal sensation Epigatric sensation Auditory sensation Respiratory sensation Emotions (fear, anxiety) Somatic pain Gustatory sensation Visual sensation Ecstatic, mirth
Series > 10 cases
 Wang et al.[26▪▪] 37 62% 11% 87% 32% 32% 11% 16% 5% 5%
 Singh et al.[24▪▪] 12 67% 25% 75% 25% 8% 8% 25% 17% 17% 8% 8%
 Peltola et al.[25▪▪] 11 36% 91% 73% 9% 18% 18% 9% 9% 9% 9%
 Wang et al.[23▪] 18a 77% 59% 5% 5% 5% 9% 5%
 Freri et al.[22] 13a 56% 31% 13% 6% 13% 6% 6%
 TOTAL series > 10 cases 91 38% 8% 79% 42% 16% 10% 10% 8% 7% 5% 5% 2% 1%
Series ≤ 10 cases
 Aldosari et al.[28] 1 100% 100% 100%
 Kharytonov et al.[31] 2 50%
 Wiwchar et al.[32] 1 100% 100%
 Kuruumbi et al.[50] 1 100% 100%
 Xiao et al.[11] 3 33% 100% 100% 33% 67%
 Thompson et al.[10] 1a 100% 100% 100% 100% 100%
 Montavont et al.[8] 4a 100% 100%
 Dylgjeri et al.[30] 10 100% 30%
 Geevasinga et al.[19] 2 50% 100% 100% 50%
 Tran et al.[12] 4 50% 25% 25%
 Tayah et al.[15] 1 100% 100% 100%
 Nesbitt et al.[18] 1 100% 100% 100% 100% 100% 100%
 Kriegel et al.[48] 2 100% 50% 50%
 Proserpio et al.[7] 8 100% 88% 50% 50% 13% 13% 13% 13% 25%
 Landtblom et al.[21] 1 100% 100% 100%
 Dioniso et al.[17] 1 100% 100% 100% 100%
 Anzellotti et al.[20] 1 100% 100% 100% 100%
 Surges et al.[14] 1 0% 100% 100% 100%
 Davis et al.[16] 1 100% 100% 100% 100%
 Zhang et al.[6] 1 100% 100% 100% 100%
 Dobeserger et al.[5] 1 100% 100% 100%
 Kaido et al.[3] 2 0%
 Rossetti et al.[47] 1 100% 100% 100%
 Isnard et al.[33] 4a 100% 100% 100% 25% 25%
 Seeck et al.[13] 1
 TOTAL series ≤ 10 cases 55 51% 11% 75% 31% 25% 5% 5% 5% 15% 15% 7% 0% 4%
 TOTAL all series 146 43% 9% 77% 38% 20% 8% 8% 7% 10% 9% 6% 1% 2%
Ictal signs were neither indicated as present or absent in a significant number of publications, in particuler case reports and small series. These are indicated in the table as ‘−’, and by default considered as absent in the calculation of their incidence in insular seizures.
aSome patients were excluded from these series due to either pure extra-insular seizure onset or involvement of the mesial temporal regions at ictal onset (see Table 1 for details).

Summary of the literature and brief description of the most recent largest series

As illustrated in Table 1, we retrieved 29 publications focusing on insular or insulo-opercular seizures, including a majority of single case reports, for a total of 146 patients (with a few counted twice in different publications) [23▪,26▪▪]. The five largest series accounted for 62% of cases. Although some differences are noted between large and small series (Table 1), the overall findings are consistent in showing that about half of patients had an insular MRI abnormality, with focal cortical dysplasia (FCD) being the most frequently identified lesion on MRI or at pathology (58%). SEEG was performed in 88% of reported cases, and surgery in 80%. An impressive 86% of Engel class I outcome was reported, showing that although challenging to operate, insular epilepsies can most often be controlled by surgery.

Synthesis of ictal signs

Regarding ictal signs detailed in Tables 1 and 2, some features were not reported as present or absent in many publications, in particular small series and case reports, with the likely possibility that such features were simply overlooked (e.g., nocturnal seizures, subtle orofacial signs, etc.), and are thus underestimated in this review. About 40% of patients have predominantly nocturnal seizures. A reflex component was observed in 9% of cases, triggered by sound (N = 6) [9–11,26▪▪], eating (N = 3) [9,26▪▪,28], or somatosensory stimuli (N = 3) [11,24▪▪]. An aura was reported in 77% of patients, with the most frequent being nonpainful somatosensory (38%) and laryngeal (20%) sensations. Epigastric, auditory, gustatory, emotional (fear, anxiety) and painful sensations were all reported between 5 and 10% of cases. Ecstatic, mirth or orgasmic sensations, visual blurring, dizziness were rare, occurring in 1–3 patients overall. Nonpainful somatosensory aura, typically described as paresthesias or tingling, and more rarely as thermal or electricity, were usually contralateral to seizure onset or bilateral, with very rare occurrence of purely ipsilateral sensation [25▪▪,29]. They often involved the face, but could affect any body parts, often including large cutaneous territories. Painful sensations could affect parts of a hemibody, at times reported as a burning sensation and associated with expression of pain and screaming [8], or be perceived over the abdominal region [25▪▪]. Laryngeal sensations were typically described as unpleasant and could culminate to throat constriction and suffocation.

Objective ictal signs, and in particular motor signs, were observed in virtually every patient with insular or insulo-opercular seizures (Table 3). The most frequent manifestations were elementary orofacial motor signs, reported in 60% of cases from the largest series, and most likely overlooked in smaller series (4%). Other elementary motor signs, dystonic posturing, hypermotor behavior, eye blinking and salivation all occurred in 20–30% of cases, including ipsilateral eye blinking in about 10%. Some authors have also stressed the value of tonic neck anteflexion [23▪]. Demonstration of lack of awareness was infrequent, only reported in 18%. Grimacing, facial flushing, head and/or eye deviation, automatic behavior (including oro-alimentary automatisms), speech disturbances, and focal to bilateral tonic-clonic seizures all occurred in 10–15% of patients. Rare ictal signs included spasms in children, usually asymmetrical [24▪▪,25▪▪,30,31], pouting [23▪,25▪▪,32], laughter [12,23▪,25▪▪], bradycardia or AV block [13–15,24▪▪], vomiting [25▪▪], and bilateral eyelid tonic closure that might wrongly point to psychogenic nonepileptic seizures [23▪].

Table 3 - Objective ictal signs in patients with insular or insulo–opercular seizures
Objectve ictal signs during insular seizures (% of patients presenting signs in each series)
Publication Number of patients Any motor sign Elementary motor sign Salivation Dystonic posturing Tachycardia Hypermotor behavior Eye blinking Lack of awareness Grimace Speech dysfunction Facial flushing Automatic behavior GTCS Head/eye deviation Spasms Ictal pouting Laugther Bradycardia Vomiting
Orofacial Other Any Ipsilateral Oroalim. Limb
Series > 10 cases
 Wang et al.[26▪▪] 37 97% 38% 32% 22% 22% 24% 22% 8% 3% 11% 16%
 Singh et al.[24▪▪] 12 100% 92% 50% 83% 92% 50% 50% 8% 92% 92% 50% 75% 33% 58% 8% 50% 17% 8%
 Peltola et al.[25▪▪] 11 100% 45% 46% 36% 27% 18% 27% 45% 45% 55% 55% 18% 18% 18% 9% 9% 9% 9%
 Wang et al.[23▪] 18a 100% 82% 59% 23% 23% 18% 18% 32% 9% 9% 5% 5% 9% 18% 9% 5%
 Freri et al.[22] 13a 100% 75% 50% 19% 19% 19%
 TOTAL series > 10 cases 91 99% 60% 31% 32% 30% 27% 24% 24% 10% 19% 14% 13% 12% 12% 9% 11% 10% 3% 3% 2% 1% 1%
Series ≤ 10 cases
 Aldosari et al.[28] 1 100% 100% 100% 0%
 Kharytonov et al.[31] 2 100% 100%
 Wiwchar et al.[32] 1 100% 100% 0% 100% 100%
 Kuruumbi et al.[50] 1 100% 100% 100% 100% 100% 100% 100% 100%
 Xiao et al.[11] 3 100% 67% 33% 33%
 Thompson et al.[10] 1a 100% 100% 100% 100% 100% 0% 100% 100% 100% 100%
 Montavont et al.[8] 4a 75% 20% 20% 20% 20% 25% 20%
 Dylgjeri et al.[30] 10 100% 50% 10% 50% 60% 60% 10% 40%
 Geevasinga et al.[19] 2 50%
 Tran et al.[12] 4 50% 25% 50% 25% 50% 100%
 Tayah et al.[15] 1 100%
 Nesbitt et al.[18] 1 0% 0%
 Kriegel et al.[48] 2 100% 50% 100% 50% 50% 50%
 Propserpio et al.[7] 8 100% 13% 13% 63% 75% 50% 25% 38% 13% 25%
 Landtblom et al.[21] 1 100%
 Dioniso et al.[17] 1 100%
 Anzellotti et al.[20] 1 100% 0%
 Surges et al.[14] 1 100% 100%
 Davis et al.[16] 1 100% 100%
 Zhang et al.[6] 1 100% 100% 100% 100% 100%
 Dobeserger et al.[5] 1 100% 100% 0%
 Kaido et al.[3] 2 100% 50% 100% 50% 50% 100%
 Rossetti et al.[47] 1 0%
 Isnard et al.[33] 4a 75% 50% 25% 25% 75% 25% 50%
 Seeck et al.[13] 1 100% 100% 100% 100% 100% 100%
 TOTAL series ≤ 10 cases 55 80% 4% 20% 13% 31% 2% 31% 20% 4% 18% 0% 16% 5% 15% 7% 20% 11% 11% 2% 7% 4% 0%
 TOTAL all series 146 92% 29% 27% 25% 30% 18% 27% 23% 6% 18% 9% 14% 10% 13% 8% 14% 10% 6% 3% 4% 2% 1%
Ictal signs were neither indicated as present or absent in a significant number of publications, in particuler case reports and small series. These are indicated in the table as ‘−’, and by default considered as absent in the calculation of their incidence in insular seizures.
aSome patients were excluded from these series due to either pure extra-insular seizure onset or involvement of the mesial temporal regions at ictal onset (see Table 1 for details).

Clinical patterns and anatomo-clinical correlations

Since the early identification of insular seizures, several patterns were identified, including a temporo-insular pattern (temporal plus) with a combination of insular and temporo-mesial ictal phenomenology including epigastric sensations and oro-alimentary automatisms, a fronto-insular pattern with hypermotor behavior, and insulo-opercular (perisylvian) patterns with preserved consciousness and prominent orofacial and laryngeal signs or auditory signs [1,4,33]. However, it was not until 2020 that sufficiently large series were collected to appropriately investigate anatomo-clinical correlations. The latter are summarized below.

Wang et al.[26▪▪] applied hierarchical clustering of symptoms in 37 patients with insular or insulo-opercular seizures, and delineated four seizure patterns:

  • (1) The most frequent pattern (46%) was primarily characterized by elementary orofacial and laryngeal signs, often associated with autonomic and other elementary focal motor signs. It involved the middle dorsal portions of the insula together with the frontal and central operculum.
  • (2) The second most frequent pattern (24%) included auditory signs and symmetric proximal/axial tonic posturing, but without other elementary motor or orofacial signs. It primarily involved the temporal operculum and adjacent ventral portion of the insula.
  • (3) The third pattern (19%) was characterized by somatosensory auras followed by nonintegrated gestures and/or asymmetric tonic posturing, without autonomic sign. Ictal onset occurred in the most dorsal and posterior part of the insula and parietal operculum.
  • (4) The rarest pattern 1 (11%) included epigastric sensations and various combination of integrated gestures, fear/rage, and gustatory sensation, without elementary motor or orofacial signs. It primarily involved the most anterior and the ventral portions of the insula, as well as the mesial temporal lobe structures.

Singh et al.[24▪▪] used agglomerative hierarchical cluster analysis to distinguish two primary clinical patterns associated with the insular gyri anterior and posterior to the central sulcus of the insula, respectively. The anterior cluster was further divided into a more anterior (anterior and middle short gyri) and middle (posterior short gyrus) pattern. The most anterior pattern, affecting 50% of patients, was characterized by lack of aura or a cognitive aura, with some elementary orofacial signs and early hypermotor behavior or spasms. The middle pattern, observed in 17% of patients, was associated with early and marked opercular motor and respiratory signs, as well as some of the features of the other patterns (brief hypermotor behavior, somatosensory aura, and posturing). The most posterior pattern involved the anterior long gyrus in 33% of cases, and included somatosensory or gustatory auras with a reflex component in most cases, as well as early posturing. Hypermotor behavior could also occur at a late stage. The two most anterior patterns were associated with seizure of shorter duration and quicker recovery than the most posterior one.

Peltola et al.[25▪▪] also distinguished the semiology of ictal onset zones located anteriorly and posteriorly to the central sulcus of the insula. Some ictal signs were only observed in anterior insular seizures, including hypermotor behaviors, viscerosensory and emotional sensations, dysprosodia and manual automatisms. Conversely, focal elementary motor signs, pain, and gustatory sensations were primarily observed in posterior insular seizures. The most prevalent somatosensory and autonomic signs as well as ipsilateral eye blinking and speech disturbances in general, were equally frequent in anterior and posterior insular ictal onset zones.

Overall, recent data confirm and refine previous knowledge of the various combinations of ictal signs observed in insular and insulo-opercular seizures, pointing to five partly overlapping patterns (Fig. 1):

  • (1) Dorsal anterior (anterior ± middle short gyri) characterized by lack of aura (or cognitive aura) with early hypermotor behavior, corresponding to previous fronto-like pattern [4].
  • (2) Dorsal mid-anterior (posterior ± middle short gyri) characterized by prominent fronto-opercular signs, including laryngeal sensations, salivation, elementary orofacial motor and other viscerosensory signs, corresponding to an anterior-dorsal perisylvian pattern.
  • (3) Dorsal posterior (anterior long gyrus) characterized by prominent somatosensory signs, pain, gustatory sensation, a reflex component, and tonic/dystonic posturing, corresponding to a posterior-dorsal perisylvian pattern. Such foci might also give rise to hypermotor behaviors [3,26▪▪].
  • (4) Ventral posterior (posterior long gyrus) characterized by auditory signs and marked tonic/dystonic posturing, corresponding to a posterior-ventral perisylvian pattern
  • (5) Ventral anterior (long gyri) characterized by epigastric sensations and emotional auras with gestures, corresponding to previous temporo-insular / temporal plus pattern [1,33].
F1
FIGURE 1:
A schematic sagittal representation of the insula, its anterior and posterior portions separated by the central sulcus of the insula (dashed white). Five clinical ictal patterns have been described in insular and insulo-opercular seizures, and associated to various insular subregions colored in blue (dorsal anterior pattern, primarily involving the dorsal part of the anterior short gyrus and overlying frontal operculum), green (dorsal mid-anterior pattern, primarily involving the dorsal portion of the middle and posterior short gyri and overlying frontal operculum), red (dorsal posterior pattern, primarily involving the posterior part of the anterior long gyrus and overlying parietal operculum), yellow (ventral posterior pattern, primarily involving the posterior long gyrus and overlying temporal operculum) and orange (ventral anterior pattern, strongly connected to the temporo-limbic structures). Note that all five patterns are overlapping, accounting for transition patterns. Also note that hypermotor behaviors are typical of the dorsal anterior pattern, but can be also observed in dorsal posterior pattern.

MINIMAL INVASIVE SURGERY OF INSULAR EPILEPSY

Open surgery of the insula is challenging due to highly functional covering opercula and a dense wall of arteries whose injury can lead to hemiparesis and dysphasia (see Table 4). Hence, alternate minimally invasive surgical options have been recently tried, including laser interstitial thermal therapy (LITT), less frequently radiofrequency thermocoagulation (RFTC), and more rarely RNS and gamma-knife surgery (GKS). As detailed below, only a limited number of patients have been treated with these procedures, 83 in total, many of whom had a complex history of failed past-surgery, and an epileptogenic zone extending beyond the borders of the targeted insula. Overall, minimal invasive surgery was associated with lower rate of seizure control (about 50%) compared to open-surgery, and yet a nonnegligible incidence of neurological deficits albeit mostly transient. In the largest surgical series of 44 insulo-opercular cortectomies performed in 43 patients, 77% achieved a class 1 outcome at a mean follow-up of 6 years [53], while 7% suffered a post-operative permanent deficit [54]. Comparisons remain however difficult for the lack of controlled study and differences in the populations treated with minimal invasive versus open-surgery.

Table 4 - Minimal invasive surgery in patients with insular or insulo-opercular seizures
Prior neurosurgery Surgical target(s), peri-operative deficit and seizure outcome
Publication Number of patients % adults Normal insula on MRI Lesion within and/or outside the insula Insular Other icEEG Pure insular ictal onset zone > insula Left side Motor and/or language deficit (% permanent) Mean follow-up (year) Engel class I outcome Still on ASM
Laser interstitial thermal therapy (LITT)
 Haswali et al.[34] 1 100% 0% Stroke 0 0 100% 100% 0 0% 100% (100%) 1.9 100% 100%
 Perry et al.[35] 20 0 70% 3 FCD, 3 TSC, 1 PMG 35% 65% 100% UNK 35% 45% 30% (0%) 1.7 50% 80%
 Hale et al.[36] 14a 0 UNK 13 FCD, 1 TSC 36% 43% 100% UNK 21% 64% 36% (0%) 2.4 43% UNK
 Alexander et al.[39,40] 3 0 100% 2 FCD, 1 cavernoma 0 66% 100% 100% 0 66% 0 1.0 100% 100%
 Easwaran et al.[38] 1 0 0% Astrocytoma 100% 0 0 0 0 100% 0 1.9 100% 100%
 Gireesh et al.[37] 6 33% 100% 0 17% 100% 50% 17% 83% 67% (0%) 1.5 67% UNK
 TOTAL 38 ∗ 5% 92% ≥ 39% FCD 29% 42% 97% UNK 24% 55% 34% (3%) 58% 84%
Radiofrequency thermocoagulation (RFTC)
 Guenot et al.[41] 7 100% 57% 3 FCD 0 0 100% 29% 71% 14% 0 UNK 0% 100%
 Catenoix et al.[42] 5 100% 0% 4 FCD, 1 Heterotopia 0 0 100% 100% 0 60% 0 3.9 20% 100%
 Wellmer et al.[51] 1 100% 0% FCD 0 0 100% 100% 0 100% 100% (100%) 1.5 0% 100%
 Yu et al.[52] 1 0 100% 0 0 100% 100% 0 100% 0 0.4 100% 100%
 Mulatti et al.[43▪] 19 70% 79% 10 FCD, 1 PMG, 2 tumors 11% 100% 0% 0 63% 16% (0%) 5.4 53% 95%
 TOTAL 33 79% 61% 55% FCD 6% 100% 27% 15% 39% 12% (3%) 36% 97%
Gamma-knife radiosurgery (GKS)
 Irisliminane et al.[46] 3 100% 33% 1 cavernoma, 1 FCD 33% 0% 66% 100% 0 100% 0 10 66% 100%
Responsive neurostimulation (RNS)
 Smith et al.[44] 1 100% 0% FCD 100% 100% 100% 0 100% 100% 0 4.0 0% 100%
 Chen et al.[45] 8 100% 88% Not specified 13% 75% 75% 0 100% 88% 0 0.9 0% 100%
 TOTAL 9 100% 78% 11% FCD 22% 78% 78% 0 100% 89% 0 0 100%
 TOTAL ALL MODALITIES 83 ∗ 49% 67% 42% FCD 19% 31% 95% 14% 28% 53% 20% (2%) 43% 93%
ASM, antiseizure medication; FCD, focal cortical dysplasia; IcEEG, intracranial EEG; PMG, polymicrogyria; UNK, unknown.
a7 out of the 14 patients from this series were also reported in Perry et al. 2017 and are thus not counted in the total number of cases

MRI-guided laser interstitial thermal therapy

LITT is particularly appealing for insular epilepsy as damage to opercula and M2 vessels/perforators are minimized by inserting the probe(s) through a parasagittal entry point, and because vascular structures with internal flow act as heat sinks. Part or the entire insula can be ablated using one to three probe(s) and by pulling back on the probe.

The first case of LITT-treated insular epilepsy was reported in 2014 [34]. Although the patient was rendered seizure-free, postoperative neuropsychological assessment at 8 months revealed a decline on numerous verbal measures. The largest LITT series was published three years later and included a pediatric cohort of 20 insular or insular plus epilepsy, including 70% MRI-negative and 85% who had previously failed previous surgery [35]. Ablations were restricted to the insula in 13 patients (65%) and extended to the opercula in 7. At a mean follow-up (FU) of 20 months, 50% of patients were in Engel Class I. Complications occurred in 35% of patients, including 6 mild hemiparesis that resolved within 6 months, and one dysphasia that resolved by 3 months. Despite temperature control settings, structures lateral to the putamen appeared injured on postoperative MRI.

More recently, the same group reported outcome from LITT and open surgery in 26 children with SEEG-proven insular or insular plus epilepsies [36]. The two groups were not randomized and significantly differed. The 14 patients who underwent LITT (including seven previously reported) had a high rate of normal MRI (60%) and failed previous surgery (71%), whereas the 12 patients who underwent open-surgery had only 17% of both normal MRI and failed past-surgery. LITT was associated with 43% of Engel Class I outcome at a mean FU of 1.9 years, with half of patients showing transient postoperative hemiparesis. Open surgery was associated with comparable results, that is, 50% Engel class 1 at a mean FU of 3 years, and 50% of transient hemiparesis.

Another series of six insular LITT reported 67% of Engel class I outcome [37], including one that required two successive procedures, one where LITT was also applied to the cingulate gyrus, and one with a large per-operative hemorrhage that led to stop the LITT procedure and proceed to an emergency large decompressive craniectomy. In total, 67% of patients developed transient motor and language deficit, which required significant rehabilitation in the patient who suffered an hemorrhage. A few other case reports and small series of successful insular LITT were reported in patients with recurrent low-grade glioma [38], and failed previous resection of a cavernoma or a FCD [39,40]

Radiofrequency thermocoagulation

RFTC is typically performed through the depth electrodes implanted for SEEG, enabling lesions of 6–8 mm in diameter at each treated electrode contact [41]. The possibility to control insular seizures with RFTC was optimized by placing a larger number of electrodes than would be needed for diagnostic purpose within the targeted insula, in five patients with epileptogenic malformation of cortical development [42]. This resulted in long-lasting seizure-freedom in 20% of cases, and >50% seizure reduction in another 40%. By increasing the number of insula-targeting electrodes with an average of 18 contacts per subject, a more recent series of 19 patients, with mostly insulo-opercular seizures, reported 53% of Engel class I outcome and another 21% of class II [43▪]. Transient postoperative deficits (mild hemiparesis, dysarthria, hypoesthesia, dysgueusia) were observed in 42% of patients, with rapid recovery in all but one with persistent dysarthria. Neurological deficits were related to higher number of RFTC procedures and greater volume of RFTC lesions. Factors associated with seizure-freedom were the presence of FCD or a localized epileptogenic zone requiring low RFTC volume. The authors suggested that an optimal volume of RFTC around 2 cm3 offered the best compromise between efficacy and safety.

Other minimal invasive surgical procedures

RNS targeting the anterior insula was first reported in a patient who had failed previous insulo-opercular surgery, and resulted in a 60% reduction in seizure frequency [44]. More recently, insular RNS was performed in a series of eight patients, most of whom suffered from an epileptogenic zone extending well beyond the insula with six having undergone prior extra-insular surgery [45]. At last visit, no patient was seizure free, whereas seizures were reduced by >50% in half of subjects. However, several confounding factors hamper interpretation of data (adjustment of other therapies, short follow-up in several patients), making it difficult to conclude on the potential of RNS to treat insular or insulo-opercular epilepsies.

GKS was reported in a small series of three patients with pure insular epilepsy [46], including two demonstrated by intracranial EEG and one with a posterior insular cavernoma. Two patients reached an Engel Class I outcome whereas the third improved but eventually underwent resective surgery to become seizure-free.

CONCLUSION

Significant progress has been made in the identification and surgical treatment of insular and insulo-opercular epilepsies. Given the diagnostic and therapeutic challenges associated with this condition, precise anatomo-clinical correlations shall help to optimally target the various insular and opercular subregions using SEEG, in particular when MRI is normal. Minimal invasive techniques appear well suited to the surgical management of patients with insular epilepsies, but have not yet demonstrated a clear advantage in terms of risk/benefit ratio. Future studies will need to determine whether such approaches are more appropriate for small well-localized insular foci, patients with significant co-morbidities, or as part of a step-wise approach, starting with less invasive procedures to move to more invasive approaches when necessary.

Acknowledgements

We thank France Ravey for her support in handling references.

Financial support and sponsorship

None.

Conflicts of interest

There are no conflicts of interest.

REFERENCES AND RECOMMENDED READING

Papers of particular interest, published within the annual period of review, have been highlighted as:

▪ of special interest

▪▪ of outstanding interest

REFERENCES

1. Ryvlin P. Avoid falling into the depths of the insular trap. Epileptic Disord 2006; 8: (Suppl 2): S37–S56.
2. Ryvlin P, Picard F. Invasive investigation of insular cortex epilepsy. J Clin Neurophysiol 2017; 34:328–332.
3. Kaido T, Otsuki T, Nakama H, et al. Complex behavioral automatism arising from insular cortex. Epilepsy Behav 2006; 8:315–319.
4. Ryvlin P, Minotti L, Demarquay G, et al. Nocturnal hypermotor seizures, suggesting frontal lobe epilepsy, can originate in the insula. Epilepsia 2006; 47:755–765.
5. Dobesberger J, Ortler M, Unterberger I, et al. Successful surgical treatment of insular epilepsy with nocturnal hypermotor seizures. Epilepsia 2008; 49:159–162.
6. Zhang H, Yao Q, Zhao X, et al. A hypermotor seizure with a focal orbital frontal lesion originating in the insula: a case report. Epilepsy Res 2008; 82:211–214.
7. Proserpio P, Cossu M, Francione S, et al. Insular-opercular seizures manifesting with sleep-related paroxysmal motor behaviors: a stereo-EEG study. Epilepsia 2011; 52:1781–1791.
8. Montavont A, Mauguière F, Mazzola L, et al. On the origin of painful somatosensory seizures. Neurology 2015; 84:594–601.
9. Blauwblomme T, Kahane P, Minotti L, et al. Multimodal imaging reveals the role of γ activity in eating-reflex seizures. J Neurol Neurosurg Psychiatry 2011; 82:1171–1173.
10. Thompson SA, Alexopoulos A, Bingaman W, et al. Auditory aura in frontal opercular epilepsy: sounds from afar. Epileptic Disord 2015; 17:150–155.
11. Xiao H, Tran TP, Pétrin M, et al. Reflex operculoinsular seizures. Epileptic Disord 2016; 18:19–25.
12. Tran TP, Truong VT, Wilk M, et al. Different localizations underlying cortical gelastic epilepsy: case series and review of literature. Epilepsy Behav 2014; 35:34–41.
13. Seeck M, Zaim S, Chaves-Vischer V, et al. Ictal bradycardia in a young child with focal cortical dysplasia in the right insular cortex. Eur J Paediatr Neurol 2003; 7:177–181.
14. Surges R, Scott CA, Walker MC. Peri-ictal atrioventricular conduction block in a patient with a lesion in the left insula: case report and review of the literature. Epilepsy Behav 2009; 16:347–349.
15. Tayah T, Savard M, Desbiens R, Nguyen DK. Ictal bradycardia and asystole in an adult with a focal left insular lesion. Clin Neurol Neurosurg 2013; 115:1885–1887.
16. Davis KA, Cantor C, Maus D, Herman ST. A neurological cause of recurrent choking during sleep. J Clin Sleep Med 2008; 4:586–587.
17. Dionisio S, Koenig A, Murray J, Somerville E. A gut feeling about insular seizures. BMJ Case Rep 2011; 2011:bcr1220103647.
18. Nesbitt AD, Kosky CA, Leschziner GD. Insular seizures causing sleep-related breathlessness. Lancet 2013; 382:1756.
19. Geevasinga N, Archer JS, Ng K. Choking, asphyxiation and the insular seizure. J Clin Neurosci 2014; 21:688–689.
20. Anzellotti F, Franciotti R, Bonanni L, et al. Persistent genital arousal disorder associated with functional hyperconnectivity of an epileptic focus. Neuroscience 2010; 167:88–96.
21. Landtblom AM, Lindehammar H, Karlsson H, Craig AD. Insular cortex activation in a patient with ‘sensed presence’/ecstatic seizures. Epilepsy Behav 2011; 20:714–718.
22. Freri E, Matricardi S, Gozzo F, et al. Perisylvian, including insular, childhood epilepsy: presurgical workup and surgical outcome. Epilepsia 2017; 58:1360–1369.
23▪. Wang X, Hu W, McGonigal A, et al. Electroclinical features of insulo-opercular epilepsy: an SEEG and PET study. Ann Clin Transl Neurol 2019; 6:1165–1177.
24▪▪. Singh R, Principe A, Tadel F, et al. Mapping the insula with stereo-electroencephalography: the emergence of semiology in Insula Lobe Seizures. Ann Neurol 2020; 88:477–488.
25▪▪. Peltola ME, Trébuchon A, Lagarde S, et al. Anatomoelectroclinical features of SEEG-confirmed pure insular-onset epilepsy. Epilepsy Behav 2020; 105:106964.
26▪▪. Wang H, McGonigal A, Zhang K, et al. Semiologic subgroups of insulo-opercular seizures based on connectional architecture atlas. Epilepsia 2020; 61:984–994.
27. Mazzola L, Mauguière F, Isnard J. Electrical stimulations of the human insula: their contribution to the ictal semiology of Insular Seizures. J Clin Neurophysiol 2017; 34:307–314.
28. Aldosari MM, Joswig H, Marti AS, et al. Nonlesional eating epilepsy with temporo-insular onset: a stereo-EEG study. Epilepsy Behav Rep 2020; 14:100368.
29. Bouthillier A, Nguyen DK. Epilepsy surgeries requiring an operculoinsular cortectomy: Operative technique and results. Neurosurgery 2017; 81:602–612.
30. Dylgjeri S, Taussig D, Chipaux M, et al. Insular and insulo- opercular epilepsy in childhood: an SEEG study. Seizure 2014; 23:300–308.
31. Kharytonov V, Dulac O. Are epileptic spasms a seizure type for the insular region? Neuropediatrics 2020; 51:295–297.
32. Wiwchar LD, Hader W, Pauranik A, et al. Focal seizures associated with the chapeau de gendarme sign or ictal pouting of insular origin. Epilepsy Behav Rep 2019; 12:100347.
33. Isnard J, Guénot M, Sindou M, Mauguière F. Clinical manifestations of insular lobe seizures: a stereo-electroencephalographic study. Epilepsia 2004; 45:1079–1090.
34. Hawasli AH, Bandt SK, Hogan RE, et al. Laser ablation as treatment strategy for medically refractory dominant insular epilepsy: therapeutic and functional considerations. Stereotact Funct Neurosurg 2014; 92:397–404.
35. Perry MS, Donahue DJ, Malik SI, et al. Magnetic resonance imaging-guided laser interstitial thermal therapy as treatment for intractable insular epilepsy in children. J Neurosurg Pediatr 2017; 20:575–582.
36. Hale AT, Sen S, Haider AS, et al. Open resection versus laser interstitial thermal therapy for the treatment of Pediatric Insular Epilepsy. Neurosurgery 2019; 85:E730–E736.
37. Gireesh ED, Lee K, Skinner H, et al. Intracranial EEG and laser interstitial thermal therapy in MRI-negative insular and/or cingulate epilepsy: case series. J Neurosurg 2020; 1–9. doi: 10.3171/2020.7.JNS201912. [Epub ahead of print].
38. Easwaran TP, Lion A, Vortmeyer AO, et al. Seizure freedom from recurrent insular low-grade glioma following laser interstitial thermal therapy. Childs Nerv Syst 2020; 36:1055–1059.
39. Alexander H, Cobourn K, Fayed I, et al. Magnetic resonance-guided laser interstitial thermal therapy for the treatment of nonlesional insular epilepsy in pediatric patients: thermal dynamic and volumetric factors influencing seizure outcomes. Childs Nerv Syst 2019; 35:453–461.
40. Alexander H, Cobourn K, Fayed I, Oluigbo CO. Magnetic resonance-guided laser interstitial thermal therapy for the treatment of nonlesional insular epilepsy in pediatric patients: technical considerations. Pediatr Neurosurg 2020; 55:155–162.
41. Guénot M, Isnard J, Catenoix H, et al. SEEG-guided RF-thermocoagulation of epileptic foci: a therapeutic alternative for drug-resistant nonoperable partial epilepsies. Adv Tech Stand Neurosurg 2011; 36:61–78.
42. Catenoix H, Mauguière F, Montavont A, et al. Seizures outcome after stereoelectroencephalography-guided thermocoagulations in malformations of cortical development poorly accessible to surgical resection. Neurosurgery 2015; 77:9–14.
43▪. Mullatti N, Landre E, Mellerio C, et al. Stereotactic thermocoagulation for insular epilepsy: lessons from successes and failures. Epilepsia 2019; 60:1565–1579.
44. Smith JR, Fountas KN, Murro AM, et al. Closed-loop stimulation in the control of focal epilepsy of insular origin. Stereotact Funct Neurosurg 2010; 88:281–287.
45. Chen H, Dugan P, Chong DJ, et al. Application of RNS in refractory epilepsy: targeting insula. Epilepsia Open 2017; 2:345–349.
46. Irislimane M, Mathieu D, Bouthillier A, et al. Gamma knife surgery for refractory insular cortex epilepsy. Stereotact Funct Neurosurg 2013; 91:170–176.
47. Rossetti AO, Mortati KA, Black PM, Bromfield EB. Simple partial seizures with hemisensory phenomena and dysgeusia: an insular pattern. Epilepsia 2005; 46:590–591.
48. Kriegel MF, Roberts DW, Jobst BC. Orbitofrontal and insular epilepsy. J Clin Neurophysiol 2012; 29:385–391.
49. Gras-Combe G, Minotti L, Hoffmann D, et al. Surgery for Nontumoral Insular Epilepsy Explored by stereoelectroencephalography. Neurosurgery 2016; 79:578–588.
50. Kurukumbi M, Leiphart J, Singer L. A rare case of insular epilepsy: not to be missed in refractory epilepsy patients. Cureus 2019; 11:e5434.
    51. Wellmer J, Parpaley Y, Rampp S, et al. Lesion guided stereotactic radiofrequency thermocoagulation for palliative, in selected cases curative epilepsy surgery. Epilepsy Res 2016; 121:39–46.
    52. Yu K, Liu C, Yu T, et al. Out-of-body experience in the anterior insular cortex during the intracranial electrodes stimulation in an epileptic child. J Clin Neurosci 2018; 54:122–125.
    53. Bouthillier A, Weil AG, Martineau L, et al. Operculoinsular cortectomy for refractory epilepsy. Part 1: Is it effective? J Neurosurg 2019; Sep 20:1-10. doi: 10.3171/2019.4.JNS1912. [Epub ahead of print].
    54. Bouthillier A, Weil AG, Martineau L, et al. Operculoinsular cortectomy for refractory epilepsy. Part 2: Is it safe? J Neurosurg 2019; Sep 20:1-11. doi: 10.3171/2019.6.JNS191126. [Epub ahead of print].
    Keywords:

    epilepsy; ictal signs; insula; semiology; surgery

    Copyright © 2021 Wolters Kluwer Health, Inc. All rights reserved.