Impact of laterality on seizure duration in mesial temporal lobe epilepsy with hippocampal sclerosis

Highlights

• •Right-sided MTLE-HS shows longer seizures, with increased longest and mean seizure durations.
• •Prolonged seizures without bilateral tonic-clonic spread occur more often in right-sided MTLE-HS.
• •Seizure duration differs between hemispheres, indicating lateralization in MTLE-HS.

Abstract

Objective

This study aims to evaluate differences in seizure durations in mesial temporal lobe epilepsy with hippocampal sclerosis (MTLE-HS) based on laterality, along with other related findings.

Methods

This retrospective trial (2009–2023) at Istanbul University-Cerrahpasa’s VEM Unit reviewed patients diagnosed with MTLE-HS via MRI and/or postoperative histopathology. Demographic data, epilepsy duration, and MTLE laterality were recorded. Seizure durations (excluding bilateral tonic-clonic periods) were analyzed, including the shortest, longest, and average seizure durations per admission. The occurrence of prolonged seizures without progression to bilateral tonic-clonic seizures (PS) was also documented.

Results

The study included 82 participants, with 58.5% (n = 48) left-sided and 41.5% (n = 34) right-sided MTLE-HS cases. The shortest recorded seizure duration showed no significant difference between sides (p > 0.05), but the longest seizure was significantly longer in right-sided MTLE-HS (p = 0.014). The mean seizure duration per patient was also significantly longer in the right-sided group (p = 0.035). PS occurred in 17.1% (n = 14) of participants, with a significantly higher incidence in right-sided MTLE-HS (p = 0.002).

Significance

This study suggests that seizure durations may be longer in right-sided (non-dominant) MTLE-HS compared to left-sided (dominant) cases, contributing new insights into the lateralization of MTLE-HS.

1. Introduction

Temporal lobe epilepsy (TLE) is the most common type of focal epilepsy and primarily involves a disorder of hippocampal function caused by increased neuronal excitability [1], [2]. Among the different underlying mechanisms, mesial temporal lobe epilepsy (MTLE) stands out in TLE, a syndrome encompassing various clinical and anatomical changes, due to its clinical manifestations, prevalence, and success in surgical treatment [3], [4]. MTLE is one of the best-known and most extensively studied epilepsy syndromes. While MTLE may arise from different etiologies, including long-term epilepsy-associated tumors or cavernomas involving limbic structures, hippocampal sclerosis (MTLE-HS) remains the most frequently observed cause [5], [6].

In cases of MTLE-HS, laterality is an important parameter in terms of the clinical findings and treatment approach of the disease. It is known that seizure semiology and neuropsychological profiles differ between right and left MTLE-HS cases. Regarding seizure semiology, automatisms such as preservation of ictal speech, spitting, musical vocalization, nose wiping and coughing are predominantly associated with the non-dominant hemisphere; whereas clinical signs such as ictal anomia and postictal dysphasia are reported to be associated with dominant hemisphere [7], [8], [9], [10], [11], [12]. From a cognitive profile perspective, non-verbal memory impairments are most prominent in patients with nondominant hemisphere MTLE-HS, whereas verbal memory impairments are more common in those with dominant hemisphere MTLE-HS [13].

A significant yet relatively neglected clinical parameter in the daily lives of individuals with epilepsy, and particularly during the process of recording seizures with video electroencephalography (EEG), is seizure duration. The underlying mechanisms of seizure termination have not been studied as extensively as seizure onsets. Although some studies have identified potential mechanisms related to seizure termination, the existing literature on this topic is sparse. Cortico-subcortical networks, particularly thalamic nuclei and their cortical connections, may contribute to seizure termination and thereby influence seizure duration [14], [15], [16].

The differences in semiology, neuropsychology, and network connections caused by laterality in MTLE-HS cases are relatively well-documented, but the variation in seizure duration between the dominant and non-dominant hemispheres is a feature that has not been previously reported in the literature. The aim of this study was to investigate whether seizure duration differs according to seizure laterality in patients with unilateral MTLE-HS, and to explore the potential clinical relevance of this observation.

2. MATERIAL AND METHODS

Patient Selection

This retrospective cohort study was conducted between 2009 and 2023 at the Video EEG Monitoring (VEM) Unit of the Department of Neurology, Istanbul University-Cerrahpasa, Faculty of Medicine. During this period, the diagnoses of patients admitted to the VEM unit were systematically reviewed. Patients diagnosed with mesial temporal lobe epilepsy with hippocampal sclerosis (MTLE-HS) based on magnetic resonance imaging (MRI) and/or postoperative histopathological examination were identified. Patients with bilateral mesial temporal lobe epilepsy or bilateral hippocampal sclerosis were excluded from the study. From this cohort, 82 adults with accessible clinical data and seizure records were included in the study. A key inclusion criterion was right-hand dominance. Handedness was assessed using standardized neuropsychological testing, and patients with non–right-hand dominance were excluded.

3. Data Collection

The demographic information of the patients included in the study, along with clinical findings such as epilepsy duration and MTLE laterality, were recorded.

Data were collected retrospectively by reviewing video-EEG monitoring reports. Antiseizure medication tapering during video-EEG monitoring was individualized based on patient-specific clinical parameters, including baseline seizure frequency, perceived efficacy of individual ASMs, prior sensitivity to medication changes, and safety considerations. Comprehensive clinical records of seizure onset and termination times were compiled. The recorded seizures needed to show some ictal EEG changes for inclusion. The onset of a seizure was defined as the first EEG change or the first subjective/objective clinical change, whichever occurred first. The end of the seizure was defined as the cessation of ictal EEG activity. All video-EEG recordings were originally reviewed and reported as part of routine clinical care by two senior epileptologists (Ç.Ö. and S.N.Y.), both board-certified neurologists with formal training in clinical neurophysiology and more than 30 years of experience in epilepsy and video-EEG interpretation. Seizure onset and termination times used in the present study were derived from these structured clinical reports. To reduce inter-rater variability and ensure reliability of seizure duration measurements, seizures with ambiguous, gradual, or poorly defined ictal EEG onset or termination were excluded from the analysis. Original EEG recordings were not systematically re-reviewed. EEG onset or termination was considered unclear when reports described ambiguous or gradual ictal EEG changes, significant EEG artifacts, delayed or poorly defined EEG correlates relative to clinical signs, or incomplete documentation preventing precise temporal determination. Such cases were excluded from analysis to ensure reliability of seizure duration measurements. For each patient, the durations of the shortest and longest seizures were recorded excluding the bilateral tonic clonic period. Additionally, the arithmetic mean of the durations of all seizures experienced during a single admission was calculated.

The occurrence of “prolonged seizures without progression to bilateral tonic-clonic seizures” (PS) was also documented during the recordings. Seizures lasting more than five minutes, accompanied by impaired awareness but without significant motor symptoms—or with only minor motor symptoms such as automatisms—were classified under this category. Additionally, seizures that progressed to bilateral tonic-clonic seizures were recorded separately.

Patients were subsequently categorized based on laterality into two groups: right (nondominant) hemisphere MTLE-HS and left (dominant) hemisphere MTLE-HS.

3.1. Statistical analysis

The NCSS (Number Cruncher Statistical System) 2007 (Kaysville, Utah, USA) software was used for statistical analyses. Descriptive statistical methods (Mean, Standard Deviation, Median, Frequency, Ratio, Minimum, Maximum) were employed to evaluate the study data. The Mann-Whitney U test and Pearson Chi-Square test were used for group comparisons. Significance was assessed at the levels of p < 0.01 and p < 0.05.

No a priori sample size calculation was performed, as this was a retrospective, exploratory study based on the available cohort of patients meeting strict inclusion criteria. The sample size was therefore determined by data availability rather than prospective power estimation.

4. Standard protocol Approvals, Registrations, and patient Consents

Informed consent was obtained from all patients/ legal guardians before they participated the study. The ethics committee approved the study protocol (Protocol No: 623691).

5. RESULTS

Sociodemographic and Clinical Data.

A total of 82 participants were included in the study, with a predominance of females (54.9%, n = 45) over males (45.1%, n = 37). The mean age of the participants was 37.1 ± 11.38 years, with a median age of 37 years (range 9–66 years).

The duration of epilepsy among participants varied widely, with a mean duration of 25.17 ± 12.58 years, ranging from 3 to 51 years (median: 23 years).

Regarding the side of MTLE-HS, 58.5% (n = 48) of the cases were left-sided, while 41.5% (n = 34) were right-sided.

Among the included patients, 46 patients underwent surgical resection and had postoperative histopathological confirmation of hippocampal sclerosis. In the remaining patients, the diagnosis of MTLE-HS was established based on characteristic MRI findings together with concordant clinical and video-EEG features. Lack of histopathological confirmation was mainly attributable to the absence of surgical intervention at the time of analysis. Among the 46 patients who underwent surgery (19 left-sided, 27 right-sided), postoperative seizure control was evaluated according to Engel class I versus non–Engel class I outcomes. No statistically significant difference was observed between the two groups (p = 0.809).

The duration of video-EEG monitoring for the entire cohort had a mean of 3.2 ± 0.9 days. When analyzed according to laterality, mean VEEG duration was 2.9 ± 1.1 days in the right-sided MTLE-HS group and 3.3 ± 1.0 days in the left-sided MTLE-HS group. There was no statistically significant difference in monitoring duration between the two groups (p = 0.097).

Correlation analyses did not reveal a significant association between epilepsy duration and mean seizure duration. Neither Pearson (r = 0.015, p = 0.900) nor Spearman correlation analysis (ρ =  − 0.034, p = 0.772) demonstrated a significant relationship.

An overview of the sociodemographic and clinical characteristics of the study population is presented in Table 1.

Table 1.

The Sociodemographic and Clinical Characteristics of the Patients.

		n	%
Gender	Female	45	54.9
Male	37	45.1
Side of TLE	Left	48	58.5
	Right	34	41.5
PSa	Present	14	17.1
	Absent	68	82.9
		Mean ± Sd	Min-Max (Median)
Age		37.1 ± 11.38	9–66 (37)
Duration Of Epilepsy (Year)		25.17 ± 12.58	3–51 (23)
Number Of The Recorded Seizures		2.18 ± 1.26	1–6 (2)
Duration Of The Shortest Seizure Recorded (Seconds)		95.06 ± 108.94	4–600 (71.5)
Duration Of The Longest Seizure Recorded (Seconds)		164.66 ± 225.61	4–1388 (96)

^aProlonged seizures without progression to bilateral tonic-clonic seizures.

The gender distribution between the right and left MTLE-HS groups did not show a statistically significant difference (p > 0.05).

During video-EEG monitoring, scalp-EEG–negative seizures were observed in both right- and left-sided MTLE-HS groups (7 patients in the left-sided group and 6 patients in the right-sided group). The number of scalp-EEG–negative seizures did not differ significantly between the two groups (p = 0.70).

Table 2 presents the distribution of variables between the two groups based on MTLE-HS laterality.

Table 2.

The Distribution of Variables Between the Two Groups Based on MTLE-HS Laterality.

		Side of MTLE-HS				ap
		Left (Dominant)		Right (Non-Dominant)
		n	%	n	%
Gender	Female	23	47.9	22	64.7	0.132
Male	25	52.1	12	35.3
PSb	Present	3	6.3	11	32.4	0.002**
	Absent	45	93.8	23	67.6
Ictal Scalp EEG Change	Present	35	72.9	25	73.5	0.950
	Absent	13	27.1	9	26.5

^aPearson Chi-Square Testi ** p < 0.01 * p < 0.05.

^bProlonged seizures without progression to bilateral tonic-clonic seizures.

The duration of epilepsy was slightly longer in the left MTLE-HS group, showing a statistically significant difference (p = 0.049). The average ages between the right and left MTLE-HS groups did not show a statistically significant difference (p > 0.05). Similarly, the number of recorded seizures did not significantly differ between the sides of MTLE-HS (p > 0.05).

Seizure types were recorded as focal seizures with preserved consciousness in 13 patients and focal seizures with impaired consciousness in 59 patients. It was observed that in a total of 19 patients (8 with right MTLE-HS and 11 with left MTLE-HS), the seizures evolved from focal to bilateral tonic-clonic seizures (FBTCS). There was no significant difference between the groups regarding laterality and the incidence of FBTCS (p = 0.96).

The average number of recorded seizures per participant was 2.18 ± 1.26, with the number of seizures ranging from 1 to 6 (median: 2). The shortest recorded seizure had a mean duration of 95.06 ± 108.94 s (median: 71.5 s). The longest recorded seizure had a mean duration of 164.66 ± 225.61 s (median: 96 s). The presence of prolonged seizures without progression to bilateral tonic-clonic seizures (PS) was reported in 17.1% (n = 14) of the participants, whereas 82.9% (n = 68) did not experience PS. The incidence rates of PS differed significantly according to the side of MTLE-HS (p = 0.002), with a higher incidence of PS observed in the nondominant (right) MTLE-HS cases.

The duration of the shortest recorded seizure did not differ significantly between the two sides (p > 0.05). In contrast, the duration of the longest recorded seizure was significantly longer in the right MTLE-HS group (p = 0.014). Additionally, the arithmetic mean of seizure durations for all seizures combined for each patient was significantly longer in the right MTLE-HS group (p = 0.035).

These findings are presented in Table 3.

Table 3.

The Measurements Regarding the Seizure Durations Between The Two Groups Based on MTLE-HS Laterality.

	Side of MTLE-HS		bp
	Left (Dominant)	Right (Non-Dominant)
	25–75 (Median)	25–75 (Median)
Age	27–49 (38)	28–43 (35)	0.201
Duration Of Epilepsy (Year)	16–38 (26)	11–30 (20)	0.049*
Number Of The Recorded Seizures	1–3 (2)	1–3 (2)	0.251
Duration Of The Shortest Seizure Recorded (Seconds)	42–96 (61)	54–104 (82)	0.226
Duration Of The Longest Seizure Recorded (Seconds)	59.5–130.5 (87.5)	79–250 (123.5)	0.014*
Average Seizure Duration Across All Seizures in Each Patient (Seconds)	55.5–103.5 (72.75)	65–187 (99)	0.035*

^bMann-Whitney U Testi ** p < 0.01 * p < 0.05.

6. DISCUSSION

It is well-established that laterality in MTLE is associated with varying clinical and cognitive outcomes. Our study has examined this phenomenon from the perspective of seizure duration. The principal finding from the current retrospective cohort is that seizure durations are longer in right-sided (non-dominant) MTLE with hippocampal sclerosis cases compared to left-sided (dominant) cases.

Animal studies have identified several parameters that influence the duration of seizures. Among the significant mechanisms demonstrated, one involves the inhibitory effects of the thalamus and subcortical structures in suppressing synchronous activity in the cortex [24]. Experimental animal studies have shown that these inhibitory effects can occur at the molecular level through GABAergic mechanisms and ions [17], [18], [19]. It is hypothesized that during seizure activity, as the intracellular and extracellular spaces become more acidic, a negative feedback mechanism is triggered, activating thalamo-cortical circuits that terminate the seizure [17], [18], [19]. While there is limited literature on the parameters affecting the activation of this inhibitory axis, a clinical study has shown that the size of the area where the seizure occurs is crucial [20]. As the seizure area increases, the time required to terminate the seizure also increases, thereby extending the overall seizure duration [20]. Although all patients shared a similar radiological diagnosis of hippocampal sclerosis, the extent of the epileptogenic network cannot be assumed to correspond directly to the size or laterality of the structural lesion. Seizure onset and propagation in MTLE-HS likely involve distributed cortico-subcortical networks, which may extend beyond the hippocampus and may not be fully captured by scalp EEG. Furthermore, prior neuropathological studies have demonstrated that bilateral mesial temporal involvement may be present despite unilateral imaging findings. Accordingly, seizure duration in MTLE-HS is more plausibly influenced by network-level dynamics rather than lesion size alone.

Over the last two decades, connectivity studies have established MTLE as a network disorder rather than a purely focal pathology. Alterations in cortico-subcortical and thalamocortical circuits influence seizure propagation, semiology, and consciousness [14], [21]. Connectivity changes at and beyond the epileptogenic focus are well documented, and clinical manifestations such as dystonic posturing, automatisms, and impaired consciousness have been linked to seizure spread involving the basal ganglia, somatosensory cortex, and thalamus in MTLE [22], [23].

From a lateralization perspective, thalamocortical circuit impairment has been shown to be more pronounced in the affected hemisphere in patients with MTLE [24]. Functional connectivity studies further demonstrate that the dominant and nondominant hemispheres differ in their physiological organization, with the dominant hemisphere generally exhibiting more extensive connectivity [15], [25]. These hemispheric differences support the concept that MTLE-HS affects distributed brain networks rather than representing a purely focal lesion, despite similar structural pathology across sides [14], [15]. Accordingly, neuroimaging studies using white matter tractography, voxel-based morphometry, and resting-state functional MRI have consistently demonstrated lateralization-related network differences, with left-sided MTLE showing a greater overall impact on network function [26], [27], [28], [29].

An additional finding of the present study was that epilepsy duration was modestly but significantly longer in patients with left-sided (dominant) MTLE-HS. Although the absolute difference was small and should be interpreted cautiously given the borderline statistical significance, this observation warrants consideration. Epilepsy duration was not correlated with mean seizure duration in our analysis, suggesting that the mechanisms governing chronic disease course and those influencing ictal termination are at least partially dissociable. Therefore, the longer epilepsy duration in left-sided MTLE-HS and the longer individual seizure durations in right-sided MTLE-HS likely reflect distinct pathophysiological dimensions—one related to diagnostic trajectory and network adaptation over time, and the other related to seizure propagation and termination dynamics [26], [27], [28], [29]. These findings further support the concept that hemispheric lateralization in MTLE-HS influences multiple levels of disease expression beyond structural pathology alone.

In MTLE-HS, lateralization-related differences in clinical manifestations are hypothesized to reflect hemispheric differences in network organization and vulnerability that evolve over the course of the disease, rather than focal pathology alone [15]. An apparent conceptual tension arises when considering prior literature reporting that left (dominant) temporal lobe epilepsy is associated with more pronounced network abnormalities on neuroimaging and worse cognitive outcomes, whereas our findings demonstrate longer seizure durations and a higher frequency of prolonged focal seizures in right (non-dominant) MTLE-HS. These observations may not be mutually exclusive. Neuroimaging and neuropsychological studies primarily reflect disruption of language- and memory-related cortico-cortical networks, which are more prominently lateralized to the dominant hemisphere. In contrast, seizure duration and termination likely depend on additional mechanisms, including subcortical modulatory pathways such as thalamocortical and basal ganglia circuits, propagation routes, and dynamic inhibitory processes. Thus, greater dominant-hemisphere vulnerability in cognitive networks does not necessarily imply less effective seizure termination, and longer seizures in non-dominant MTLE-HS may reflect differences in propagation and termination dynamics rather than a simple gradient of overall network impairment. Furthermore, scalp EEG may fail to detect early mesial temporal ictal activity and often identifies seizures only after propagation to neocortical regions, potentially influencing apparent seizure duration in a hemisphere-dependent manner. Accordingly, our mechanistic interpretation should be considered hypothesis-generating. Future studies incorporating intracranial EEG and multimodal connectivity analyses will be required to clarify whether hemispheric differences in seizure termination pathways underlie the observed laterality effects.

Another noteworthy finding of our study is that the incidence of prolonged seizures without progression to bilateral tonic-clonic seizures“ (PS) is markedly higher in right-sided (non-dominant) MTLE-HS cases compared to left-sided (dominant) cases.

Although not all prolonged focal seizures observed in this cohort fulfill formal diagnostic criteria for non-convulsive status epilepticus (NCSE), these events represent the severe end of the seizure-duration spectrum and provide a useful conceptual framework for interpreting our findings. Prolonged ictal activity without prominent convulsive features likely reflects alterations in seizure termination mechanisms rather than seizure initiation alone [30], [31], [32]. Experimental and limited human data suggest that prolonged seizures and NCSE are associated with network-level dysfunction, including impaired inhibitory modulation, altered thalamocortical and limbic connectivity, and disrupted synchronization dynamics within epileptogenic networks [33]. In this context, the findings from our study suggest that the observed differences in seizure durations may be due to the distinct wiring of the dominant and nondominant hemispheres. This interpretation, however, necessitates further investigation to confirm its validity. Additionally, our results suggest that onset in the nondominant hemisphere may be considered as a risk factor for NCSE, such as advanced age and presence of critical illness/coma.

7. LIMITATIONS

The major limitation of the current study is its reliance on retrospective data derived from the patients' video EEG monitoring reports. Additionally, the relatively small sample size is a limitation; however, to maintain homogeneity, only MTLE cases with the same pathology (HS) were selected. and inclusion criteria were stringently applied to ensure the reliability of the data. Also, seizure detection relied on scalp EEG, which may underestimate or delay the identification of focal mesial temporal seizures due to the depth and orientation of hippocampal generators. Second, seizure onset was defined using a combined clinical and EEG-based approach, which may introduce heterogeneity in seizure duration measurements and potentially favor longer seizures with slower propagation. Additionally, scalp-EEG–negative seizures were excluded from duration analyses, which may have influenced overall seizure length estimates. Yet, future studies incorporating intracranial EEG recordings will be essential to more precisely characterize seizure onset, propagation, and termination mechanisms. Another important limitation relates to ASM tapering during video-EEG monitoring. Medication withdrawal was individualized according to clinical parameters such as baseline seizure frequency, perceived efficacy of specific ASMs, prior sensitivity to dose changes, and patient safety considerations. Due to substantial heterogeneity in ASM regimens, formulations, and dosing schedules, it was not feasible to standardize or quantitatively compare the degree or rate of ASM reduction across patients. Although ASM tapering was applied according to similar clinical principles regardless of seizure laterality and no patient was completely withdrawn from medication, we cannot objectively exclude differences in ASM reduction between the right- and left-sided MTLE-HS groups. Therefore, the potential confounding effect of ASM tapering on seizure duration cannot be fully ruled out. Although only right-hand-dominant patients were included to reduce variability, the absence of systematic dominance testing (e.g., Wada or fMRI) represents an important limitation and precludes definitive conclusions regarding dominant versus non-dominant hemisphere function.

8. Conclusion

Our study is the first clinical research to demonstrate the impact of lateralization on seizure duration in cases of mesial temporal lobe epilepsy with hippocampal sclerosis (MTLE-HS). Seizure durations were found to be significantly longer in nondominant hemisphere MTLE-HS cases compared to the dominant hemisphere. Additionally, the incidence of prolonged seizures was higher in patients with nondominant hemisphere MTLE-HS. These clinical findings highlight the potential importance of lateralization in the etiopathogenesis of MTLE-HS. However, further functional imaging studies are needed to substantiate these findings.

Ethical Statement.

Informed consent was obtained from all patients/ legal guardians before they participated the study. The ethics committee approved the study protocol.

CRediT authorship contribution statement

Bengi Gul Turk: Writing – original draft, Project administration, Investigation. Yesim Abanoz: Investigation. Yasin Abanoz: Investigation. Bahadir Efe Arik: Investigation. Berke Ukus: Investigation. Elif Nur Yildirim: Investigation. Menekse Perk: Investigation. M.Sakir Delil: . Cigdem Ozkara: Visualization, Supervision, Investigation. S.Naz Yeni: Visualization, Supervision, Methodology, Investigation, Data curation, Conceptualization.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.