Journal Pre-proof

Comparative analysis of Lennox-Gastaut Syndrome with different subtypes of tonic

seizures: a single-center retrospective cohort study

Shiyu Wang, Xuan Zhao, Ting Li, Yu Jia, Liping Zhang, Xiaohong Qi, Yicong Lin,
Yuping Wang

PII: S0887-8994(24)00282-0
DOI: https://doi.org/10.1016/j.pediatrneurol.2024.07.018
Reference: PNU 10700

To appear in: Pediatric Neurology

Received Date: 2 August 2023

Revised Date: 2 July 2024
Accepted Date: 29 July 2024

Please cite this article as: Wang S, Zhao X, Li T, Jia Y, Zhang L, Qi X, Lin Y, Wang Y, Comparative
analysis of Lennox-Gastaut Syndrome with different subtypes of tonic seizures: a single-
center retrospective cohort study, Pediatric Neurology (2024), doi: https://doi.org/10.1016/
j.pediatrneurol.2024.07.018.

This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition
of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of
record. This version will undergo additional copyediting, typesetting and review before it is published
in its final form, but we are providing this version to give early visibility of the article. Please note that,
during the production process, errors may be discovered which could affect the content, and all legal
disclaimers that apply to the journal pertain.

© 2024 Published by Elsevier Inc.

Comparative analysis of Lennox-Gastaut Syndrome with different

subtypes of tonic seizures: a single-center retrospective cohort study

Shiyu Wang1,3,6, Xuan Zhao1,3,6, Ting Li1,3,6, Yu Jia2,3,4,5,6, Liping Zhang1,3,6, Xiaohong
Qi1,3,6, Yicong Lin2,3,4,5,6, Yuping Wang2,3,4,5,6,7,8*

1 Department of Pediatrics, Xuanwu Hospital, Capital Medical University, Beijing

100053, China.

f
2 Department of Neurology, Xuanwu Hospital, Capital Medical University, Beijing

oo
100053, China.

r
3 Center of Epilepsy, Beijing Institute for Brain Disorders, Capital Medical University,
-p
Beijing 100053, China.
re
4 Beijing Key Laboratory of Neuromodulation, Beijing 100053, China;
lP

5 Center for sleep and consciousness disorders, Beijing Institute for Brain Disorders,
Beijing 100053, China;
na

6 Collaborative Innovation Center for Brain Disorders, Capital Medical University,

ur

Beijing 100053, China;

7 Hebei Hospital of Xuanwu Hospital, Capital Medical University, Shijiazhuang
Jo

050000, China.
8 Neuromedical Technology Innovation Center of Hebei Province, Shijiazhuang
050000, China

*Corresponding author:
Yuping Wang
Department of Neurology, Xuanwu Hospital Capital Medical University, No.45
Changchun Street, Xicheng District, Beijing 100053, China, Fax: +86 83157841
E-mail address: [email protected].
 1 Comparative analysis of Lennox-Gastaut Syndrome with different

2 subtypes of tonic seizures: a single-center retrospective cohort study

3

4 Shiyu Wang1,3,6, Xuan Zhao1,3,6, Ting Li1,3,6, Yu Jia2,3,4,5,6, Liping Zhang1,3,6, Xiaohong
5 Qi1,3,6, Yicong Lin2,3,4,5,6, Yuping Wang2,3,4,5,6,7,8*
6

7 1 Department of Pediatrics, Xuanwu Hospital, Capital Medical University, Beijing

8 100053, China.

f
9 2 Department of Neurology, Xuanwu Hospital, Capital Medical University, Beijing

oo
10 100053, China.

r
11 3 Center of Epilepsy, Beijing Institute for Brain Disorders, Capital Medical University,
-p
12 Beijing 100053, China.
re
13 4 Beijing Key Laboratory of Neuromodulation, Beijing 100053, China;
lP

14 5 Center for sleep and consciousness disorders, Beijing Institute for Brain Disorders,
15 Beijing 100053, China;
na

16 6 Collaborative Innovation Center for Brain Disorders, Capital Medical University,

ur

17 Beijing 100053, China;

18 7 Hebei Hospital of Xuanwu Hospital, Capital Medical University, Shijiazhuang
Jo

19 050000, China.
20 8 Neuromedical Technology Innovation Center of Hebei Province, Shijiazhuang
21 050000, China
22 *Corresponding author:
23 Yuping Wang
24 Department of Neurology, Xuanwu Hospital Capital Medical University, Beijing,
25 100053, China, Fax: +86 83157841
26 E-mail address: [email protected].
27

28 The word count of abstract: 233; The word count of main text: 2530
29 The number of references: 37; The number of tables: 3; The number of figures: 2
1
 1 ABSTRACT
2 Objective: Lennox-Gastaut syndrome (LGS) is one of the most severe childhood-onset
3 epileptic encephalopathies, primarily characterized by tonic seizures. In clinical
4 practice, we have identified various subtypes of tonic seizures in LGS. This study aimed
5 to analyze the clinical characteristics, electrographic features, treatment responses, and
6 prognosis across different subtypes of LGS.
7 Methods: This retrospective cohort study included 46 patients diagnosed with LGS at
8 our center between January 2017 and January 2020. Patients were classified into four

f
9 groups based on tonic seizure subtypes: Group A (tonic), Group B (spasm-tonic), Group

oo
10 C (myoclonic-tonic), and Group D (combination of spasm-tonic and myoclonic-tonic).

r
11 Comprehensive clinical data were collected and analyzed.
12
-p
Results: Of the 46 patients, 33 were male. The mean age of onset for Group B (12.38 ±
re
13 7.85 months) was significantly younger than the other three groups (P = 0.02). No
lP

14 significant differences in etiology were found among the groups. Genetic analysis
15 identified mutations in SCN8A, MCCC2, STXBP1, GABRB3, and CACNA1H. After a
na

16 minimum follow-up of 24 months, the treatment outcomes were more favorable in

17 Groups A and C, whereas psychomotor development was notably poorer in Groups B
ur

18 and D.
Jo

19 Conclusions: The findings of this study suggest that LGS may present with distinct
20 subtypes of tonic seizures, with spasm-tonic seizures presenting at an earlier age. LGS
21 patients experiencing spasm-tonic seizures, with or without myoclonic-tonic seizures
22 exhibited poorer treatment responses and psychomotor development than those with
23 other subtypes.
24

25 Keywords: Lennox-Gastaut syndrome; Tonic seizures; Epileptic spasm;

26 Myoclonic seizures; Subtype of seizures
27
28

2
 1 Introduction
2 Lennox-Gastaut syndrome (LGS) is a severe form of developmental epileptic
3 encephalopathy (DEE). Typically, LGS manifests before the age of 8,1-3 although a
4 small number of patients present with late-onset LGS after this age.4,5 The etiology of
5 LGS remains unclear in 25%- 33% of cases,6 with no single underlying cause identified.
6 Seizures associated with LGS are typically resistant to anti-seizure medications (ASMs),
7 resulting in infrequent complete seizure control.7 The incidence of LGS is estimated at
8 2 per 100,000 children, accounting for approximately 2-5% of all childhood epilepsy
9 cases.6,8 Furthermore, over 90% of patients with LGS eventually develop moderate to

f
oo
10 severe intellectual disabilities. 2,9,10
11 According to the 2022 ILAE classification and definition, LGS is characterized by

r
12 -p
the presence of (1) multiple types of drug-resistant seizures with onset prior to 18 years
re
13 (one of which must include tonic seizures and at least one additional seizure type), (2)
14 cognitive and behavioral impairments, which may not be present at seizure onset, and
lP

15 (3) diffuse slow spike-and-wave and generalized paroxysmal fast activity on

na

16 electroencephalographic (EEG).1
17 Tonic seizures are a defining feature of LGS, occurring in approximately 56% to
ur

18 80% of cases2,9,11. In clinical practice, we have observed that some patients present with
Jo

19 other subtypes of tonic seizures, such as spasm-tonic and myoclonic-tonic seizures.

20 While these subtypes have been documented in DEEs, they are often overlooked at
21 initial presentation due to their subtle manifestations.12-14 Moreover, comparative
22 studies focusing on these different subtypes of tonic seizures within LGS are limited.
23 Therefore, this study aimed to meticulously analyze patients exhibiting these typical
24 features to evaluate the clinical characteristics, electrographic features, treatment
25 responses and prognosis across different subtypes of LGS.
26 Methods
27 Subject selection
28 This retrospective cohort study analyzed 1286 patients diagnosed with epilepsy at
29 the Department of Pediatrics, Xuanwu Hospital from January 2017 to January 2020.
30 Patients with LGS were selected according to medical records on the information
3
 1 system and EEG data sourced from the EEG database at Xuanwu Hospital. The
2 inclusion criteria were as follows: (1) seizure onset age ranging from 1 month to 14
3 years, (2) fulfillment of diagnostic criteria for LGS,1 (3) documented tonic seizures
4 during video-electroencephalogram (VEEG) and electromyography (EMG) monitoring,
5 and (4) VEEG recordings exceeding 18 hours, encompassing periods of both
6 wakefulness and sleep. The exclusion criteria included (1) patients with incomplete
7 medical records and (2) cases of repeated hospitalizations. Fouty-six patients were
8 ultimately enrolled in this study after obtaining informed consent from all patients’
9 guardians. The study was approved by the Ethics Committee of Xuanwu Hospital,

f
oo
10 Capital Medical University.
11 Clinical data and outcome measures

r
12 -p
We collected and reviewed comprehensive clinical data for each patient, including
demographic data, etiology, age at seizure onset, age at EEG recording, medical and
re
13

14 family history, brain magnetic resonance imaging (MRI) results, gene testing outcomes,
lP

15 treatment regimens, and findings from VEEG and magnetoencephalogram (MEG). The
na

16 EEG signals were bandpass filtered between 0.5 and 70Hz, digitized at a sampling rate
17 of 1024 Hz, and stored along with synchronous EMG for subsequent analysis. EEG
ur

18 recordings included both waking and sleeping periods and were analyzed using bipolar
Jo

19 and common average reference montages. Three experienced neurophysiologists (Xuan.

20 Z, Yu. J, and Ting. L.) independently analyzed all the recordings, and discrepancies in
21 EEG interpretations were collaboratively discussed and reassessed. The final
22 interpretations were decided by a senior neurophysiologist (Yi cong. L). Considering
23 the variability in seizure types and EEG abnormalities over time in some patients, we
24 focused on the most recurrent types and sites of abnormalities observed in the EEG
25 recordings. Genetic analysis was conducted for twelve patients. All patients underwent
26 clinical psychometric evaluations at the time of their LGS diagnosis.
27 Epileptic seizure types were classified according to the 2017 International League
28 Against Epilepsy (ILAE) guidelines. We divided all patients into four groups based on
29 the types of tonic seizures and EEG features. Group A consisted of patients with only
30 tonic seizures, Group B included those with spasm-tonic seizures, Group C comprised
4
 1 patients with myoclonic-tonic seizures, and Group D included patients with both
2 spasm-tonic seizures and myoclonic-tonic seizures. All patients were followed for a
3 minimum of 24 months post-LGS diagnosis, with outcome data obtained from
4 outpatient visits or telephone interviews. Assessment of seizure frequency and
5 psychomotor development were either conducted by physicians or reported by
6 guardians. Patients were categorized based on seizure frequency reduction—either
7 more than 50% or less than 50%. Additionally, psychomotor development was
8 evaluated and classified as normal, delayed, and regressed.
9 Statistical Analysis

f
oo
10 Statistical analyses were performed using IBM SPSS V.23.0 (SPSS Inc, Chicago,
11 IL, USA). Continuous variables were presented as mean±standard deviation or median

r
12 -p
(range), while categorical variables were expressed as number (percentage). The
comparison of means were analyzed using ANOVA or Welch’s test, and proportions
re
13

14 were assessed with the chi-square test or Fisher’s exact test. A P-value <0.05 was
lP

15 considered statistically significant. For multiple comparisons among the four groups,
na

16 the difference between any two groups was compared with multiple comparisons,
17 statistical significance was determined by a false discovery rate (FDR) corrected q–
ur

18 value <0.05.
Jo

19

20 Results
21 Demographics and etiologies of LGS with different subtypes of tonic seizures
22 In this study, 46 patients were enrolled and classified into four groups: Group A
23 (tonic seizures, n=20), Group B (spasm-tonic seizures, n=8), Group C (myoclonic-tonic
24 seizures, n=14), and Group D (combination of spasm-tonic seizures and myoclonic-
25 tonic seizures, n=4). As shown in Table 1, the mean age of onset for group B was 12.38
26 ± 7.85 months, significantly younger than the other groups (P = 0.02). In addition, nine
27 (19.5%) patients had a history of infantile epileptic spasms syndrome (IESS), with the
28 majority (6/9, 66.7%) later developing LGS characterized by spasm-tonic seizures (P
29 < 0.01). More than half of the patients (28/46, 60.9%) had an identifiable etiology.
30 Hypoxic ischemic encephalopathy (HIE) was the most frequent cause (10/46, 21.7%),
5
 1 followed by structural abnormalities and neurogenetic metabolic disease (each 5/46,
2 10.9%), viral encephalitis (4/46, 8.7%), cerebral hemorrhage (3/46, 6.5%), and tuberous
3 sclerosis (1/46, 2.1%). There were no significant differences in etiology among the four
4 groups (P > 0.05). Genetic testing was performed in 12 patients, yielding positive
5 results in five: two in group A (mutations in SCN8A, MCCC2), two in group B
6 (mutations in STXBP1, GABRB3), and one in group C (mutation in CACNA1H). The
7 SCN8A mutation was associated with a gain of function, and the patient with the
8 MCCC2 mutation was also diagnosed with a metabolic disorder (3-Methylcrotonyl-
9 coenzyme A carboxylase deficiency, MCCD). Psychomotor development was delayed

f
oo
10 in 40 patients (87.0%) and regressed in 6 patients (13.0%), with no statistically
11 significant differences among the four groups (P > 0.05).

r
12 -p
Characteristics of seizure types and interictal epileptiform discharges
re
13 Patients in Group A, characterized by tonic seizures, presented with sustained
14 increases in axial and limb muscle contractions lasting from 3 to 20 seconds. The
lP

15 corresponding ictal EEG, as shown in Figure 1A, revealed diffuse or bilateral fast
na

16 rhythm patterns (20-25 Hz) lasting for several seconds, while the EMG showed
17 sustained synchronous myoelectric activity. In Group B, patients with spasm-tonic
ur

18 seizures mainly presented with head nodding and sudden, sustained contractions of the
Jo

19 proximal limbs. As shown in Figure 1B, the ictal EEG showed generalized slow waves,
20 occasionally accompanied by superimposed fast activity, followed by diffuse fast
21 activity lasting a few seconds. The EMG indicated a crescendo-decrescendo activity
22 pattern preceding sustained myoelectric activity.14,15 In Group C, patients with
23 myoclonic-tonic seizures exhibited brief jerks of the limbs or trunk, followed by
24 sustained muscle contractions. As shown in Figure 1C, the ictal EEG displayed
25 polyspike-wave complexes, followed by generalized fast activity lasting a few seconds,
26 while the EMG showed burst activity followed by sustained myoelectric activity.13 In
27 addition, patients in Group D exhibited both spasm-tonic seizures and myoclonic-tonic
28 seizures, encompassing the seizure characteristics observed in Groups B and C.
29 In addition to tonic seizures, all patients exhibited one or more other types of
30 seizures, as shown in Table 2. These included epileptic spasms (13 patients, 28.3%),
6
 1 myoclonic seizures (21 patients, 45.7%), atypical absence (15 patients, 32.6%), atonic
2 seizures (6 patients, 13.0%), and focal seizures (7 patients, 15.2%). Notably, a higher
3 incidence of myoclonic seizures was observed in Group C (11patients, 78.6%; P = 0.01).
4 Interictal EEG patterns showed slow spike-wave complexes in 42 patients (91.3%),
5 spike rhythms in 28 patients (60.9%), and focal discharges in 21 patients (45.7%), with
6 16 of these (76.2%) predominantly localized over the frontal to temporal regions. There
7 were no statistically significant differences in interictal EEG patterns among the four
8 groups.
9 Treatment and outcomes

f
oo
10 As shown in Table 3, all patients were treated with at least 2 ASMs and some tried
11 alternative therapies, including the ketogenic diet in two patients and vagus nerve

r
12 -p
stimulation (VNS) in five patients. Of the cohort, 38 patients (82.6%) were followed
re
13 for minimum of 24 months post-initial consultation (median follow-up: 30 months,
14 range: 24-39 months), while eight patients were lost to follow-up. Notably, 22 patients
lP

15 (57.9%) achieved seizure freedom or experienced a reduction in seizure frequency of

na

16 more than 50%. The treatment response was significantly better Groups A (12 patients,
17 70.6%) and C (9 patients, 75%) compared to Groups B (1 patient, 16.7%) and D (0
ur

18 patient, 0%) (P = 0.01). In terms of psychomotor development, sixteen patients (42.1%)

Jo

19 were assessed as normal, eighteen (47.4%) as delayed, and four (10.5%) as regressed.
20 Psychomotor outcomes were significantly poorer in Groups B and D compared to
21 Groups A and C (P = 0.01).
22 Discussion
23 LGS is a rare, age-related epileptic encephalopathy, characterized by multiple
24 types of intractable seizures, cognitive and behavioral impairments, and characteristic
25 EEG abnormalities.1,2,6,9,10,16 While tonic seizures are the most defining type of seizure
26 in LGS, spasms and myoclonic seizures also frequently occur throughout the course of
27 the disease.2,9 In clinical practice, it is observed that LGS patients typically exhibit a
28 variety of seizure subtypes, each presenting with distinct clinical manifestations,
29 treatment responses, and outcomes. In this study, we analyzed the clinical data of 46
30 LGS patients, classifying them into four groups according to their seizure types and
7
 1 EEG patterns. We compared the clinical profiles across these groups and explored
2 potential underlying mechanisms.
3 In this study, 42 patients (91.3%) developed LGS before the age of 8 years,
4 aligning with findings from previous studies. Notably, the age of onset was
5 significantly earlier in Group B compared to the other groups (P < 0.05), with six
6 patients (75%) in Group B previously diagnosed with IESS (P < 0.05). Given that the
7 electroclinical features of IESS typically resolve by the age of 2 and LGS manifests
8 between the ages of 2 and 8, there appears to be an electroclinical overlap period
9 between IESS and LGS.17-21 Previous research by Rantala et al. indicated an average

f
oo
10 transition period from West syndrome to LGS of 2.8 years.22 In IESS, epileptic spasms
11 are prominent, whereas tonic seizures are the typical seizure type in LGS.23,24 It is

r
12 -p
probably that the transition from epileptic spasms to tonic seizures occurred earlier in
re
13 the patients of Group B, leading to an earlier documented age of LGS onset. Moreover,
14 it is suggested that spasm-tonic seizures may represent a transitional marker in LGS
lP

15 patients evolving from infantile spasms.

na

16 Regarding the etiology of LGS, it has been reported that over half of the patients
have an identifiable cause.25 In this study, 60.9% of patients had identifiable causes,
ur

17

18 including perinatal complications, structural abnormalities, CNS infections, and

Jo

19 neurogenetic metabolic disorders, with no significant differences observed among the

20 groups. Mutations in five genes, including SCN8A, STXBP1, MCCC2, GABRB3, and
21 CACNA1H were identified, with the SCN8A mutation resulting in a gain of function. A
22 previous study4 has indicated perinatal complications as the most frequent cause of LGS,
23 accounting for 25% of cases, followed by CNS infections (3.7%), head trauma (less
24 than 1%), and metabolic disorders (rare).2,9 Besides, pathogenic variants in several
25 genes such as SCN8A, SCN1A, STXBP1, GABRB3, ALG13, FOXG1, DNM1, and
26 CHD226-30 have been implicated in the etiology of LGS,2,27 illustrating the genetic
27 complexity associated with this condition. In our cohort, gene mutations in STXBP1
28 and GABRB3 were identified in Group B, SCN8A and MCCC2 in Group A, and
29 CACNA1H in Group C. Notably, mutations in SCN8A, STXBP1, and GABRB3 have
8
 1 been reported in both patients with IESS and LGS,27 suggesting a potential link between
2 these mutations and specific LGS subtypes. Furthermore, we identified a specific
3 mutation in MCCC2, known to be associated with generalized seizures and mental
4 retardation in patients with MCC2 deficiency. However, the relationship between
5 MCC2 deficiency and LGS remains unclear.31
6 In this study, over half of the patients showed improvement following treatment,
7 with 21.1% achieving seizure freedom and 42.1% exhibiting normal psychomotor
8 development. Particularly, more than 70% of LGS patients with either isolated tonic or
9 myoclonic-tonic seizures responded positively to treatments, indicating a more

f
oo
10 favorable prognosis for these patients. This finding aligns with previous reports
11 highlighting a better prognosis for the myoclonic variant of LGS.32 Conversely, patients

r
12 -p
in Group B and Group D, who experienced spasms-tonic with or without myoclonic-
re
13 tonic seizures, did not show significant treatment effects and exhibited poorer
14 psychomotor development. This suggests that LGS patients presenting with spasm-
lP

15 tonic seizure, with or without myoclonic-tonic seizures, may have worse outcomes.
na

16 In our study of LGS patients, we identified specific subtypes of tonic seizures,

17 including spasm-tonic seizures and myoclonic-tonic seizures, which were associated
ur

18 with variations in onset ages, outcomes, and psychomotor development. Given the
Jo

19 established progression from IESS to LGS, we propose the existence of a

20 developmental continuum encompassing both IESS and LGS. This suggests that the
21 epileptic syndrome may evolve within the same patient over time. The identification of
22 distinct subtypes of tonic seizures underscores the need for a more precise definition of
23 the electroclinical spectrum. In this study, interictal epileptiform discharges were
24 predominantly observed over the frontal to temporal regions in most patients. Previous
25 studies suggest that tonic, epileptic spasm, and myoclonic seizures may activate a
26 common brain network, with the frontal lobe playing a crucial role in this network.33-35
27 The rapid maturation of the frontal lobe during childhood is known to be epileptogenic
28 and could influence the progression of epilepsy from various causes.36,37 Based on our
29 observations that LGS frequently evolves into subtypes of tonic seizures, we propose
30 that these developments may be linked to the process of brain maturation.
9
 1 This study has several limitations. Firstly, it reflected clinical practice in a single
2 institution with a relatively small sample size, making it difficult to match baseline
3 characteristics across subgroups. This limitation introduced significant bias,
4 complicating any definitive conclusions about the association of specific tonic seizure
5 types with poor prognosis. Future studies should include more medical centers to
6 enhance the robustness of the findings. Secondly, the classification of different
7 subtypes of tonic seizures was somewhat subjective, and currently, there are no
8 established guidelines to standardize these classifications. Thirdly, given the
9 retrospective nature of this study, there is a possibility that patients may exhibit

f
oo
10 different types of seizures in the future. We plan to continue monitoring the
11 electroclinical features of LGS patients and consider including more diverse groups in

r
12 our research. -p
Conclusion
re
13

14 In our cohort, LGS frequently manifests as distinct subtypes of tonic seizures, with
lP

15 spasm-tonic seizures occurring at an earlier age than other subtypes. Patients with
na

16 spasm-tonic seizures, with or without concurrent myoclonic-tonic seizures, tend to

17 exhibit a less favorable response to treatment and poorer psychomotor development
ur

18 compared to those with other subtypes.

Jo

19
20
21 References
22 1. Specchio N, Wirrell EC, Scheffer IE, et al. International League Against Epilepsy classification
23 and definition of epilepsy syndromes with onset in childhood: Position paper by the ILAE Task
24 Force on Nosology and Definitions. Epilepsia. 2022;63(6):1398-1442.
25 2. Asadi-Pooya AA. Lennox-Gastaut syndrome: a comprehensive review. Neurological Sciences.
26 2017;39(3):403-414.
27 3. Bourgeois BFD, Douglass LM, Sankar R. Lennox-Gastaut syndrome: a consensus approach to
28 differential diagnosis. Epilepsia. 2014;55 Suppl 4:4-9.
29 4. Asadi-Pooya AA, Sharifzade M. Lennox-Gastaut syndrome in south Iran: electro-clinical
30 manifestations. Seizure. 2012;21(10):760-763.
31 5. Goldsmith IL, Zupanc ML, Buchhalter JR. Long-term seizure outcome in 74 patients with
32 Lennox-Gastaut syndrome: effects of incorporating MRI head imaging in defining the
33 cryptogenic subgroup. Epilepsia. 2000;41(4):395-399.
34 6. Camfield PR. Definition and natural history of Lennox-Gastaut syndrome. Epilepsia. 2011;52
35 Suppl 5:3-9.

10
 1 7. Michoulas A, Farrell K. Medical management of Lennox-Gastaut syndrome. CNS drugs.
2 2010;24(5):363-374.
3 8. Trevathan E, Murphy CC, Yeargin-Allsopp M. Prevalence and descriptive epidemiology of
4 Lennox-Gastaut syndrome among Atlanta children. Epilepsia. 1997;38(12):1283-1288.
5 9. Arzimanoglou A, French J, Blume WT, et al. Lennox-Gastaut syndrome: a consensus approach
6 on diagnosis, assessment, management, and trial methodology. The Lancet Neurology.
7 2009;8(1):82-93.
8 10. Mastrangelo M. Lennox–Gastaut Syndrome: A State of the Art Review. Neuropediatrics.
9 2017;48(03):143-151.
10 11. Cross JH, Auvin S, Falip M, Striano P, Arzimanoglou A. Expert Opinion on the Management of
11 Lennox-Gastaut Syndrome: Treatment Algorithms and Practical Considerations. Frontiers in
12 neurology. 2017;8:505.
13 12. Ronzano N, Valvo G, Ferrari AR, Guerrini R, Sicca F. Late-onset epileptic spasms: clinical

f
14 evidence and outcome in 34 patients. Journal of child neurology. 2015;30(2):153-159.

oo
15 13. Nordli DR. Epileptic encephalopathies in infants and children. Journal of clinical
16 neurophysiology : official publication of the American Electroencephalographic Society.

r
17 2012;29(5):420-424.
18
19
14.
-p
Ishikawa N, Kobayashi Y, Fujii Y, Tajima G, Kobayashi M. Ictal electroencephalography and
electromyography features in symptomatic infantile epileptic encephalopathy with late-onset
re
20 spasms. Neuropediatrics. 2014;45(1):36-41.
21 15. Frost JD, Hrachovy RA, Kellaway P, Zion T. Quantitative analysis and characterization of infantile
lP

22 spasms. Epilepsia. 1978;19(3):273-282.

23 16. Heiskala H. Community-based study of Lennox-Gastaut syndrome. Epilepsia. 1997;38(5):526-
na

24 531.
25 17. Alba GO-Gd, Valdez JM, Crespo FV. West syndrome evolving into the Lennox-Gastaut syndrome.
ur

26 Clinical EEG (electroencephalography). 1984;15(1):61-68.

27 18. Blume WT. Pathogenesis of Lennox-Gastaut syndrome: considerations and hypotheses.
Jo

28 Epileptic disorders : international epilepsy journal with videotape. 2001;3(4):183-196.

29 19. You SJ, Kim HD, Kang H-C. Factors influencing the evolution of West syndrome to Lennox-
30 Gastaut syndrome. Pediatric neurology. 2009;41(2):111-113.
31 20. Delmiro A, Rivera H, García-Silva MT, et al. Whole-exome sequencing identifies a variant of the
32 mitochondrial MT-ND1 gene associated with epileptic encephalopathy: west syndrome
33 evolving to Lennox-Gastaut syndrome. Human mutation. 2013;34(12):1623-1627.
34 21. Valdez JM. [Transional syndrome: from West to Lennox-Gastaut syndromes]. Medicina.
35 2018;78 Suppl 2:2-5.
36 22. Rantala H, Putkonen T. Occurrence, outcome, and prognostic factors of infantile spasms and
37 Lennox-Gastaut syndrome. Epilepsia. 1999;40(3):286-289.
38 23. Fusco L, Vigevano F. Ictal clinical electroencephalographic findings of spasms in West syndrome.
39 Epilepsia. 1993;34(4):671-678.
40 24. Fusco L, Chiron, C., Trevisano, M., Vigevano, F.,Chugani H., , Bureau M, Genton, P., Dravet, C.,
41 et al. Infantile Spasms. In: Epileptic Syndromes in Infancy, Childhood and Adolescence, 6 edicion.
42 John In eurotext2019:103–118.
43 25. Crumrine PK. Lennox-Gastaut syndrome. Journal of child neurology. 2002;17 Suppl 1:S70-75.
44 26. Selmer KK, Lund C, Brandal K, Undlien DE, Brodtkorb E. SCN1A mutation screening in adult

11
 1 patients with Lennox-Gastaut syndrome features. Epilepsy & behavior : E&B. 2009;16(3):555-
2 557.
3 27. Consortium EK, Project EPG, Allen AS, et al. De novo mutations in epileptic encephalopathies.
4 Nature. 2013;501(7466):217-221.
5 28. Terrone G, Bienvenu T, Germanaud D, et al. A case of Lennox-Gastaut syndrome in a patient
6 with FOXG1-related disorder. Epilepsia. 2014;55(11):e116-119.
7 29. Spiczak Sv, Helbig KL, Shinde DN, et al. DNM1 encephalopathy: A new disease of vesicle fission.
8 Neurology. 2017;89(4):385-394.
9 30. Lund C, Brodtkorb E, Øye A-M, Røsby O, Selmer KK. CHD2 mutations in Lennox-Gastaut
10 syndrome. Epilepsy & behavior : E&B. 2014;33:18-21.
11 31. Uematsu M, Sakamoto O, Sugawara N, et al. Novel mutations in five Japanese patients with 3-
12 methylcrotonyl-CoA carboxylase deficiency. Journal of human genetics. 2007;52(12):1040-
13 1043.

f
14 32. Kaminska A, Oguni H. Lennox-Gastaut syndrome and epilepsy with myoclonic-astatic seizures.

oo
15 Handbook of clinical neurology. 2013;111:641-652.
16 33. Intusoma U, Abbott DF, Masterton RAJ, et al. Tonic seizures of Lennox-Gastaut syndrome:

r
17 periictal single-photon emission computed tomography suggests a corticopontine network.
18
19 34.
Epilepsia. 2013;54(12):2151-2157.
-p
Benuzzi F, Mirandola L, Pugnaghi M, et al. Increased cortical BOLD signal anticipates
re
20 generalized spike and wave discharges in adolescents and adults with idiopathic generalized
21 epilepsies. Epilepsia. 2012;53(4):622-630.
lP

22 35. Inoue T, Kobayashi K, Matsumoto R, et al. Engagement of cortico-cortical and cortico-

23 subcortical networks in a patient with epileptic spasms: An integrated neurophysiological study.
na

24 Clinical neurophysiology : official journal of the International Federation of Clinical

25 Neurophysiology. 2020;131(9):2255-2264.
ur

26 36. Chiron C, Raynaud C, Mazière B, et al. Changes in regional cerebral blood flow during brain
27 maturation in children and adolescents. Journal of nuclear medicine : official publication,
Jo

28 Society of Nuclear Medicine. 1992;33(5):696-703.

29 37. Dulac O, Milh M, Holmes GL. Brain maturation and epilepsy. Handbook of clinical neurology.
30 2013;111:441-446.

31 Acknowledgment
32 We extend our gratitude to all the patients who participated in this study and to
33 Xuanwu Hospital for their substantial support.
34
35 Funding
36 This work was supported by National Key Research and Development Program of
37 China [Grant No. 2021YFC2501400], Special Fund of the Pediatric Medical
38 Coordinated Development Center of Beijing Hospitals Authority [Grant No.
39 XTYB201810], National Natural Science Foundation of China [Grant No. 81771398],
40 Beijing Key Clinical Specialty Excellence Project and National Support Provincial

12
 1 Major Disease Medical Services and Social Capability Enhancement Project.
2

3 Declaration of Competing Interest

4 The authors declared that they have no known competing financial interests or
5 personal relationships that could have appeared to influence the work reported in this
6 paper.
7

8 Abbreviations
9 LGS Lennox-Gastaut Syndrome; DEE developmental epileptic encephalopathy;

f
oo
10 ASM anti-seizure medication; EEG electroencephalogram; EMG electromyography;
11 MRI magnetic resonance imaging; FDR false discovery rate; IESS infantile epileptic

r
12 -p
spasms syndrome; HIE hypoxic ischemic encephalopathy; MCCD 3-Methylcrotonyl-
coenzyme A carboxylase deficiency; VNS vagus nerve stimulation
re
13
14
lP

15
16 Figure Legends
na

17 Figure1. The patterns of EEG with EMG in LGS patients. The ictal patterns with
18 clear EEG/EMG correspondence. A: myoclonic-tonic seizures – myoclonic activity
ur

19 (Arrow), tonic activity (bracket). B: Spams-tonic seizures - spasm activity (Arrow),

Jo

20 tonic activity (bracket). C: Tonic seizures(bracket).

21 Figure 2. Onset age of patients. The onset age of patients in the four groups. The
22 Group B demonstrates a statistically significant earlier onset age (P < 0.05).

13
 Table 1. General characteristics, clinical features in LGS patients in different
groups

Group A Group B Group Group D P value*

(n=20) (n=8) C(n=14) (n=4)

Gender Male, n(%) 13(65%) 6(75%) 12(85%) 2(50%) 0.42

Median
Onset 33.5(15.5- 12.0(7.2- 27.0(15.7- 42(27.0-
（IQR）， 0.02*
age(months) 91.5) 17.5) # 45.0) 84.0)
month

Age at EEG Median

92.4(24.0- 57.1(20. 57.64(18.0 102.8(66
recording(mon （IQR）， 0.054
180.0) 0-132.0) -133.0) .0-131.0)
ths) month

f
Diagnosed as

oo
Number, n(%) 1（5%） 6（75%）# 1（7%） 1（25%）# <0.01*
IESS before

r
HIE, n(%) 5（25%） 1（12.5%） 4（28.6%） 0 0.73

Structural
-p
re
abnormalities, 1（5%） 1（12.5%） 1（7.1%） 2（50%） 0.09
n(%)
lP

Tuberous
sclerosis, 0 0 1（7.1%） 0 0.56
na

n(%)
Etiologies Viral
ur

encephalitis, 2（10.0%） 1（12.5%） 1（7.1%） 0 1.00

n(%)
Jo

Cerebral
hemorrhage, 1（5.0%） 0 1（7.1%） 1（25.0%） 0.38
n(%)
Neurogenetic
metabolic 3(15.0%) 0 2(15.0%) 0 0.64
disease, n(%)
Unknow, n(%) 8(40.0%) 5(62.5%) 4(28.6%) 1(25.0%) 0.42
Delay n(%) 17(85.0%) 7(87.5%) 13(92.8%) 3(75%)
Motor mental
Regression 0.80
development 3(15%) 1(12.5%) 1(7.2%) 1(15%)
n(%)

The continuous variables were described by median and standard deviation with
range, and the categorical variables by number and percentage.
 *The data was analyzed by the chi-square test and Fisher’s exact test or Kruskal–
Wallis test were used for comparison of parametric and non-parametric data. P value
<0.05 was considered as statistically significant.

#The difference between any two groups were compared using multiple comparison.
FDR corrected q–value <0.05 was considered statistically significant.

f
oo
Table 2: Seizure types and Interictal EEG discharges in LGS patients in

r
different groups

Group A
-p
Group B Group Group D P value*
re
(n=20) (n=8) C(n=14) (n=4)
lP

Seizure type epileptic spasms 5(20%) 3(37.5%) 2(14.3%) 3(75%) 0.11

myoclonic 8(40%) 1(12.5%) 11(78.6%) 1(25%) 0.01*
na

atypical absent 8(40%) 2(25%) 4(28.6%) 1(25%) 0.82

ur

atonic 2(10%) 1(12.5%) 2(14.3%) 1(25%) 0.87

focal 4(20%) 0(0%) 2(14.3%) 1(25%) 0.88
Jo

Interictal
slow spike-wave
EEG 17(85%) 8(100%) 13(92.9%) 4(100%) 0.53
complex
discharges
spike rhythm 11(55.5%) 6(75%) 8(57.1) 3(75%) 0.71
focal discharges 10(50%) 3(37.5%) 6(42.9%) 2(50%) 0.93

The categorical variables by number and percentage.

*The data was analyzed by the chi-square test and Fisher’s exact test or Kruskal–
Wallis test were used for comparison of parametric and non-parametric data. P value
<0.05 was considered as statistically significant.
 Table 3: Treatment and outcomes in LGS patients in different groups

Group A Group B Group P value*

Group D (n=4)
(n=20) (n=8) C(n=14)

Follow-up, n(%) 17(85.0%) 6(75.0%) 12(85.7%) 3(75.0%) 0.79

Therapy Antiepileptic drugs, 17(85.0%) 6(75.0%) 12(85.7%) 3(75.0%)
Neuromodulation 1(5.0%) 2(25.0%) 1(7.1%) 1(25.0%)
0.34

f
ketogenic diet 0 1(12.5%) 1(7.1%) 0

oo
Seizure frequency
reduction more than 12(70.6%) 1(16.7%) 9(75%) 0(0%)

r
Therapy 50% -p
effects Seizure frequency 0.01*
re
reduction less than 5(29.4%) 5(83.3%) 3 (25%) 3(100.0%)
50%
lP

Motor Normal n(%) 10(58.8%) 0(0%) 6(50.0%) 0(0%)

mental
na

Delay n(%) 7(41.2%) 4(66.7%) 6(50.0%) 1(33.3%) 0.01*

develop
ment Regression n(%) 0(0%) 2(33.3%) 0(0%) 2(66.70%)
ur

The variables were described by the categorical variables by number and percentage.
*The data was analyzed by the chi-square test and Fisher’s exact test were used for
Jo

comparison of non-parametric data. P value <0.05 was considered as statistically

significant.
Jo
ur
na
lP
re
-p
ro
of
Jo
ur
na
lP
re
-p
ro
of
Acknowledgment
We would like to thank all patients participated in this study and the strong
support of Xuanwu Hospital.

Funding
This work was supported by National Key Research and Development Program
of China [Grant No. 2021YFC2501400], National Natural Science Foundation of
China [Grant No. 81771398], Beijing Key Clinical Specialty Excellence Project and
National Support Provincial Major Disease Medical Services and Social Capability

f
Enhancement Project.

r oo
Declaration of Competing Interest
-p
The authors declared that they have no known competing financial interests or
re
personal relationships that could have appeared to influence the work reported in this
lP

paper.
na
ur
Jo