Could prevention of infantile spasms have been possible in a historical cohort of 31 tuberous sclerosis patients?

(1)

UEF//eRepository

DSpace https://erepo.uef.fi

Rinnakkaistallenteet Terveystieteiden tiedekunta

2021

Could prevention of infantile spasms have been possible in a historical

cohort of 31 tuberous sclerosis patients?

Riikonen, Raili

Elsevier BV

Tieteelliset aikakauslehtiartikkelit

© 2021 The Author

CC BY http://creativecommons.org/licenses/by/4.0/

http://dx.doi.org/10.1016/j.ejpn.2021.10.010

https://erepo.uef.fi/handle/123456789/26922

Downloaded from University of Eastern Finland's eRepository

(2)

Could prevention of infantile spasms have been possible in a historical cohort of 31 tuberous sclerosis patients?

Raili Riikonen

University of Eastern Finland and Kuopio University Hospital, Kuopio, Finland, Postal Address: Yliopistonranta 1, FI-70110, Kuopio, Finland

a r t i c l e i n f o

Article history:

Received 24 May 2021 Received in revised form 7 October 2021

Accepted 23 October 2021

Keywords:

Infantile spasms Tuberous sclerosis Prevention Retrospective study

a b s t r a c t

Efforts to prevent epilepsy in infants with tuberous sclerosis complex (TSC) has been the focus of EPISTOP.

Purpose: The present study was carried out to evaluate whether prevention could have been realistic.

Methods:A retrospective analysis by hospital chart review of 31 patients with TSC and infantile spasms (practically all patients) admitted to two tertiary hospitals, Children's Hospital, University of Helsinki and Kuopio in 1980e2000.

Clinical history, early cognitive development, early clinical signs of TSC, clinical signs of suspicious seizures,ﬁrst seizures and EEG, response to adrenocorticotropic hormone (ACTH) therapy, EEG and brain imaging were evaluated.

Results:Early development prior the spasms was apparently normal in 25 (80%). Theﬁrst EEG ever performed for a child showed hypsarrhythmia in 16 (51%) or modiﬁed hypsarrhythmia in 10 (32%).

Treatment lag was short (0e4, mean 2 weeks) and the primary response to ACTH favorable in 19 (64%).

Etiological diagnostic workup of IS revealed TSC. In one single case (3%) the diagnosis of TSC could be made at birth due to a congenital cardiac rhabdomyoma. Three other rhabomyomas were diagnosed later.

In brain imaging, subependymal periventricular calciﬁcations or hypodense areas were seen in every patient at onset of IS. Other organ manifestations of TSC were retinal phakomas (6), polycystic kidneys (2), and renal angiolipomatosis (1).

Conclusions: Preventive treatment of epileptic discharges could have been possible in a single case of neonatal rhabdomyoma suggesting that preventive treatment is challenging in everyday practice. The main obstacle is the delay of TSC diagnosis.

1. Introduction

Prevention of epilepsy has long been an important goal of research. Earlier clinical trials aimed at preventing epilepsy after brain injury or stroke have failed [1].

Early prediction of infantile spasms (IS) and subsequent intervention would be of signiﬁcant prognostic importance.

Walther et al., 1984 [2] tried by toﬁnd predictive signs of IS in early infancy among high-risk patients using neonatal EEGs and polygraphic tracings. Several studies from Japan report predicting the development of IS based on epileptiform discharges of serial EEGs in high-risk infants including infants with hypoxic-ischemic encephalopathy [3], infants with symptomatic IS [4], and preterm

babies with periventricular leukomalacia [5,6]. In the series by Yoshinagi et al. [6] development of IS could be prevented success- fully by valproate in two infants, but in the majority of infants it was not possible. Epileptic activity was deﬁned as“deﬁnite sharp waves and spikes”.

The understanding and treatment of TSC have advanced signiﬁcantly in the last two decades. Clinical electroencephalographic biomarkers for epilepsy in infants with tuberous sclerosis (TSC) were prospectively studied inﬁve cases by Domanska-Pakiela et al. [7] and by Wu et al. [8] in a multicenter study of serial EEGs of 40 infants.

Prevention of IS in patients with TSC has recently been a hot topic in US and Europe. Two projects PREVENT [9] and EPISTOP [10] (ClinicalTrials.gov Identiﬁer: NCT02849457 (USA) and NCT02098759) have been registered. The latter has been recently published as EPISTOP (Long-Term, Prospective Study Evaluating E-mail address:rriikone@uef.ﬁ.

Contents lists available atScienceDirect

European Journal of Paediatric Neurology

https://doi.org/10.1016/j.ejpn.2021.10.010

creativecommons.org/licenses/by/4.0/).

(3)

Clinical and Molecular Biomarkers of Epileptogenesis in a Genetic Model of EpilepsyeTuberous Sclerosis Complex) study [10]. It shows that prophylactic vigabatrin (VGB) treatment can alter the outcome of TSC patients.

The children of that trial were patients in hospitals with special services to TSC. We asked, based on our historical experience of 31 infants with TSC, whether preventive VGB treatment could have been realistic. Our cohort may serve as an example of patients in an average tertiary care hospital.

2. Material and methods

Thirty-one patients (practically all patients with TSC) were admitted to two tertiary care hospitals, Children's Hospital, Uni- versity of Helsinki and Kuopio University Hospital in 1980e2000 at a mean age of 6 months for treatment of IS. IS was defined by IS clusters as the first or main seizure type with either hypsarrhythmia or modified hypsarrhythmia (focality, multifocality and asymmetry included) on interictal EEG). Careful etiological investigations revealed the diagnosis of TSC, and the patients fulfilled the clinical criteria for definite TSC [11]. Both hospitals have multidisciplinary approach for the patients. Early outcome of the patients up to 12 months of age were described.

Hospital records were reviewed for signs of suspicious seizures, early cognitive outcome, onset of IS, treatment, and manifestations of TSC.

Patients were admitted urgently to EEG registration if IS were suspected. The treatment lag was usually short (0e4, mean 2 weeks) in 20 patients, but exceptionally long in 4 patients (3 months, 3 months, 11 months and 14 months, respectively). In one patient treatment lag was not known. The reason for delay in the treatment were urinary infection (1), Salmonella infection(1), delay in transport from one local hospital to the tertiary care hospital (1) and severe developmental mental delay and risk/beneﬁt consideration (1). The conventional antiepileptic drugs were used before a decision of ACTH treatment. In patients treated with VGB the lag was a few days.

The EEG included recordings taken while always awake, awaking and sleeping.

Adrenocorticotropic hormone (ACTH) therapy was given to 25 patients, VGB to three, and three received polytherapy. At that time, ACTH was given according to existing guidelines to all IS patients.

Our schedule was later changed to recommend VGB treatment in TSC because it was shown to be a drug of choice [12].

The etiological work-up included clinical survey, ophthalmo- logical examinations, an if necessary cardiac echography and ul- trasound of the kidneys, EEG and brain imaging (CT/NMRI), the latter always carried out before starting ACTH because of transient brain shrinkage caused by ACTH to avoid the false diagnosis of brain atrophy [13].

All infants were seen by child neurologists at admission to the hospital for treatment of IS. The developmental skills were evaluated by child neurologists and by collecting data from child welfare clinics and the caregivers. Standardized tests were usually not done before starting treatment because during the hypsarrhythmic state their value is in our opinion questionable. Psychometric tests from the Bayley Scales of Infant Development third edition (BSID-III) were carried out at the age of 24 months and (Cattel, Vineland, Bayley, Bender, Terman-Merrill or Wechsler Intelligence Scale for Children (WISC) after a few years.

3. Ethics

Approval from the Ethics Committee of the Children's Hospital University of Helsinki and Kuopio was obtained for this retrospective data analysis.

4. Results

Clinical data of 31 patients with TSC are shown inTable 1. The diagnosis of TSC was based on careful investigations at admission.

There have not been any known familial cases before treatment.

However, there were four familial cases but this was detected only after the proband (infantile spasms). In one family, in addition to the proband, the father was affected and the diagnosis was veriﬁed by sequencing of the TSC2 gene. In the second family the brother of the proband, in the third family the twin sister and father and in the fourth family the father, the father's father were affected having white spots in the skin, but because they lacked other symptoms, they were not genetically tested.

The first manifestations of TSC were seizures. The first EEG showed hypsarrhythmia in 16 (51%) or modified hypsarrhythmia in 10 (32%) and diagnosis of IS was made and treatment started.

Prevention of IS could not have been possible. There was one newborn with focal spikes at age of one month with congenital cardiac rhabdomyoma. This patient could have been a candidate for preventive treatment. Three others had early subtle symptoms such as jerks or eye blinking retrospectively reported by the parents but they were ignored until clinical and electro-clinical focal seizures started. Two of them had cardiac rhabdomyomas not identiﬁed as newborns but only later, after 16 weeks, past the critical time for prevention (see later). Cardiac rhabdomyomas were present in 13%

of the infants. Nine of 31 infants had onset of seizures before 16 weeks.

Hypomelanotic maculae were sought with Wood's lamp but this was not done earlier than at admission for treatment of IS.

In brain imaging, subependymal periventricular calciﬁcations or hypodense areas were seen in every patient at the onset of IS. Other organ manifestations of TSC were retinal phakomas (6), polycystic kidneys (2), renal angiolipomatosis (1) and cardiac rhabdomyomas (4).

Topographical correlation of electroclinical and neuroradiological focalﬁndings were observed inﬁve cases, suggesting possible epileptogenicity of certain tubers. The temporal lobe was the most commonly affected localization of abnormalities (19/30) in 63%.

5. Discussion

To prevent the development of epileptic encephalopathy in infants with IS, a prompt treatment response is crucial. The risk of mental retardation increases if hypsarrhythmia continues more than three weeks [14]. Epilepsy develops in 70e80% of children with TSC and is often refractory to medication [15]. Scientists and clinicians share a great interest in discovering methods to prevent seizures. EPISTOP [10] was a novel prospective, multicenter, randomized study. Ninety-four patients were included in the study of which 54 were eligible. Infants without seizure history were followed with monthly video-telemetry, and received VGB either as conventional therapy, or preventively when epileptiform EEG activity was detected before electrical or clinical seizures. The randomized groups included preventive treatment (N¼13) and conventional treatment (14); and open-label trial: preventive (12) and conventional treatments (15).

The effect of preventive treatment could be shown at 24 months of age: Early intervention with VGB delayed the onset of epilepsy, reduced the risk of seizures, and prevented IS [10].

Epileptic activity was deﬁned as unifocal discharges occupying 10% or more of the recording time, multifocal discharges (involving 2 areas of the brain), or widespread/generalized epileptiform activity, including hypsarrhythmia. Electrographic seizures were deﬁned as ictal EEG activity with no clinical correlate on video [10].

Ictal abnormalities on EEG recordings predate clinical seizures [8].

R. Riikonen European Journal of Paediatric Neurology 35 (2021) 153e157

154

(4)

Very recently, early epileptiform EEG activity of the patients in EPISTOP were more exactly characterized [16]. Early appearance of epileptiform discharges was associated with worse developmental outcome. Infants presenting with multifocal interictal epileptic discharges beneﬁted more from preventive VGB treatment in delaying seizure onset than infants with focal EEGﬁndings [16].

Also previously, Gataullina et al. [17] has shown that in 3/15 infants diagnosed prenatally with TSC and showing subclinical discharges (sharp waves) before any seizures, VGB prevented the development of seizures until follow-up at 18e36 months. How- ever, eight patients treated after seizure onset developed IS or focal seizures that were pharmacoresistant inﬁve.

5.1. Would prevention be realistic in everyday practice?

We collected data of 31 patients with TSC and IS to evaluate whether prevention would have been possible in this cohort which represents a quite typical tertiary hospital setting.

For possibilities of epilepsy prevention an early diagnosis of TSC before clinical and electrographic seizures is needed, but this pre- sents a big challenge. The symptoms and signs of TSC are commonly missed [18]. The most common symptoms and signs of TSC are the seizure [18] as was the cases also in this cohort. Signs of focal seizures were reported but ignored, without EEG registration or video- telemetry. Furthermore, the hypomelanotic maculae of the skin, typical for TSC were not noted because the infants seemed to be apparently well and normal in development. At admission to hospital the patients were speculated to have IS without known etiology (cryptogenic, idiopathic cases). Only careful etiological investigations ﬁrst revealed the diagnosis of TSC. The response to ACTH was nearly as high as in patients with unknown etiology, 73 vs 80%, respectively [19]. This is in accordance with results of Elterman et al. [20]. The

high response rate by ACTH is not surprising due to the possible inflammatory activity of cortical tubers [21], and the anti- inflammatory action of ACTH. VGB is used in our country because VGB is effective [12] and can be given for prolonged period but ACTH only for a limited time. The patients with TSC have frequent relapses [20]. VGB is the drug of choice for TSC [12] although one third of the patients will have visualfield defects [22].

5.2. Focality of EEG and neuroradiology

Given the focal pathology it is natural that most seizures were of focal onset and with or without rapid secondary generalization. In hindsight, some early signs reported by the parents might have been subclinical partial seizures. After treatment the focality was often observed in EEG and brain imaging. Focality in EEG has been observed in series where patients were followed after hypsarrhythmia or modiﬁed hypsarrhythmia for 24 months or longer [19,23]. In the largest international cohort of TSC patients with an age range of 0e30 years, epilepsy was reported in 83.6% of patients (1885/2216) at baseline; 38.9% presented IS and 67.5% focal seizures. The mean age at diagnosis of focal seizures was 2.7 years, range 1e66 years whereas mean age at diagnosis of spasms was 0.4 years [23].

In another retrospective chart review of TSC patients seizure onset occurred within theﬁrst 4 weeks of life in 15 patients (6.6%), before 6 months of age in 106 patients (45.8%), by the end of the ﬁrst year of life in 146 patients (63%), and before 3 years of age in 231 patients (88%) [24]. Focal impaired awareness seizures (complex partial seizures) were the dominant seizure type in the follow- up of all 291 TSC patients [24].

In the present study, although patients were admitted to two University Hospitals having special expertise in pediatric Table 1

Data on epilepsy parameters in the study population.

Characteristics of tuberous sclerosis patients Number of patients (N¼31) Percentage

Normal early development 25 81

Male 17 55

Female 14 45

Suspicion of signs of seizures before IS 3

Cardiac rhabdomyomas diagnosed before seizure onset 1 3

Family history of tuberous sclerosis # 0 0

Age at IS onset (mean) 6.1 months

Early IS (<4 months) 9 29

First EEG

hypsarrhythmia (HA) 16 52

multifocal spike activity 4

focalþHA 6

slow waves and spikes 3

Good response to ACTH 16 of 25 64

EEG prior and/or during IS treatment

Focality in EEG prior or during ACTH 14 56

Temporal 6

frontotemporal 5

parietotemporal 1

occipital 1

frontal 4

Laterality in EEG

right hemisphere 12

left hemisphere 10

Neuroradiological localization of tubers

temporal 6

frontotemporal or frontal 5

occipitotemporal or occipital 4

Laterality in neuroradiology

right hemisphere 11

left hemisphere 6

Correlation between EEG and neuroradiologicalﬁndings 5

Abbrevations: IS¼infantile spasms, HA¼hypsarrhythmia, ACTH¼adrenocorticotropic hormone, # known before the treatment. See the Text Page 5,3rd prgh

(5)

neurology, the diagnosis of TSC came only after seizure onset, except in one case (3%). A similar delay of diagnosis of TSC may exist even today, if congenital cardiac rhabdomyomas are not detected, or there is no family history of TSC. However, the majority of cases arede novomutations [25].

Prenatal diagnosis of TSC was done in 5.9% in the largest case series of TS with 2093 patients suggesting that prenatal diagnosis was rare [26].

5.3. Differences between TSC clinics and everyday practice

TSC clinics have 1) a much larger numbers of TSC patients, 2) education of the parents about recognition of subclinical spasms (e.g. using videos), 3) increased vigilance of the clinicians for epileptic signs and hypopigmented maculae, 4) serial EEG/video- monitoring 5) prenatal diagnosis of cardiac rhabdomyoma or brain tumors using MRI, 6) mutational analysis of TSC genes.

Early signs of TSC were presented in a study which included only children in whom theﬁrst sign of TSC was cardiac tumors [27]. In one study including both children <2 years and adults up to 23 years old, rhabdomyomas occurred in 74/154 (48%) patients [28], and in TSC registry (patients aged 0e69 years 717/2093 (34.3%) showed cardiac tumors [26].

The incidence was dependent on the patient age. The tumors are usually clinically silent and show regression in 68% [28].

The diagnosis of TSC was regarded as feasible in the TSC center of Warsaw, but this series included only patients with cardiac tumors [27]. In the vast majority, 82/100 (82%) the diagnosis was established before the end of the 16 weeks of age, termed as“early diagnosis”. The usual time of seizure onset are theﬁrst 16 weeks and represent also a critical window for potential disease- modifying antiepileptogenic treatment. Prenatal MRI of the brain is another possibility for prenatal TSC diagnosis.“Early diagnosis of TSC could be made without cardiac tumors although it might be more challenging”[28]. Skin examinations showed hypomelanotic maculae in 35/75 infants studied before<16 weeks of age. Davis et al. [29] in a prospective multicenter US study emphasize that most patients with TSC can be identiﬁed with echocardiography for cardiac rhabdomyomas and skin examinations for hypomelanotic maculae; both are noninvasive and do not require sedation.

In our study most patients 22/31 (70%) with TSC were diagnosed

“late”i.e. after 16 weeks, and would not have been suitable for preventive treatment.

5.4. Cognition

The ultimate goal of treatment of IS will be good cognitive outcome. Since epilepsy is the main contributor to intellectual disability in TSC the question remains why cognitive outcome was not better in the prevention group of EPISTOP and in the preventive study by Humphrey et al. [30]. In an open-label pilot study which was the basis of EPISTOP study, cognitive outcome was better in the preventive treatment group (N¼14) [31]. However, the study can be criticized for selection bias because the preventive group included four neurologically “normal” infants who had no seizures nor abnormal EEG. Congenital rhabdomyomas were present in 12/14 cases, an exceptional proportion, which allowed an early diagnosis of TSC [31]. However, there are great differences between the Polish 2011 open-label study [31] and the prospective EPISTOP study. The reason for lacking effect of preventive VGB treatment on cognition in the EPISTOP study might be explained by the following: 1) It is possible that the seizures are merely a marker of something else that is determining developmental outcome but are not actually the cause of developmental stagnation/regression. TSC manifests itself during fetal development prenatally and in the brain

permanent structural malformations and metabolic changes, such as active expression of nerve growth factors [32], take place. 2) Those randomized to the conventional treatment arm received intensive follow up and were treated almost immediately after any kind clinical seizures occurred (i.e. there was a very short lead-time to treatment).

A controlled randomized study of TSC showed that the mean IQ dropped from 92 prior onset of IS to 62 after exposure to IS for more than one month [30]. In our study the apparently normal children were later moderately to severely delayed in most cases [19].

There are several studies showing that the lead time to treatment and response, both, are important for cognitive outcome in the group of patients with unknown etiology and previous normal development [14,33e36]. For patients with TSC it seems also to be important [37]. I sincerely hope, but it is not proven, that prophylactic VGB treatment gives a better cognitive outcome or lack of autistic behavior when epileptic discharges before sentinel seizures have started.

In summary, an early diagnosis of TSC is crucial for prevention.

5.5. Strengths of the present study

This study deals with the careful diagnosis of TSC patients at onset of IS. Data on infants before, during, and shortly after treatment (with ACTH) up to 12 months of age is presented, in contrast to most studies dealing with older children and adults. Although the retrospective nature of the work involves signiﬁcant limitations and ignores changes in diagnostic and therapeutic approaches in TSC over time, this study allows to address an important topic in the ﬁeld of epilepsy research.

5.6. Limitations of the present study

This study had a limited number of patients due to the small size of the Finnish population (5 million) [38]. Limitations of our retrospective study are due to the fact that data is dealing with patients diagnosed 20e30 years ago and many advances have been made in the diagnostics of TSC since then. In addition, awareness of TSC and awareness of subtle signs of epilepsy by the medical community, availability of high quality EEG in neonates and infants, and their interpretation (recognition of subclinical and electroclinical spasms), serial EEGs and video monitoring, availability of prenatal MRI for diagnostics of fetal rhabdomyomas, and increased screening of variants of TSC genes may have improved the pre- requisites for preventive treatment of TSC such as suggested by the EPISTOP study. The seizures were the presenting symptom of TSC in our study. In the EPISTOP trials, the large majority showed EEG abnormalities before onset of seizures, which is the ideal port for preventive treatment.

6. Conclusion

It is a great challenge to ﬁnd early epileptic activity before clinical and electroclinical seizure onset, and treat IS preventively to reverse disease progression in TSC. In our cohort it could have been possible in only 3% of the patients. The seizures were the presenting symptom of TSC. Pediatricians and pediatric neurologists should be educated to pay attention to the signs of TSC. They should be aware of the possibility of IS. Once TSC is recognized, frequent monitoring of EEG, as suggested by EPISTOP, and consequently early treatment may predict a more favorable outcome. Advances in prenatal, perinatal and postnatal care may hopefully increase the number of infants with TSC diagnosed prior to manifestation of seizures. Searching for TSC signs key for early diagnosis.

R. Riikonen European Journal of Paediatric Neurology 35 (2021) 153e157

156

(6)

Our retrospective analysis of TSC patients in typical tertiary care settings sheds light on the current discussion on the possibilities and the challenges of epilepsy prevention.

Ethical publication statement

I conﬁrm that I have read the Journal's position on issues involved in ethical publication and afﬁrm that this report is consistent with those guidelines.

Funding disclosure

This research did not receive any speciﬁc grant from funding agencies in the public, commercial, or not-for-proﬁt sectors.

Declaration of competing interest

The authors declare that they have no known competing ﬁnancial interests or with other people or organizations that could have appeared to inﬂuence the work reported in this paper.

Acknowledgements

I thank Dr. Jaana L€ahdetie, for help in writing this manuscript.

References

[1] K. Thompson, B. Pohlmann-Eden, L.A. Campbell, H. Abel, Pharmacological treatments for preventing epilepsy following traumatic head injury, Cochrane Database Syst. Rev. 8 (2015) CD009900.

[2] B. Walther, T. Schmitt, B. Reitter, Identiﬁcation of infants at risk for infantile spasms by neonatal polygraphy, Brain Dev. 9 (1987) 377e390.

[3] K. Watanabe, T. Takeuchi, S. Hakamada, F. Hayakawa, Neurophysiological and neuroradiological features proceeding infantile spasms, Brain Dev. 9 (1987) 391e398.

[4] F. Endoh, H. Yoshinaga, K. Kobayashi, Y. Ohtsuka, Electroencephalographic changes before the onset of symptomatic West syndrome, Brain Dev. 29 (2007) 630e638.

[5] M. Suzuki, A. Okumura, K. Watanabe, T. Negoro, F. Hayakawa, T. Kato, et al., The predictive value of electroencephalogram during early infancy for later development of west syndrome in infants with cystic periventricular leukomalacia, Epilepsia 44 (2003) 443e446.

[6] H. Yoshinaga, F. Endo, K. Kikumoto, T. Inoue, M. Oka, Y. Ohtsuka, Epilepsy development in infancy with epileptic discharges, Brain Dev. 29 (2007) 17e23.

[7] D. Domanska-Pakieła, M. Kaczorowska, Kotulska K. Jurkiewicz, D. Dunin- Wa˛sowicz, S. Jozwiak, EEG abnormalities preceding the epilepsy onset in tuberous sclerosis complex patientsdprospective study of 5 patients, Eur. J.

Paediatr. Neurol. 18 (2014) 458e468.

[8] J.Y. Wu, M. Goyal, J.M. Peters, D. Krueger, M. Sahin, H. Northrup, et al., Scalp EEG spikes predict impending epilepsy in TSC infants: a longitudinal obser- vational study, Epilepsia 60 (2019) 2428e2436.

[9] PREVENT. Preventing Epilepsy Using Vigabatrin in Infants with Tuberous Sclerosis Complex. ClinicalTrials.gov ID: NCT02849457. NIH U.S. National Li- brary of Medicine.

[10] K. Kotulska, D. Kwiatkowski, P. Curatolo, B. Wesche, K. Riney, F. Jansen, et al., Prevention of epilepsy in infants with tuberous sclerosis complex in the EPISTOP trial, Ann. Neurol. 89 (2021) 3004e3014,https://doi.org/10.1002/

ana.25956.

[11] H. Northrup, D.A. Krueger, Tuberous sclerosis complex diagnostic criteria update: recommendations of the 2012 international tuberous sclerosis complex consensus conference, Pediatr. Neurol. 49 (2013) 243e254.

[12] E.C. Hancock, J.P. Osborne, S.W. Edwards, Treatment of infantile spasms Cochrane, Database Syst Rev 6 (2013) CD001770.

[13] Y. Konishi, M. Yasujima, M. Kuriyama, K. Konishi, K. Hayakawa, Y. Fujii, et al., Magnetic resonance imaging in infantile spasms: effects of hormonal therapy, Epilepsia 33 (2) (1992) 304e309, https://doi.org/10.1111/j.1528- 1157.1992.tb02320.x.

[14] Z. Rener Primec, J. Stare, D. Neubauer, The risk of lower mental outcome in infantile spasms increases after three weeks of hypsarrhythmia duration, Epilepsia 47 (2007) 2202e2205, https://doi.org/10.1111/j.1528- 1167.2006.00888.x.

[15] P. Curatolo, R. Bombardieri, S. Jowiack, Tuberous sclerosis, Lancet 37 (23)

(2008) 657e688,https://doi.org/10.1016/S0140-6736(08)61279-9, submitted for publication.

[16] J. De Ridder, B. Verhelle, J. Vervisch, K. Lemmens, K. Kotulska, R. Moavero, et al., The EPISTOP consortium Early epileptoform EEG activity in infants with tuberous sclerosis complex predicts epilepsy and neurodevelopmental outcomes, Epilepsia 62 (2021) 1208e1219,https://doi.org/10.1111/epi.16892.

[17] S. Gataullina, E. Lemaire, F. Wendling, A. Kaminska, F. Watrin, A. Riquet, et al., Epilepsy in young Tsc1þ/- mice exhibits age-dependent expression that mimics that of human tuberous sclerosis complex, Epilepsia 7 (2016) 648e669,https://doi.org/10.1111/epi.135.

[18] B. Staley, E. Vail, E. Thiele, Tuberous sclerosis complex: diagnostic challenges, presenting symptoms, and commonly missed signs, Pediatrics 17 (2011) 117e215,https://doi.org/10.1542/peds.2010-0192.

[19] R. Riikonen, O. Simell, Tuberous sclerosis and infantile spams, Dev. Med. Child Neurol. 32 (1990) 203e209.

[20] R. Elterman, W. Shield, S. Collins, Vigabatrin is effective in multiple etiologies of infantile spasms, Epilepsia 47 (s4) (2006) 179 ([Abstract]).

[21] K. Boer, F. Jansen, M. Nellist, M. Redeker, A.M.W. van den Ovelan, W. Spielt, et al., Inﬂammatory processes in cortical tubers and subependymal giant cell tumors of tuberous sclerosis complex, Epilepsy Res. 78 (2008) 7e21.

[22] R. Riikonen, Rener-Primec, L. Carmant, M. Dorofeeva, K. Hollody, I. Szabo, et al., Does vigabatrin treatment for infantile spasms cause visualﬁeld defects?

An international multicentre study, Dev. Med. Child Neurol. 57 (2015) 60e67, https://doi.org/10.1111/dmcn.12573.

[23] R. Nabbout, E. Belousova, M.P. Benedik, T. Carter, V. Cottin, P. Curatolo, et al., Epilepsy in tuberous sclerosis complex:ﬁndings from Thompson K, Pohl- mann-Eden B, Campbell LA, Abel H. Pharmacological treatments for preventing epilepsy following traumatic head injury, Cochrane Database Syst.

Rev. 8 (2015) CD009900.

[24] C.J. Chu-Shore, P. Major, S. Camposano, D. Muzykewicz, E. Thiele, The natural history of epilepsy in tuberous sclerosis complex, Epilepsia 51 (2010) 1236e1241,https://doi.org/10.1111/j.1528-1167.

[25] J. Liu, L. Tong, S. Song, Y. Niu, J. Li, J. Zhang, et al., Novel andde novomutations in pediatric refractory epilepsy, Mol. Brain 11 (1) (2018) 48e66,https://

doi.org/10.1186/s13041-018-0392-5.

[26] J. Kingswood, G. d’Augeres, E. Belousova, J. Ferreira, T. Carter, R. Castellana, et al., And on behalf of TOSCA consortium and TOSCA investigators. Tuberous Sclerosis registry to increase disease Awareness (TOSCA)ebaseline data on 2093 patients, Orphanet J. Rare Dis. 12 (2017) 2, https://doi.org/10.1186/

s13023-016-0553-5. Published online 2017 Jan 5.

[27] M. Slowinska, S. Jozwiak, A. Peron, J. Borkowska, D. Chmielewski, K. Sadowski, et al., Early diagnosis of tuberous sclerosis complex: a race against time. How to make the diagnosis before seizures? Orphanet J. Rare Dis. 13 (2018) 25, https://doi.org/10.1186/s1323-018-0764-z.

[28] S. Jozwiak, K. Kotulska, Kaspryk-Obara, D. Domanska- Pakiela, M. Tomyn- Drabik, P. Roberts, et al., Clinical and genotype studies of cardiac tumors in 14 patients with tuberous sclerosis complex, Pediatrics 118 (2006) e1146e1151, https://doi.org/10.1542/peds.2006-0504.

[29] P. Davis, R. Filip-Dhima, G. Sideridis, J. Peters, K. Sing Au, H. Northrup, et al., Presentation and diagnosis of tuberous sclerosis complex in infants, Pediatrics 140 (2017), e20164040,https://doi.org/10.1542/peds.2016-4040.

[30] A. Humphrey, C. MacLean, G. Ploubidis, Y. Granader, M. Clifford, M. Haslop, et al., Intellectual development before and after the onset of infantile spasms: a controlled prospective longitudinal study in tuberous sclerosis, Epilepsia 55 (2014) 108e116.

[31] S. Jozwiak, K. Kotulska, D. Domanska-Pakila, B. Lojszczyk, M. Syczewska, D. Chmielewski, et al., Antiepileptic treatment before the onset of seizures reduces epilepsy severity and risk of mental retardation in infants with tuberous sclerosis complex, Eur J Pediatr Neurol 15 (2011) 424e431.

[32] R. Riikonen, H. Kokki, CSF nerve growth factor (b-NGF) is increased but CSF insulin-like growth factor-(IGF-1) is normal in children with tuberous sclerosis complex and infantile spasms, Eur J Pediatr Neurol (2018) 191e196, https://doi.org/10.1016/ejpn.2018.11.001.

[33] R. Riikonen, Long-term outcome of patients with West syndrome Brain, Dev 23 (2001) 683e687.

[34] S. Kivity, P. Lerman, B. Ariel, Y. Danziger, M. Mimouni, S. Shinnar, Long-term cognitive outcomes of a cohort of children with cryptogenic infantile spasms treated with high-dose adrenocorticotropic hormone, Epilepsia 45 (2004) 255e262.

[35] F. O'Callaghan, A. Lux, K. Darke, S. Edwards, E. Hancock, A. Johnson, et al., The effect of lead time to treatment and of age of onset on developmental outcome at 4 years in infantile spasms: evi- dence from the United Kingdom Infantile Spasms Study, Epilepsia 52 (2011) 1359e1364.

[36] S. Auvin, A. Hartman, B. Desnous, A.-C. Moreau, C. Alberti, C. Delanoe, Diag- nosis delay in West syndrome. Misdiagnosis and consequences, Eur. J. Pediatr.

171 (2012) 1695e1701.

[37] R. Cusmai, R. Moavero, R. Bombardieri, F. Vigevano, P. Curatolo, Long-term meurological outcome in children with early-onset epilepsy associated with tuberous sclerosis, Epilepsy Behav. 22 (2011) 735e739.

[38] R. Riikonen, Decreasing perinatal mortality: unchanged infantile spasm morbidity, Dev. Med. Child Neurol. 37 (1995) 232e238.