Ketogenic Diet for the Management of Epilepsy

Associated with Tuberous Sclerosis Complex in Children

1 2 3 4 4 4
Soyoung Park , Eun Joo Lee , Soyong Eom , Hoon-Chul Kang , Joon Soo Lee , Heung Dong Kim
1
Department of Pediatrics, Soonchunhyang University Bucheon Hospital, Soonchunhyang University College of
2
Medicine, Bucheon; Division of Dietetics, Severance Children's Hospital, Yonsei University College of Medicine,
3 4
Original Article Seoul; Epilepsy Research Institute, Yonsei University College of Medicine, Seoul; Division of Pediatric Neurology,
Journal of Epilepsy Research Epilepsy Research Institute, Severance Children's Hospital, Department of Pediatrics, Yonsei University College of
pISSN 2233-6249 / eISSN 2233-6257 Medicine, Seoul, Korea

Background and Purpose: In the present study, we reviewed the outcome of ketogenic diet (KD) use for the
management of epilepsy in children with tuberous sclerosis complex (TSC).
Methods: A total of 12 children with intractable epilepsy associated with TSC who were treated with KD at
our hospital between March 1, 2008 and February 28, 2015 were retrospectively enrolled.
Results: The mean age at the time of KD initiation was 73.1 ± 38.0 months. Patients were medically
refractory to a mean of 4.8 ± 1.7 antiepileptic drugs. Nine patients (75.0%) had a history of infantile spasms.
Received April 27, 2017
Accepted May 17, 2017 At 3 months after KD initiation, 10 patients (83.3%) had > 50% seizure reduction. Moreover, 7 patients
Corresponding author: Heung Dong Kim (58.3%) exhibited qualitative improvements in cognition and behavior after KD initiation, as reported by
Division of Pediatric Neurology, Epilepsy caregivers/parents. The mean duration of dietary therapy was 14.8 ± 12.8 months. Half of the patients in this
Research Institute, Severance Children's
Hospital, Department of Pediatrics, Yonsei study eventually underwent epilepsy surgery due to persistent seizures or seizure relapse.
University College of Medicine, 50-1 Conclusion: KD is an important non-pharmacological treatment option for patients with intractable epilepsy
Yonsei-ro, Seodaemun-gu, Seoul 03722,
associated with TSC. KD may improve cognition and behavior in addition to reducing seizure frequency.
Korea
Tel. +82-2-2228-2061 (2017;7:45-49)
Fax. +82-2-393-9118
E-mail; hdkimmd@yuhs.ac Key words: Ketogenic diet, Epilepsy, Tuberous sclerosis complex, Children

Introduction Several treatment options are available to optimize neuro-

developmental function and manage epilepsy in patients with TSC,
Tuberous sclerosis complex (TSC) is an inherited multisystem dis- including anti-epileptic drugs (AEDs), ketogenic diet (KD), epilepsy
order that is typically caused by mutations in TSC1 or TSC2, affecting surgery (resective surgery or corpus callosotomy), vagus nerve stim-
1 in every 6,000 or 10,000 persons, respectively.1-3 TSC is charac- ulation, and mammalian target of rapamycin (mTOR) inhibitor
terized by pleomorphic features involving several organ systems and therapy.6,9,12,14,15 In this study, we reviewed use of a KD in children
hallmark central nervous system (CNS) findings including cortical tu- with TSC and drug-resistant epilepsy and evaluated its effects on
bers, subependymal nodules (SEN), and subependymal giant cell as- seizure burden and cognitive as well as behavioral outcomes.
trocytoma (SEGA).1,2,4,5 Neurologic involvement in TSC is associated
with disease-related disability such as epilepsy, behavioral disorders, Methods
and cognitive deficits,4‐8 with epilepsy being the most common, re-
ported in 75% to 90% of patients.6,9 Moreover, 30% to 50% of pa- We performed a retrospective chart review of all patients seen for
tients with TSC are diagnosed with infantile spasms (IS), an epileptic TSC at the pediatric neurology department of Severance Children’s
encephalopathy syndrome, in their 1st year of life.10‐13 TSC-asso- Hospital between March 1, 2008 and February 28, 2015. Patients
ciated neuropsychiatric disorder is another major pathological fea- were included if they had clinically definite TSC, had been on a KD for
ture in patients, which affects clinical outcome as well as quality of > 3 months, had seizures intractable to at least 2 AEDs, and were
life.5,7 available for formal neuropsychological (NP) evaluation and parental

This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/3.0/)
which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
46 Journal of Epilepsy Research Vol. 7, No. 1, 2017

global assessment for diet outcome. We excluded patients who had The institutional review board of the Severance Hospital approved
been on the low glycemic index diet, those with poor compliance for this study.
KD maintenance < 3 months, and those whose parents had difficulty
counting seizures to determine seizure frequency. Finally, 12 patients Results
were enrolled.
A KD (3:1 or 4:1 ratio of fat to carbohydrate and protein) or a Patient demographic and clinical characteristics
modified Atkin’s diet was started without fasting. Patient data in- Of 156 children with TSC who were seen at our hospital during the
cluding gender, age at last follow up, brain magnetic resonance study period, 133 (85.3%) were treated for epilepsy with multiple
imaging (MRI) findings, age at seizure onset, number of AEDs used, treatment modalities such as AEDs, KD, and surgery. KD was initiated
history of infantile spasms, developmental status, epilepsy classi- in 18 patients. 6 patients were excluded due to loss to follow up or
fication at KD initiation, age at KD initiation, lead-time to KD ini- early discontinuation of the KD within 3 months. Ultimately, 12 of
tiation, peak effect of KD, duration of KD, seizure outcome after KD, these patients were included in our analysis.
and parental global assessment of overall KD outcome were col- The 12 patients enrolled in this study were 5 boys and 7 girls with
lected by chart review. Minor adjustments of AEDs were done during a mean age of 73.1 ± 38.0 months (range, 26.4-144.2 months). All
KD maintenance in some cases. Seizure outcomes were assessed at 3 12 patients of had cortical tubers and SEN, and 3 patients (25.0%)
months, 6 months, and 12 months after starting the KD, and were had SEGA on brain MRI. Ten patients (83.3%) had experienced seiz-
classified as follows: seizure free, > 90% reduction, 50% to 90% re- ures in the 1st year of life and the mean age at seizure onset was 9.6
duction, and < 50% reduction in seizures. The parental global as- ± 12.5 months (range, 1.0-47.4 months). Patients were treated with
sessment of overall KD outcome regarded the patient’s overall con- a mean of 4.8 ± 1.7 AEDs (range, 2-7) for seizure control. Nine pa-
dition, including seizure burden as well as cognitive and behavioral tients (75.0%) had a history of IS and all of them (9 patients with IS)
features. Parental assessments were obtained at clinic visits, and failed to control spasms with vigabatrin, which is an effective in the
were classified as much improved, somewhat improved, no interval treatment of TSC children with IS. Upon KD initiation, 6 patients
change, and worsened. (50%) had IS, 1 patient had Lennox-Gastaut Syndrome (LGS), and 5
NP evaluations were available in all 12 patients. As per formal NP (41.7%) patients had focal epilepsy. The mean age at KD initiation
evaluations, patients with an intelligence quotient (IQ) < 70 (or was 23.2 ± 21.7 months (range, 2.9-76.5 months) and the lead-time
sometimes a Developmental quotient [DQ] or social quotient [SQ] in to KD initiation (i.e., the time interval from seizure onset to KD ini-
cases where an IQ was not obtained) were considered to be cogni- tiation) was 13.7 ± 13.5 months (range, 1.5-38.0 months). The
tively impaired. mean duration of KD maintenance was 14.8 ± 12.8 months (range,

Table 1. Demographics of patients

Parameters Value
Gender, n (%)
Male 5 (41.7)
Female 7 (58.3)
Mean age at last follow up, mean ± SD (months) 73.1 ± 38.0 (range, 26.4-144.2)
Brain MRI findings, n (%)
Cortical tubers 12 (100.0)
SEN 12 (100.0)
SEGA 3 (25.0)
st
Seizure onset in the 1 year of life, n (%) 10 (83.3)
Seizure onset age, mean ± SD (months) 9.6 ± 12.5 (range, 1.0-47.4)
No. of tried AEDs, mean ± SD (months) 4.8 ± 1.7 (range, 2-7)
History of infantile spasms, n (%) 9 (75.0)
Developmental status at KD initiation, n (%)
IQ, DQ, or SQ ≥ 70 5 (41.7)
IQ, DQ, or SQ < 70 7 (58.3)
SD, standard deviation; MRI, magnetic resonance imaging; SEN, subependymal nodules; SEGA, subependymal giant cell astrocytoma; No.,
Number; AEDs, anti-epileptic drugs;IQ, intelligence quotient; DQ, developmental quotient; SQ, social quotient.

Copyright ⓒ 2017 Korean Epilepsy Society

 Soyoung Park, et al. KD for Management of Epilepsy with TSC in Children 47

3.3-43.6 months). Seven patients (58.3%) had cognitive impairment 6 (60.0%) were seizure free and 8 (80%) achieved a > 50% reduc-
at the time of KD initiation (Table 1, 2). tion in seizures. Reasons for KD discontinuation after 6 months were
limited effectiveness in 2 patients (50.0%), general weakness and
Seizure outcomes weight loss in 1 patient (25.0%), and parental will in 1 patient
After 3 months of the KD, 4 patients (33.3%) were seizure free (25.0%). Half of the initial 12 patients maintained the dietary ther-
and 10 patients (83.3%) achieved a > 50% reduction in seizures. apy for more than 12 months. Of these, 2 (33.3%) were seizure free,
Two patients discontinued the KD after 3 months due to seizure 2 achieved a 90% reduction in seizures, and 2 had a < 50% reduc-
aggravation. Among 10 patients who continued the KD for 6 months, tion in seizures after 12 months of the KD (Table 2, 3).
Of the 12 patients included in our analysis, 6 patients underwent
Table 2. Clinical characteristics and seizure outcomes related to KD epilepsy surgery after trying the KD; 3 of these patients (50.0%) ex-
Parameters Value perienced no effect on KD, 2 of them (33.3%) were seizure free with
Mean age at initiation of KD, 23.2 ± 21.7 (range, 2.9-76.5) KD for initial several months, but recurred, and 1 of them (16.7%)
mean ± SD (months)
had only partial effect by KD (75% reduction of seizure frequency for
Lead-time to KD, 13.7 ± 13.5 (range, 1.5-38.0)
mean ± SD (months) less than 3 months). 5 of above mentioned 6 patients (83.3%) were
(i.e the time interval from seizure free after surgery and 1 patient still experienced weekly
seizure onset to KD initiation)
seizures. Therefore, 8 of the 12 original patients included in our anal-
Mean duration of KD, 14.8 ± 12.8 (range, 3.3-43.6)
mean ± SD (months) ysis achieved seizure free status at last follow up; 1 (12.5%) was
Epilepsy classification at KD, n (%) seizure free with KD maintenance, 5 (62.5%) were seizure free after
IS 6 (50.0) surgery, and 2 (25.0%) were seizure free after finding an appropriate
LGS 1 (8.3) AED regimen, even after KD discontinuation.
Focal epilepsy 5 (41.7)
Seizure free after 5 (41.7)
Cognitive and behavioral features
3 months of KD, n (%)
Seizure reduction more than 50% 10 (83.3) Efficacy of the KD was mainly assessed by seizure outcome;
after 3 months of KD, n (%) however, the global outcome by parental assessment included
Epilepsy surgery after KD, n (%) 6 (50.0)
cognitive and behavioral features in addition to seizure burden.
KD, ketogenic diet; SD, standard deviation; IS, infantile spasms; Four patients (33.3%) were “much improved,” 5 patients (41.7%)
LGS, Lennox-Gastaut syndrome.
were “somewhat improved,” 2 patients (16.7%) showed “no in-

Table 3. Detailed profiles of all patients: tuberous sclerosis complex patients with epilepsy treated by KD

Age at Epilepsy
No. of Lead time to KD Duration of KD
KD initiation classification at Ratio of KD Peak effect of KD
patients (months) (months)
(months) KD initiation
1 2.9 IS 1.8 3:1 → 4:1 3.3 No interval change
90% Reduction for 3 months →
2 4.5 IS 1.5 4:1 3.9
aggravation
3 76.5 Focal epilepsy 29.1 MAD → 4:1 6.1 No interval change
4 6.0 IS 3.7 3:1 → 4:1 7.6 50% Reduction for 6 months
5 3.4 Focal epilepsy 38.0 4:1 7.8 No interval change
6 15.0 Focal epilepsy 10.7 4:1 10.3 Seizure free for 6 months → recur
7 4.0 IS 2.3 3:1 → 4:1 → 2:1 12.2 Seizure free for 3 months → recur
90% Reduction for 12 months →
8 8.0 IS 6.1 4:1 12.5
aggravation
9 3.9 IS 4.5 3:1 → MAD 12.5 Seizure free for 9 months → recur
75% Reduction for 20 months →
10 6.9 Focal epilepsy 13.3 4:1 20.9
aggravation
11 7.0 Focal epilepsy 17.0 4:1 → MAD 36.5 Seizure free for 8 months → recur
12 9.0 LGS 35.8 4:1 → 3:1 43.6 Seizure free over 60 months
KD, ketogenic diet; No., number; IS, infantile spasms; LGS, Lennox-Gastaut syndrome; MAD, modified Atkin’s diet.

www.kes.or.kr
48 Journal of Epilepsy Research Vol. 7, No. 1, 2017

a history of IS or active IS.

Neuropsychological deficits of TSC are associated with abnormal-
ities in the mTOR signaling pathway, which has a crucial role in brain
development. Previous studies have identified the ability of a KD to
decrease mTOR activation in animal models, providing a biological
3,20
basis for the mechanism of action for dietary treatment. Yet, the
exact mechanism of KD-mediated benefits in TSC remains unknown
Figure 1. Global outcome by parental assessment for ketogenic diet on
seizure, cognition and behavior.
and warrants further research.
This study had several limitations. First, we enrolled a small num-
terval change,” and 1 patient was (8.3%) “worsened” after 3 ber of patients and lacked precise data about seizure types, electro-
months of the KD (Fig. 1). encephalogram findings, and genetic studies. Second, considering
the retrospective nature of the study, data may have been incomplete
Discussion despite our efforts to be thorough in chart/record collection and
review.
The present results suggest that KD maintenance is useful for the In conclusion, KD therapy is an important treatment option in cas-
management of intractable seizures in patients with TSC. KD main- es of medically intractable epilepsy related to TSC. Considering the
tenance not only had beneficial effects on seizure frequency, but also efficacy of the KD for seizure reduction and neuropsychological im-
produced overall improvements in cognition or behavior as rated by provement in our study, additional studies should explore the ther-
patient’s caretakers/parents. One-third of patients were seizure free apeutic value and mechanism of a KD in TSC.
and 83.3% patients showed a > 50% reduction in seizure frequency
after 3 months of the KD. After 12 months of the KD, 4 of 6 patients Acknowledgements
(66.7%) achieved a > 50% reduction in seizures. Furthermore,
75.0% patients who maintained the KD for at least 3 months were This study was supported by the Soonchunhyang University
reported to have overall improvement (much or somewhat improved) Research Fund.
as per parental assessments. Given that cognitive and behavioral
problems are main features of CNS involvement in TSC,5,8,16 and that Conflicts of Interest
58.3% of patients had cognitive impairment defined by an IQ, SQ, or
DQ of < 70 prior to dietary therapy in our study, it is possible that the The authors have no financial conflicts of interest.
KD has multifaceted utility for addressing CNS symptoms in TSC.
Few previous studies have reported the efficacy of a KD for the References
treatment of epilepsy associated with TSC.17,18 Our finding of a 50%
to 90% reduction in seizure frequency at 6 months after diet ini- 1. Kassiri J, Snyder TJ, Bhargava R, Wheatley BM, Sinclair DB. Cortical
tiation is consistent with that a previous study by Kossoff et al.17 tubers, cognition, and epilepsy in tuberous sclerosis. Pediatr Neurol
Another study indicated that low glycemic index dietary therapy also 2011;44:328-32.
2. Krueger DA. Management of CNS-related Disease Manifestations in
produced good responses with minimal side effects in patients with
Patients With Tuberous Sclerosis Complex. Curr Treat Options Neurol
epilepsy related to TSC.19
2013;15:618-33.
Patients with TSC are at high risk for early-onset seizures including 3. Jülich K, Sahin M. Mechanism-based treatment in tuberous sclerosis
focal seizures and IS.9 Approximately 30% of infants with TSC pres- complex. Pediatr Neurol 2014;50:290-6.
ent with epileptic encephalopathy, which has a poor prognosis.6 In 4. Krueger DA, Northrup H, International Tuberous Sclerosis Complex
our study, 83.3% patients developed seizures in the 1st year of life Consensus Group. Tuberous sclerosis complex surveillance and man-
agement: recommendations of the 2012 international tuberous scle-
and 75.0% of patients had a history of IS. Indeed, half of our study
rosis complex consensus conference. Pediatr Neurol 2013;49:255-65.
population started KD to treat IS. Future studies should examine
5. Curatolo P, Moavero R, de Vries PJ. Neurological and neuropsychiatric
whether the KD has particular usefulness in a subset of patients with aspects of tuberous sclerosis complex. Lancet Neurol 2015;14:733-45.

Copyright ⓒ 2017 Korean Epilepsy Society

 Soyoung Park, et al. KD for Management of Epilepsy with TSC in Children 49

6. Saxena A, Sampson JR. Epilepsy in Tuberous Sclerosis: Phenotypes, lepsy associated with tuberous sclerosis. Epilepsia 1998;39:1158-63.
Mechanisms, and Treatments. Semin Neurol 2015;35:269-76. 14. Curatolo P. Mechanistic target of rapamycin (mTOR) in tuberous scle-
7. de Vries PJ, Whittemore VH, Leclezio L, et al. Tuberous sclerosis asso- rosis complex-associated epilepsy. Pediatr Neurol 2015;52:281-9.
ciated neuropsychiatric disorders (TAND) and the TAND Checklist. 15. Holmes GL, Stafstrom CE. Tuberous sclerosis complex and epilepsy: re-
Pediatr Neurol 2015;52:25-35. cent developments and future challenges. Epilepsia 2007;48:617-30.
8. Zaroff CM, Devinsky O, Miles D, Barr WB. Cognitive and behavioral cor- 16. Cusmai R, Moavero R, Bombardieri R, Vigevano F, Curatolo P. Long-term
relates of tuberous sclerosis complex. J Child Neurol 2004;19:847-52. neurological outcome in children with early-onset epilepsy associated
9. Curatolo P, Jóźwiak S, Nabbout R. Management of epilepsy asso- with tuberous sclerosis. Epilepsy Behav 2011;22:735-9.
ciated with tuberous sclerosis complex (TSC): clinical recommen- 17. Kossoff EH, Thiele EA, Pfeifer HH, McGrogan JR, Freeman JM.
dations. Eur J Paediatr Neurol 2012;16:582-6. Tuberous sclerosis complex and the ketogenic diet. Epilepsia
10. Thiele EA. Managing and understanding epilepsy in tuberous sclerosis 2005;46:1684-6.
complex. Epilepsia 2010;51 Suppl 1:90-1. 18. Coppola G, Klepper J, Ammendola E, et al. The effects of the keto-
11. Chu-Shore CJ, Major P, Camposano S, Muzykewicz D, Thiele EA. The genic diet in refractory partial seizures with reference to tuberous
natural history of epilepsy in tuberous sclerosis complex. Epilepsia sclerosis. Eur J Paediatr Neurol 2006;10:148-51.
2010;51:1236-41. 19. Larson AM, Pfeifer HH, Thiele EA. Low glycemic index treatment for
12. Moavero R, Cerminara C, Curatolo P. Epilepsy secondary to tuberous epilepsy in tuberous sclerosis complex. Epilepsy Res 2012;99:180-2.
sclerosis: lessons learned and current challenges. Childs Nerv Syst 20. McDaniel SS, Rensing NR, Thio LL, Yamada KA, Wong M. The keto-
2010;26:1495-504. genic diet inhibits the mammalian target of rapamycin (mTOR)
13. Ohtsuka Y, Ohmori I, Oka E. Long-term follow-up of childhood epi- pathway. Epilepsia 2011;52:e7-11.

www.kes.or.kr