If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Data on the efficacy of KD in the treatment of structural DRE are limited.
•
KD is effective and safe in Chinese children with structural DRE.
•
Patients with DRE secondary to HIE may be particularly responsive to the KD.
Abstract
Objectives
In this study, we aimed to evaluate the efficacy and tolerability of ketogenic diet (KD) in Chinese children with drug-resistant epilepsy (DRE) due to structural etiology.
Methods
We retrospectively analyzed data from 23 pediatric patients with DRE due to structural etiology who were treated with KD at Department of Neurology, between May 2014 and December 2020. Based on etiological classifications, the patients were divided into a neonatal brain injury (Group 1), an intracranial infection group (Group2) and a group that showed malformations of cortical development (MCDs) (Group 3).
Results
The 23 patients remained on the KD for a mean duration of 15.3 ± 9.7 months. The response rates for the control of seizures were 60.9% (14/23), 52.2 % (12/23), 47.8% (11/23) at 3, 6 and 12 months, respectively. Subjective improvements in cognition were observed in 87.0% (20/23) of the children during follow-up. Reductions in the frequency of seizures of > 50% were more commonly achieved by patients in group 1 (75.0%, 9/12) compared to the patients in groups 2 (60.0%, 3/5) and 3 (33.4%, 2/6). Further analysis of the patients in Group 1 showed that children with a history of hypoxic ischemic encephalopathy (HIE) (100.0%, 6/6) had the highest rate of > 50% seizure reduction. The main reasons for the discontinuation of the KD were due to lack of efficacy and poor compliance. Most of the side effects associated with the KD diet were minor and easily corrected by appropriately adjusting the diet. Only 1 patient discontinued the diet due to severe refusal to eat.
Conclusions
KD is an effective and safe treatment for Chinese children with DRE due to structural etiology. Better efficacy of seizure control was observed in patients with a history of neonatal brain injury. Patients with DRE secondary to HIE may be particularly responsive to the KD therapy, and so KD should be considered earlier in those patients.
Since the mid-1990s, ketogenic diet (KD) have become increasingly established as an effective and safe alternative therapy for drug-resistant epilepsy (DRE) [
Optimal clinical management of children receiving dietary therapies for epilepsy: updated recommendations of the International Ketogenic Diet Study Group.
]. Patients with structural etiologies including malformations of cortical development (MCDs) and structural brain lesions acquired in infancy (HIE, intra-ventricular hemorrhage, infections) are at a high risk of developing DRE [
]. Currently, data on the efficacy of KD in the treatment of structural DRE are limited. For children with structural epilepsy who are not suitable for surgery, it is very important to determine if the structural epilepsy is responsive to the KD after the failure of conventional anti-seizure medicines (ASMs). In this study, we analyzed the response to KD in a cohort of 23 pediatric patients with DRE related to structural etiology.
2. Methods
This study was performed as a single-center retrospective study that included 23 patients with DRE due to structural etiology who were treated with KD at the Department of Neurology of the Xi'an Children's Hospital, Xi'an (China) between May 2014 and December 2020. This study was approved by the Ethics Committee of Xi'an Children's Hospital.
Children with DRE secondary to neonatal structural lesions or infectious structural lesions or MCDs were enrolled in this study. Patients included in our study had failed two or more suitable ASMs. Seizures were classified and epileptic syndromes were diagnosed according to the 2017 International League Against Epilepsy classification [
Operational classification of seizure types by the international league against epilepsy: position paper of the ILAE commission for classification and terminology.
]. 12 patients had acquired structural epilepsy due to post neonatal brain injury (6 with HIE, 4 with hypoglycemic encephalopathy and 2 with cerebral hemorrhage) (Group 1), 5 patients had DRE involving intracranial infections (2 with viral encephalitis, 3 with bacterial meningitis) (Group 2), and 6 patients had malformations of cortical development (MCDs) (Group 3). The 6 patients with MCDs were stratified according to the etiological classification of MCD by Barkovich et al. [
] Of the 6 patients in this group, 2 of the patients had malformations of abnormal neural proliferation (1 with microlissencephaly and 1 with hemimegalencephaly), 2 had malformations of abnormal neural migration (1 with nodular heterotopia and 1 with lissencephaly), and 2 patients had malformations of abnormal post-migrational development (both with polymicrogyria).
All of the patients underwent detailed clinical history evaluation and physical examinations before starting the KD. Metabolic and renal conditions tests were performed to identify unsuitable patients from initiating the KD. Electroencephalogram (EEG) and intracranial magnetic resonance imaging (MRI) results were recorded.
The KD therapy was initiated using a classical protocol under inpatient care. A lipid to protein and carbohydrate ratio of 3:1 to 4:1 was used to achieve a beta-hydroxybutyrate (BOH) concentration of 2–4 mmol/L in the blood and a ketosis level of 3+ to 4+ in urine. Formula feeds were used for the children < 1 year during the initiation phase of the KD treatment. If the ketosis reached a high steady-state level (urine ketone body >3+), the recipe was partially or completely changed to Chinese food catering provided by a dietician. To maintain blood ketosis and appropriate growth, the energy, dietary ratio, and protein intake were modified as required. Daily supplements of potassium citrate and multiple vitamins were also given to the patients along with the KD. After discharge, blood glucose, blood BOH and urine ketones were monitored twice daily at home for all patients following the initial weeks of the treatment. Urinary ketones were monitored once a day after the first review.
AEDs were continued and remained stable during the first 3 months of the KD. All the guardians of the patients had telephone access to the dietician throughout the initiation and maintenance phases of the KD. Seizure diaries were recorded for 28 days before initiation of the KD and during entire interventional phase. The average daily seizure frequency was compared to the average baseline measures at 3, 6 and 12 months, and at the last follow-up. Responders were defined as patients who achieved >50% reduction in seizure frequency. Gesell Development Schedules (GDS) were used to measure the cognitive and motor development of the patients before initiation of the KD. During the follow-up period, cognitive changes were reported based on clinical observations and guardian's reports.
Categorical variables were described using frequencies and percentages. Continuous variables were described as the mean ± standard deviation or the median (range). Categorical variables were compared using the Fisher exact test and continuous variables were analyzed using a Kruskal-Wallis test across the 3 groups. Correlations between the reductions in seizures and serum BOH levels were assessed using the Spearman rank correlation. P values < 0.05 were considered statistically significant.
3. Results
3.1 Clinical Characteristics
Of the 23 patients (17 males, 6 females), 13 patients (56.5%) were diagnosed with West syndrome, 1 patient (4.3%) was diagnosed with Lennox-Gastaut syndrome, and 9 patients (39.1%) were diagnosed with other DRE. The median age of the patients at the first seizure was 6.0 (range: 0.6–59.0) months. The median duration of epilepsy before the KD was 7.0 (range: 1.5–40.0) months. Patients started the KD at a median age of 13.3 (range: 5.3–76.8) months. The median number of ASMs that the children had been prescribed before KD initiation was 4 (range: 3–5). Before the initiation of the KD, 21 patients had daily seizures, and 2 patients had weekly seizures. Developmental delay was severe in 17 patients and moderate in 6 patients. Brain MRI showed variable degrees of brain injury including multi-cortical damage, white matter gliosis, hydrocephalus, parietal-occipital lobe atrophy, microlissencephaly, hemimegalencephaly, lissencephaly, nodular heterotopia and multi-cortical polymicrogyria. The clinical details of the patients are summarized in Table 1. The brain MRI findings of several patients are shown in Fig. 1.
Table1Clinical profile of patients with drug-resistant structural epilepsy at baseline.
No.
Age(months)
Sex
Etiology
Seizure types and frequenceis, epileptic syndrome
Intellectual disability
MRI
Age at seizure onset (months)
Duration between seizure onset and KD starting (months)
AEDs at KDinitiation
Group 1: post neonatal brain injury (n = 12)
1
11.5
M
Hypoglycemic encephalopathy
Spasm: 2-4/day Tonic-clonic:0-1/month West syndrome
Yes
Bilateral parioccipital atrophy
6
5.5
VPA TPM LEV
2
13.3
M
HIE
Spasm: 6-20/day Tonic: 0-2/month West syndrome
Yes
Bilateral parietal atrophy
8.3
5.0
VPA TPM LEV
3
8.0
F
HIE
Spasm: 10-20/day Tonic: 0-2/month West syndrome
Yes
Bilateral temporal lobe atrophy
6.5
1.5
VPA TPM
4
8.4
M
Hypoglycemic encephalopathy
Spasm: 100-300/day West syndrome
Yes
Bilateral parioccipital atrophy
3.4
5.0
VPA TPM
5
5.3
M
HIE
Focal: 17-23/day Tonic: 0-2/month
Yes
Multi-cortex malacia and atrophy located in bilateral frontal, parietal and occipital lobes, combined with gliosis
2.0
3.3
VPA TPM VGB
6
12.6
M
Cerebral hemorrhage
Spasm: 10-20/day Absent:3-7/d West syndrome
Yes
Diffused brain atrophy and right basal ganglia malacia
7.6
5.0
VPA TPM LEV CLB
7
17.3
M
Hypoglycemic encephalopathy
Tonic-clonic: 2/month Spasm: 10-21/day West syndrome
Yes
Bilateral parioccipital atrophy
8.0
9.3
VPA TPM CLB
8
11.2
M
HIE
Spasm: 10-20/d West syndrome
Yes
Multi-cortex malacia located in bilateral frontal, parietal and occipatal lobes
8.2
3.0
VPA TPM Prednisolone
9
14.6
M
HIE
Spasm: 40-50/d West syndrome
Yes
Multi-cortex malacia and atrophy located in right temporal,occipital,and left frontal, temporal, parietal and occipital lobes
6.0
8.6
VPA Prednisolone
10
9.5
M
HIE
Focal: 3-5/week
Yes
Multi-cortex malacia and atrophy located in bilateral frontal, temporal and parietal lobes
3.5
6.0
TPM LEV VGB
11
29
M
Hypoglycemic encephalopathy
Spasm: 2-4/d, Tonic: 1-3/month West syndrome
Yes
Bilateral parioccipital atrophy
7.2
21.8
VPA TPM LTG
12
30.6
M
Cerebral hemorrhage
Spasm: 32-40/d, Focal: 1-3/day West syndrome
Multi-cortex malacia and atrophy located in bilateral frontal lobes
1.5
29.1
VPA TPM LEV CLB
Group 2: post intracranial infection (n = 5)
13
76.8
F
Viral encephalitis
Focal: 4-5/week Myoclonic: 0-2/day
Yes
Bilateral parioccipital atrophy
59.0
17.8
VPA TPM CLB
14
31.6
F
Viral encephalitis
Focal: 5-6/day Tonic: 0-2/month
Yes
Multi-cortex malacia and atrophy located in bilateral thalamus, frontal, temporal and parietal lobes
21.0
10.6
LEV VPA OXC
15
39.4
M
Bacterial meningitis
Spasm: 20-25/d
Yes
Hydrocephalus and diffused brain atrophy
18.0
21.4
VPA TPM LEV
16
29.3
M
Bacterial meningitis
Focal: 12-13/d
Yes
Hydrocephalus and diffused brain atrophy
6.7
22.6
VPA TPM CLB
17
12.7
F
Bacterial meningitis
Focal: 11-13/d
Yes
Hydrocephalus and diffused brain atrophy
5.7
7.0
VPA TPM LEV
Group 3: malformations of cortical development (n = 6)
18
6.1
F
Microlissencephaly
Spasm: 3-6/day Focal: 10-20/day West syndrome
Yes
Microlissencephaly
2.0
4.1
VPA TPM LEV Prednisolone
19
8.9
M
Hemimegalencephaly
Spasm: 17-36/day Tonic-clonic:0-1/month West syndrome
Yes
Right hemisphere dysgenesis with dilation of the right lateral ventricle
Fig. 1Brain MRI findings of DRE due to structural etiology: A (Patient 9, HIE) showed multi-cortex malacia and atrophy located in right temporal, occipital, and left frontal, temporal, parietal and occipital lobes postneonatal HIE. B (Patient 11, hypoglycemic encephalopathy) showed bilateral parietal-occipital atrophy due to neonatal hypoglycemic encephalopathy. C (Patient 15, bacterial meningitis) showed hydrocephalus and diffused brain atrophy secondary to bacterial meningitis. D (Patient 19, dysplastic hemimegalencephaly) showed right hemisphere dysgenesis with dilation of the right lateral ventricle. E (Patient 23, lissencephaly) showed a lissencephalic cortex.
The mean duration of patients who remained on the KD treatment was 15.3 ± 9.7 months. The responder rates of seizure control were 60.9% (14/23), 52.2 % (12/23), 47.8% (11/23) respectively at 3, 6 and 12 months. At the last follow-up, 14 patients (60.9%) were responders, 5 patients (21.7%) obtained seizure freedom. > 50% seizure reduction was observed in 66.7% (10/15) of the patients with epileptic spasms and 50% (5/10) of the patients who had focal seizures. The responder rates of seizure control were 61.5% (8/13) in 13 patients with West syndrome, and 66.7% (6/9) in the 9 patients with DRE at 3 months. The patient with Lennox-Gastaut syndrome did not report improvements in the control of seizures. Subjective improvements in cognition were reported including increased alertness, increased vocalization and developmental improvements in 87.0% (20/23) of the children during follow-up. No significant correlation between the serum BOH and reduction in seizures was found at 3 months (p = 0.151). The median levels of serum BOH were 2.8 mmol/L in Group 1, 2.9 mmol/L in Group 2 and 2.65 mmol/L in Group 3 at 3 months of the KD. There was no significant difference in the level of ketosis in the 3 different patient groups.
In Group 1, 9 (75.0%) patients had a seizure reduction of > 50% at 3 months that included 3 patients (25.0%) who obtained seizure freedom. In Group 2, 3 patients (60.0%) reported a reduction of > 50% in seizure and no patient was seizure free at 3 months. In Group 3, 2 patients (33.3%) were responders, and 1 patient was seizure free (16.7%) at 3 months (Table 2). The data were analyzed using Fisher exact test: despite no statistical significance being found amongst the 3 groups, a positive trend suggested a better response to the KD in patients from Group 1 than those in Groups 2 and 3. The details of the response to the KD in all of the 23 patients are summarized in Table 2.
Table 2The responses to the KD of patients with drug-resistant structural epilepsy during follow-up.
No.
Response to KD at 3 months
Response to KD at 6 months
Response to KD at 12 months
Response to KD at last follow-up
Subjective improvements in cognition
Duration of KD retention(months)
KD ratio at maintenance phase
serum BOH at 3 months(mm ol/L)
Last follow-up
Side effects
Reasons for diet discontinuation
Group 1: post neonatal brain injury (n = 12)
1
25% reduction
25% reduction
30% reduction
25% reduction
More alert Developmental improvement
39.6
3:1
3.5
Discontinued
No
Lack of efficacy
2
70% reduction
80% reduction
75% reduction
75% reduction
More alert Developmental improvement
27.5
3.5:1
3.0
Discontinued
No
Poor compliance
3
50% reduction
Discontinued
Discontinued
50% reduction
More alert Increased vocalization
3.7
3:1
2.0
Discontinued
No
Swich to a new antiepileptic drug
4
No reduction
Discontinued
Discontinued
No reduction
More alert Increased vocalization Sitting up without support
Of the 12 patients from Group 1, children with a history of HIE (100.0%, 6/6) had the highest rate of seizure reduction. However, children with a history of hypoglycemic encephalopathy had poor responses to the KD. Patients in Group 2 who had a history of bacterial meningitis (66.7%, 2/3) showed improved responses to the KD. The absolute numbers were not sufficiently large to compare the differences in effective rates of response amongst the patients with different malformation types in Group 3. The responses to KD of different etiologies are summarized in Fig. 2, Fig. 3.
Fig. 2The responses to the KD treatment of different etiologies at 3 months.
8 patients (34.8%) discontinued KD before 12 months. This was due to lack of efficacy in 3 patients, poor compliance in 1 patient, severe refusal to eat in 1 patient, decreased seizure control in 1 patient, 1 patient requiring surgery, and 1 patient who was switched to a new ASM. 6 patients (26.1%) remained on the diet at the last follow-up. The KD was discontinued (> 1 year) due to poor compliance in 5 patients, lack of efficacy in 2 patients, decreased seizure control in 1 patient, and the death of 1 patient. The specific reasons for the discontinuation of the KD are shown in Table 2.
3.4 Side effects
During the initiation phase of the KD treatment, 1 patient experienced hypoglycemia, 2 patients experienced vomiting, and 2 patients presented refusal to eat. During the maintenance phase of the diet, 1 patient had diarrhea and 2 patients refused to eat. The KD was well tolerated in most cases, and most of the above side effects were minor and disappeared by adjusting the diet. Only 1 patient discontinued the diet due to severe refusal to eat.
4. Discussion
KD has been reported as a tolerable and effective diet intervention for children with DRE in different countries [
A pragmatic study on efficacy, tolerability and long term acceptance of ketogenic diet therapy in 74 South Indian children with pharmacoresistant epilepsy.
]. However, little is known about the efficacy of KD in the treatment of structural DRE. In this study, we report on the efficacy and tolerability of the KD in a retrospective series of 23 children with structural DRE that were recruited over 6.5 years.
In our study, >50% reduction in the frequency of seizures was found in up to 60.9% of the patients. The efficacy of the KD in our patients was consistent with the general data of all seizures with both structural or idiopathic etiology [
], subjective improvements in cognition were obtained in 20 (87.0%) patients during the follow-up period in our study. Also, another study focused on the patients with MCDs reported that only 46.7% of the patients improved after 3 months on the KD [
]. These results may be due to etiological and individual differences in patients within the different study groups. Also, the subjective nature of reporting cognition may impact the reliability of these data.
], 66.7% of the patients with epileptic spasms and 50% of the patients with focal seizures had a response to KD in our study. For other seizure types, the number of patients was too small or the seizure frequency was not sufficiently large to differentiate the efficacy of the KD treatment. At 3, 6, and 12 months, 30.8%, 37.5% and 33.3%, respectively, of our patients with West syndrome who remained on the diet at that time point were seizure free; 61.5%, 75.0 % and 66.7% were > 50% improved. Our data were consistent with the general data for West syndrome patients with both structural or idiopathic etiologies [
The relationship between d-beta-hydroxybutyrate blood concentrations and seizure control in children treated with the ketogenic diet for medically intractable epilepsy.
], and may be explained by the the small sample in our study. A BOH concentration of > 3mmol/L was difficult to achieve and maintain for most of the children in our study who were used to a carbohydrate rich diet.
Our results suggested that patients with acquired structural epilepsy due to postneonatal brain injury had the best response to KD, with a reduction in seizures of > 50% in 9 (75.0%) patients at 3 months. Our results agreed with those reported by Villaluz et al. [
] who observed that 7 out of 9 patients with acquired structural epileptic encephalopathy were responders at 3 months. The median age, gender, median duration of epilepsy, median age of seizure onset and the median level of serum BOH were not significantly different amongst the 3 groups. It is unclear why patients with a history of neonatal brain injury had the best response and further investigations are needed to better understand these data.
Of the 9 patients who responded to the KD in Group 1, 6 (6/6) patients with HIE, 2 (2/2) patients with cerebral hemorrhage and 1 (1/4) patient with hypoglycemic encephalopathy showed > 50% response at 3 months. At the last follow-up, 6 patients with HIE were still responders. The responder rate decreased to 1/2 in patients with cerebral hemorrhage and still only one patient with hypoglycemic encephalopathy responded to the KD. The responder rate of patients with HIE was considerably greater than in patients with other structural etiologies which was consistent with the previous reports [
]. Unexpectedly, we found that the responder rate of patients with hypoglycemic encephalopathy was much lower than the other patients in Group 1 caused by HIE or cerebral hemorrhage. These data suggest that patients with a history of HIE may be particularly responsive to the KD whilst the patients with a history of hypoglycemic encephalopathy may not respond well. Studies with larger sample sizes are required to confirm these findings.
Based on our data, KD is a promising therapy for patients with a history of intracranial infection as observed in 5 of the patients in this study. Two-thirds of the patients with bacterial meningitis and 1 out of 2 patients with viral encephalitis had a response with a >50% reduction in seizures. Thammongkol et al. [
] reported that 1 (1/1) patient with pneumococcal meningitis had a response of a >50% seizure reduction, but the patient with viral encephalitis did not respond to KD.
Three-sixths of the patients with MCDs showed a response at the last follow-up. 1 child with hemimegalencephaly was seizure freedom for over 20 months. 1 patient with nodular heterotopia and 1 patient with polymicrogyria had responses to the KD. Unfortunately, the number of patients in the analysis was too small to differentiate the efficacy of the KD treatment according to specific types of MCD. Malformations with good responses included bilateral focal cortical dysplasia, lissencephaly, perisylvian polymicrogyria and hemispheric dysplasia [
]. In contrast to the previous studies, 1 (1/2) patient with polymicrogyria and 1 (1/1) patient with lissencephaly didn't respond to the KD therapy in our study which may also be explained by the small sample size.
To date, the mechanisms underlying the efficacy of the KD in the control of seizures remain incompletely understood despite being of interest for the past 20 years [
]. The relevance between the anti-inflammatory effects of KD and the efficacy of KD in patients with structural epilepsy secondary to brain injury or intracranial infection should also be considered. Inhibition of the mammalian target of rapamycin (mTOR) pathway might be one potential mechanism underlying the efficacy of the KD in these patients [
]. Other hypothesis like immature cerebral cortex using ketone bodies more effectively and KD having inhibitory effect on the kainate model of epilepsy were also investigated in previous research [
The main reasons for the discontinuation of the KD were poor compliance and lack of efficacy. The rate of patients who discontinued the KD due to poor compliance was much higher than the study of Pasca et al. [
]. The diet is a strict restriction of carbohydrates which conflicts with carbohydrate-based diet in China making the KD difficult to accept and maintain. Less restrictive diets may be an alternative approach. Unfortunately, 1 patient (Patient 14) died after 14.4 months of the KD treatment. There were no KD related side effects reported during follow-up before the patient died, and the cause of the patient's death was unknown.
Most of the side effects were minor and could be easily corrected by adjusting the fatty ratio or changing the consistency of the meals in our study. These findings were in consistent with previous studies [
A pragmatic study on efficacy, tolerability and long term acceptance of ketogenic diet therapy in 74 South Indian children with pharmacoresistant epilepsy.
Our program provides the first case series focused on reporting the efficacy of KD therapy in patients with a structural etiology in China. However, this study had several limitations. Firstly, the results of this study may be impacted by the small sample size and the retrospective nature of the data extraction. Secondly, the absolute numbers of patients with different malformation types in Group 3 were too small and so meaningful comparisons could not be made. Thirdly, no standardized questionnaire was used to assess developmental outcomes during the follow-up period. Finally, the subjective nature of the data concerning cognitive improvement makes it difficult to draw conclusions based on the observed trends.
5. Conclusions
The results from this study show that KD is effective in reducing the frequency of seizures in Chinese children with DRE secondary to structural etiology. The group of patients with a history of neonatal brain injury showed a better response to the KD. Patients with DRE secondary to HIE may be particularly responsive to the KD, whilst the patients with a history of hypoglycemic encephalopathy are less responsive. In the future, larger case-series studies are needed to better evaluate the potential of KD interventions and to identify factors to predict response for epilepsy of structural etiology.
Compliance with ethical standards
The clinical data was extracted after the necessary approvals from the hospital ethics committee. Informed consent was obtained from the guardians of all the participants involved in the study.
Optimal clinical management of children receiving dietary therapies for epilepsy: updated recommendations of the International Ketogenic Diet Study Group.
Operational classification of seizure types by the international league against epilepsy: position paper of the ILAE commission for classification and terminology.
A pragmatic study on efficacy, tolerability and long term acceptance of ketogenic diet therapy in 74 South Indian children with pharmacoresistant epilepsy.
The relationship between d-beta-hydroxybutyrate blood concentrations and seizure control in children treated with the ketogenic diet for medically intractable epilepsy.
00367-8&id=gr1.jpg){kind=link}
00367-8&id=gr2.jpg){kind=link}
00367-8&id=gr3.jpg){kind=link}