Bachelors thesis conducted at the Erasmus MC Sophia Children’s Hospital
Fine-tuning of the
ketogenic diet: treatment
of intractable epilepsy in
infants
Carolina Broek
Thesis number 2018122
Erasmus MC Sophia’s Children’s Hospital
University of applied sciences Amsterdam
Bachelors Nutrition and Dietetics
1
Fine-tuning of the ketogenic diet: treatment of intractable
epilepsy in infants
Author
Thesis number
Client
Thesis counselor
Teacher counselor
Carolina Broek 500701971 carolinabroek@hotmail.com 2018122Erasmus MC Sophia’s Children’s Hospital
Elles van der Louw
2
Preface
This is my thesis on the fine-tuning of the ketogenic diet in infants with intractable epilepsy. The thesis was conducted at the Erasmus MC Sophia’s Children’s Hospital as part of my bachelor’s degree at the University of applied sciences in Amsterdam and was written in the period of September 2016 to January of 2017.
For the course multidisciplinary treatment, I made a short film about epilepsy and the ketogenic diet. It fascinated me how the right nutrition could help where medicine could not. This, combined with my fondness for children, made this thesis appealing to me. Conducting my thesis at an academic hospital offered me a huge challenge which I was looking for. Although it certainly was not easy, it was an honor to be given the opportunity to conduct my thesis at the Erasmus MC. I would like to give a special thanks to Elles van der Louw, my thesis counselor at the Erasmus MC, for providing me with continuous feedback and supporting me throughout this process. I would also like to thank Annemarie Both-Zuur, my teacher counselor, for the feedback and support she has given.
I hope you enjoy your reading. Carolina Broek, Haarlem 16 December 2017
3
Abstract
Introduction: The ketogenic diet can be used for the treatment of intractable epilepsy. This is a high fat, low protein and carbohydrate diet. There has been little research into the ketogenic diet for children under the age of 2. For infants it is expected that, due to development combined with frequent illness and the goal of maximal efficacy, the diet must be adjusted frequently. The following question will be answered in this thesis: What determinants are reasons for dietary adjustments and how effective are these adjustments, in short (0-3 months) and long-term (3-6 months) after the initiation of the ketogenic diet in infants 0-2 years of age treated for intractable epilepsy in the Erasmus MC Sophia Children’s Hospital?
Methods: In this retrospective study infants treated with the ketogenic diet for intractable epilepsy were followed for three to six months after diet initiation. For each patient a case report form was filled out containing information on patient characteristics and determinants for fine-tuning and effectiveness. Chi-square tests were done to determine if there were more changes in short-term or long-term and if changes were more effective.
Results: In total 45 patients were included in the study. The diet was adjusted 454 times for 43 different determinants. Overall the diet was mostly fine-tuned for the category Development. In both short and long-term, the diet was most often adjusted for growth, illness and high ketosis. The number of changes was significantly different for illness (p=0.043) and high ketosis (p=0.006). In both T1 and T2, growth and illness were significantly different in effectiveness (p=0.00). In T2, development was also significantly different (p=0.001)
Conclusion: The study confirms the expectancy that development is often reason for dietary adjustments. In both short and long-term growth, illness and high ketosis were adjusted for most often. There was a significant difference in effectiveness for the determinants growth and illness. Implementing medicine as primary treatment for gastrointestinal problems could improve overall treatment. A do’s and don’ts list could be compiled for the dietary changes to improve overall effectiveness. Possibly leading to a less burdensome experience of the diet and increased compliance. Keywords: ketogenic diet, intractable epilepsy, infants, fine-tuning.
5
Table of contents
Preface ... 2
Abstract ... 3
1.
Introduction ... 7
2.
Methods ... 10
2.1 Study design ... 10 2.2 Subjects ... 10 2.3 Data collection ... 10 2.4 Data analysis ... 113
Results ... 13
3.1 Patient characteristics ... 133.2 Number of dietary changes... 14
3.2.1 Dietary changes in total follow up period ... 14
3.2.2 Dietary changes within 0-3 months (T1) and 3-6 months (T2) after diet initiation ... 14
First three months after initiation ... 14
Three to six months after initiation ... 14
3.3 Determinants for fine-tuning the diet ... 15
3.3.1 Efficacy ... 15
3.3.2 Side effects... 16
3.3.3 Development ... 17
3.3.4 Other ... 18
3.3.5 Top four determinants ... 19
3.3.5.1 Chi-square test frequency ... 20
3.4 Effectiveness dietary changes ... 21
3.4.1 Chi-square tests effectiveness ... 22
3.4.2 Dietary changes ... 23
4.
Discussion ... 25
5.
Conclusion and recommendations ... 27
6.
References ... 28
7.
Appendix ... 30
Appendix 1. Fine-tuning measurements ... 30
Appendix 2. Indications and contra-indications for the ketogenic diet in infants ... 31
Appendix 3. Case report form ... 32
Appendix 4. Determinants and count dietary changes ... 38
Appendix 5. Criteria effective and not effective dietary changes ... 41
Appendix 6. Dietary changes effective and not effective ... 42
7
1. Introduction
The ketogenic diet was developed in the 1920’s after it was discovered that fasting could help with the prevention of seizures (1). During fasting there is not enough glucose for the brain to fully function. Therefore, the brain switches to ketones as its main energy source. Ketones are a byproduct of the fat metabolism. When the body uses ketones as main energy source, the body is in state of ketosis. Since this discovery, the ketogenic diet has been used to treat patients with epilepsy. Epilepsy is a neurological condition which is characterized by two or more epileptic seizures with no direct identifiable cause. These seizures are a result of an abnormal and excessive discharge of neurons in the brain (2,3,4). When epilepsy is diagnosed, the main treatment is with antiepileptic drugs (AED’s). These AED’s can completely control the seizures in up to 70% of epileptic patients (5). The group unable to be treated with the correct use of multiple AED’s, is said to have intractable epilepsy. Of all patients with epilepsy 20-30% develop intractable epilepsy (1,6).
The ketogenic diet can be used as alternative treatment for this type of epilepsy. The physiological working of this diet has not yet been scientifically determined. One hypothesis is that the combination of less glucose and higher levels of free fatty acids cause less neuronal (over)stimulation and therefore, reduces the prevalence of seizures (7).
There are different versions of the ketogenic diet. The Classical ketogenic diet, the medium chain triglyceride (MCT)-ketogenic diet, the modified Atkins diet and the low glycemic index ketogenic diet. This thesis focuses mainly on the Classical ketogenic diet and the MCT-version of the diet. These are the two main versions of the ketogenic diet and are used for the treatment of intractable epilepsy (8). In this Classical version, the fat, protein and carbohydrate ratio is 3:1. For every 3 grams of fat, there is 1 gram of combined carbohydrate and protein. The fats are mainly long chain triglyceride fats. The ratio can be adjusted to the individual patient needs, side effects and the correct level of ketosis. In the second version, the MCT version, less fat is required to achieve the same ketone levels. This allows for more protein and carbohydrates in the diet, therefore the diet is less strict and for the patient easier to maintain (10). The outcome of seizure control can be affected when ketosis is not within the optimum range of 2-5 mmol/L (9). Maximum efficacy being the main goal of this diet, it is always attempted to achieve the 2-5 mmol/L. To maintain this range the diet can be fine-tuned with different dietary changes. These dietary changes are mentioned in Appendix 1. In most cases the ketogenic diet does not contain enough micronutrients. To ensure daily required intake, these should be supplemented (9).
A full physical check-up is essential before the initiation of the diet, to ensure the patient does not have any contraindication for the use of the ketogenic diet. One example of these contra-indications is fatty acid oxidation deficiencies, all the contraindications are included in Appendix 2. When the full check-up is completed and there are no contraindications, the diet can be initiated. An individual diet plan is compiled for each patient. Parents and caregivers are instructed regarding the different aspects of the ketogenic diet. These aspects include calculating with the diet ratio, monitoring ketose levels, monitoring blood glucose levels and actions needed when levels are not within normal ranges. Furthermore, an emergency plan is discussed in event of a patient becoming ill. Under the age of one, initiation of the diet requires most patients to be admitted to the hospital for close monitoring. If the patient is above the age of one and the medical situation allows it, the ketogenic diet can be initiated as outpatient treatment. The full diet is gradually built up with a personal step-by-step plan. This takes 3-5 days to implement during clinical initiation and 28 days during outpatient initiation.
As side effect, hypoglycemia may appear during diet initiation because the number of carbohydrates is decreased and the body switches to fat as its main energy source. Along with the blood glucose levels, ketone levels are monitored. This is done to monitor the state of ketosis and prevent hyperketosis. Symptoms of hyperketosis can be rapid breathing, increased heart rate, facial flushing, irritable, vomiting, lethargy and poor feeding. Literature has shown that various adverse side effects
8 may occur whilst on this diet, such as diarrhea, obstipation, kidney stones, nausea, vomiting and hypoglycemia. By adjusting the diet, these side effects can often be remedied (9,11,12,13,14). While following the diet regular check-ups are essential and fine-tuning is an important part of the process, not only to monitor side effects but also ensure maximum efficacy towards seizure reduction (10).
In children ages 2 to 16, the ketogenic diet has been scientifically proven to be safe and effective (14). There is little to no research into treatment with the ketogenic diet under the age of 2 years. Research that is available for these infants is mainly focused on the efficacy and the side effects that might occur. However, in this age group there are many other factors that need to be considered when on the ketogenic diet. From birth until they are two years old, infants gain 3.5 times their body weight and grow approximately 38 cm, which is more than three quarters of their length at birth (15). In this short period of time they also develop motor- and sensory skills. These processes require an adequate intake of calories and nutrients. These factors combined with the appearance of frequent intercurrent illness and pursuing maximum efficacy of this strict diet makes it very difficult to design an ideal diet. Therefore, it is expected that the diet needs to be frequently adjusted to meet all these requirements. The adjustment of the diet is also called fine-tuning the diet. In this thesis it will be researched if this expectancy corresponds to practice. This will entail researching the frequency of dietary adjustments and the reasons therefore. The effectiveness of these changes will also be evaluated.
This descriptive study is conducted in infants (0-2 years of age) with intractable epilepsy at the Erasmus MC Sophia Children’s Hospital. They are one of five centers in the Netherlands that treat infants with the ketogenic diet. The Erasmus MC Sophia Children's Hospital is an expertise center in the field of ketogenic diet treatment in the Netherlands. Under intensive supervision of a multidisciplinary team, children with intractable epilepsy or metabolic diseases are treated with the ketogenic diet. In addition, Erasmus MC researches the treatment of the ketogenic diet to optimize treatment and the efficacy of the diet. This study is part of a larger research project on the ketogenic diet in infants as the PhD project of Elles van der Louw.
Outcomes of this study can result in the implementation of structural changes of the treatment protocol of the ketogenic diet in infants. For instance, prevention of adverse side effects can be a reason for structural changes. If this study shows that determinants related to the development of motor- and sensory skills are frequently reason for dietary adjustments, then increased monitoring for these specific determinants can be implemented. Outcomes related to effect of the dietary changes can be of use to increase effectiveness and possibly reduce the number of dietary changes. This might reduce time and efforts of both parents and the multidisciplinary team.
The main question of this thesis is: What determinants are reasons for dietary adjustments (fine-tuning) and how effective are these adjustments, in short (0-3 months) and long-term (3-6 months) after the initiation of the ketogenic diet in infants 0-2 years of age treated for intractable epilepsy in the Erasmus MC Sophia Children’s Hospital?
The sub questions of this thesis are:
- How often was the diet adjusted in the total follow-up period (0-6 months after diet initiation)?
- How often was the diet adjusted in short-term (0-3 months after initiation)? - How often was the diet adjusted in long-term (3-6 months after initiation)? - What determinants were reason for dietary adjustments?
- Which determinants occur most often in the total follow-up period (0-6 months after initiation?)
- Which determinants occur most often in short-term (0-3 months after initiation)? - Which determinants occur most often in long-term (3-6 months after initiation)?
9 - Is there a significant difference between the number of adjustments in short and long-term? - How often are the dietary adjustments effective and not effective?
- Is there a significant difference in effectiveness for the top four determinants?
- What were the specific dietary adjustments for the determinants with a significant difference in effectiveness?
The study design will be discussed in chapter 2 methods. In chapter 3 the results, the outcomes of this study will be further addressed. The remarkable outcomes will be compared to literature in chapter 4 the discussion. This chapter will also include the strengths and weaknesses of this study. answer to the main question and recommendations can be found in chapter 5 conclusion and recommendations.
10
2. Methods
2.1 Study design
This thesis was a descriptive study, with a retrospective longitudinal design conducted at the Erasmus MC Sophia Children’s Hospital. To answer the main question of this thesis, infants following the ketogenic diet as treatment for intractable epilepsy were observed for three to six months. The term determinants used in the main question includes all reasons and variables for dietary adjustments. The time periods (short and long-term) have been chosen based on the international guidelines for infants with refractory epilepsy. These indicate that after diet initiation the diet must be followed for at least three months to enable the evaluation of efficacy. Whereby, efficacy is defined as at least 50% seizure reduction (9).
2.2 Subjects
The inclusion and exclusion criteria are schematically noted in Table 1. The age was limited to 0-2 years because there is very little research on the ketogenic diet in infants. The other inclusion criteria were based on the main question.
Table 1. Inclusion and exclusion criteria
Inclusion criteria
❖ Infants, 0-2 years old
❖ Treated with the ketogenic diet for intractable epilepsy for at least 3 months
❖ Treated between 1-1-2008 and 1-10-2017 ❖ Treated at the Erasmus MC-Sophia
Exclusion criteria
❖ Missing data regarding determinants for dietary adjustments
2.3 Data collection
Case report formsFor each patient a case report form was filled out. This case report form is included in Appendix 3. The form contained general patient information, prematurity, growth retardation, ethnicity, etiology, electro-clinical syndrome, number of AED’s at start, manner of feeding, start and stop date of the ketogenic diet and reason of stop. This information was used for a Table containing the patient characteristics. These forms also included the determinants for dietary adjustments, the frequency of dietary adjustments and effectiveness of these adjustments.
Determinants
The reasons for dietary adjustments were determined by current literature. This list was supplemented with reasons recorded by the dietitians in the electronic patient case files during treatment. The literature was found in PubMed, ScienceDirect, Google Scholar and Cochrane library databases. The following search terms were used: ketogenic diet, infants, epilepsy and side effects. They were used in different combinations (e.g. ketogenic diet AND infants, ketogenic diet AND side effects AND infants.) Since determinants are described differently in each study, finding articles was difficult. To find more articles on these determinants the snowball method was used. Only free full-text articles written in English were included. All the determinants are included in Table 2.
11
2.4 Data analysis
The patient characteristics which were included in Table 3 were tested for skewness. When normally distributed the variable was presented as mean (SD), when skewed presented as median (IQR). Number of adjustments per determinant
All the determinants and the frequency that these determinants were reason for dietary changes, were listed in a table. This was done for the total follow up period of six months and further divided into the subgroups short (T1) and long-term (T2). From this table, the total number of adjustments was calculated for each time period. For each time period pie charts were made. These pie charts contained the number of adjustments divided into the four categories as noted in Table 2.
The number of dietary changes for each determinant were further discussed in the four categories as noted in Table 2. Hereafter, each category and its determinants were further explained:
- Efficacy: determinants were related to the level of ketosis and especially to the fine-tuning of the diet as noted in Appendix 1.
- Side effects: determinants were mainly related to adverse side effects. Other side effects are insufficient growth and weight loss. With the strictness of the diet, sufficient intake can be challenging. This can result in weight loss and insufficient growth. Insufficient growth was scored when growth was less than the expected growth curve.
- Development: determinants were part of general improved motor skills. The specific determinants were growth, development, teething, prolonging overnight fast and increased physical activity. This requires an adjustment in type, volume and structure of the diet. The determinant development was scored when patients started with a vegetable bite or expanded their solid food. These are called complementary foods. Within the category of Development, the determinant development was used. This was done because a certain level of physical development is needed to expand the diet with complementary foods. Therefore, using the determinant complementary food would not be adequate as reason for dietary adjustments.
- Other: all miscellaneous determinants were included in this category.
For each category a stacked column was made which contained all the determinants. In these stacked columns the number of dietary changes in T1, per category, add up to 100%. The same applies to T2.
Table 2. Reasons for dietary adjustments (Determinants)
❖ Efficacy
Ketose too low (<2mmol/L) *2 Ketose too high (>5mmol/L) *2 Optimizing ketosis *1 Fluctuating ketosis*1 ❖ Side effects Diarrhea*5 Obstipation*3 Nausea*2 Vomiting*4 Kidney stones*4 Reflux *3 Gastric acid*1 General gastric problems*2 Cramps*1 Hypoglycemia*2 Weight loss*3 Insufficient growth*6 ❖ Development Teething *1 Increased physical activity *1 Development*1 Growth *1 Prolonged night fast *1 ❖ Other Illness *2 Tube feeding *1 Coughing (vomiting) *1 Rejecting food *1 Food saturation*1 Tired (crying) *1 Hungry *1 Blood levels*2 Urine levels*2 Insufficient fluid *1 OK (operation) *1
*1= Patient case files
*2= Van der Louw et al, 2006 (9)
*3= Kossoff et al, 2008 (24) *4= Nordli et al 2001 (19) *5= Hong et al, 2010 (20) *6= Dressler et al, 2015 (22)
12 Furthermore, the number of dietary adjustments were sorted descending. This descending list was used to compose pie charts for the three time periods. For every ten patients, one variable could be tested. Therefore, the pie charts contained the top four determinants and the remainder of the determinants which were named ‘other’ in this pie chart.
A chi-square test for distribution was done with the top four determinants to determine if there was a significant difference between the number of dietary changes in subgroups T1 and T2. The outcome was significant when p<0.05.
Effectiveness
The effectiveness of the dietary adjustments was also assessed. The effect was scored in terms of effective and not effective. Since effectiveness is different for each determinant, the exact criteria for scoring effective was included in Appendix 5. Generally, the criteria was scored effective if the situation improved and not effective if the situation was unchanged or worsened. When data was missing regarding the effectiveness, it was scored as unknown. This was used when patients stopped the diet, passed away or if changes were made in the last consult (only in T2).
A table was composed containing the information on effectiveness. The table was divided into T1 and T2. It was further divided into effective and not effective. For each determinant the frequency and specific dietary adjustment were noted. When calories were added this was noted as a percentage of the whole diet e.g. total calories are 1000 kcal, adding 50 calories equals adding (5%). Current literature shows that 1.5g protein/100 calories should be sufficient for adequate growth (23). Therefore, also a calorie/protein ratio was included (e.g. when 50 calories and 2 grams of protein are added this will be noted as 4/100). The full table on effectiveness was included in Appendix 6. The effectiveness of the top four determinants, divided into the subgroups T1 and T2, were included in the results. A chi-square test for distribution was done to determine if there was a significant difference between effectiveness in each subgroup. The outcome was significant when p<0.05. If more that 20% of the values were under 5, the Fischer’s exact test was used to determine significance. All statistical analyses were done in IBM SPSS Statistics 24.
13
3 Results
3.1 Patient characteristics
In total 45 patients were included in the study. Of these 21 (46.7%) were male and 24 (53.3%) female. The patients started the diet at a mean age of 11.9 (5.8) months. Their seizures started at a median of 3.5 (5) months. At diet initiation they were taking a median number of 3 (2) AED’s. The other patient characteristics regarding prematurity, adequate growth, ethnicity, etiology, electro-clinical syndrome, manner of feeding (at start) and reason for stopping the diet are included in Table 3.
The patient characteristics show a very young group of patients whom started the ketogenic diet before the first year of age. Most patients were not premature and had no growth retardation at start. They followed the ketogenic diet for a period of 7 months (range 4-16).
Table 3. Patient characteristics
Number of patients in study 45
Gender
- Male
- Female
21 (46.7%) 24 (53.3%)
Median age of seizure onset in months (IQR) 3.5 (5)
Median number of AED’s at start (IQR) 3 (2)
Mean age at start of ketogenic diet in months (SD) 11.9 (5.8)
Median time on diet in months (IQR) 7 (13)
Prematurity - Yes - No - Unknown 6 (13,3%) 37 (82.2%) 2 (4.4%) Ethnicity - Caucasian - African - Asian - Indo-Mediterranean - Other 31 (68.9%) 3 (6.7%) 1 (2.2%) 7 (15.6%) 3 (6.7%) Growth retardation - Yes - No 13 (28.9%) 32 (71.1%) Etiology of epilepsy - Genetic - Structural/ metabolic - unknown 15 (33.3%) 10 (22.2%) 20 (44.4%) Electro-clinical syndrome: - Non-syndromic - Lennox-Gastaut - Dravet - West - Ohtahara 12 (26.7%) 1 (2.2% 1 (2.2%) 27 (60%) 4 (8.9%) Manner of feeding (at start)
- Oral - Partially tube - Fully tube 33 (73.3%) 5 (11.1%) 7 (15.6%) Reason stop diet
- Insufficient or no effect
- Medical reason
- No compliance/too restrictive
- Ceasing the diet after successful treatment
- Deceased
- Other reason
- Has not stopped yet
- Epilepsy surgery 14 (31.1%) 2 (4.4%) 3 (6.7%) 6 (13.3%) 6 (13.3%) 5 (11.1%) (17.8%) 1 (2.2%)
14
3.2 Number of dietary changes
3.2.1 Dietary changes in total follow up period
In total 454 dietary changes were made in the first six months after diet initiation. For these dietary changes 43 different determinants were listed. Most of the adjustments were made for the category Development with 31% (n=142). Although differences between groups were small, with the lowest being the category Efficacy with 21% (n=95). A mean number of 10.1 (SD 5.5) changes was made for each patient.3.2.2 Dietary changes within 0-3 months (T1) and 3-6 months (T2) after diet initiation
First three months after initiation
In T1, 65% (n=295) of the total changes (n=454) were made. As seen in Figure 2, the adjustments were spread almost equally between the four categories. The category Side effects required the most dietary changes, with 28% (n=84). A mean number of 6.6 (SD 3.6) changes were made for each patient.
Three to six months after initiation
In T2, the 35% (n=159) of all changes (n=454) were made. Figure 3 shows that Development was the largest category, with 40% (n=63). A median number of 3 (IQR 5) changes were made for each patient. This is half the number of changes in T1 per patient.
In summary, most of the dietary changes were made in T1. The category Development being the largest in the whole follow-up period and in T2. In T1, side effects of the diet were most often reason for dietary adjustment.
The determinants within the four categories will be described in the following section. This section will also include how often these determinants were reason for fine-tuning.
Figure 2. Total dietary changes per category in T1 (n=295) Figure 3. Total dietary changes per category in T2 (n=159)
Figure 1. Total dietary changes (n=454)
21% 25% 31% 23% Efficacy Side effects Development Other 23% 28% 27% 22% Efficacy Side effects Development Other 17% 17% 40% 26% Efficacy Side effects Development Other
15
3.3 Determinants for fine-tuning the diet
3.3.1 Efficacy
In 21% (n=95) of all the cases (n=454) the category Efficacy was reason for fine-tuning the diet. In T1 this was 23% (n=68) and in T2 this is 17% (n=27). There were three major reasons for adjusting the diet. Two of these reasons were too low ketosis and too high ketosis. The other main reason was the optimizing of ketosis. For instance, when ketosis was within the optimum range of 2-5 mmol/L, but a certain range was pursued for optimum efficacy (e.g. 4-5mmol/L) for the individual patient.
Ketosis too low
In 19% of the cases (n=11) the ratio was increased to fine tune low ketosis during T1. In 15% of the cases (n=4), the diet ratio was maintained but calories decreased during T2.
Ketosis too high
In both T1 and T2, when ketosis was too high the diet was fine-tuned by decreasing the ratio. In T1 this was done in 43% of the cases (n=29), in T2 this was done in 63% of the cases (n=17). In T2, decreasing the ratio was the only fine-tuning measurement that was applied.
Optimizing ketosis
In T1, 7% of the cases (n=5) the level of ketosis was optimized. Optimizing ketosis was only done in T1. Fluctuating ketosis levels were in 6% of the cases (n=4) reason for dietary adjustments in T1. In T2, this was 4% (n=1).
Overall, the majority of the changes for the category Efficacy were done in T1. Optimizing ketosis was more often done in T1 than in T2 as well. In T2 only decreasing the ratio was used when ketosis was too high.
Figure 4. Determinants within the category Effectivity
19% 7% 43% 63% 1% 7% 1% 3% 15% 7% 4% 4% 3% 4% 6% 4% 7% 0% 10% 20% 30% 40% 50% 60% 70%
Ketosis too low Ketosis too high Optimizing ketosis Ketosis too low Ketose too high Optimizing ketosis
T1 (n=68) T2 (n=27) N u m b er o f ch an es p er d eterm in an t in %
Increase ratio Decrease ratio
Add 5% energy as MCT Decrease 5% energy from MCT
Decrease 5-10% of calories maintaining diet ratio Increase 5-10% of calories, maintaining same diet ratio Start carnitine supplementation Other (hyperkytosis)
Other (hypoketosis) Fluctuating ketosis Optimizing ketosis
16
3.3.2 Side effects
In 25% (n=122) of all cases (n=454) the diet was fine-tuned to correct for side effects. In T1, the category was fine-tuned for in 28% of the cases (n=84). In T2, it made up for 17% of all the cases (n=28).
Figure 5. Determinants within the category Side Effects
Gastrointestinal
Over half of the side effects were related to gastrointestinal problems such as gastric acid, cramps, vomiting, diarrhea, reflux, nausea and obstipation. Obstipation was the most prominent factor for dietary changes. In T1 the diet was fine-tuned for obstipation in 26% of the cases (n=22), in T2 in 25% of the cases (n=7). Of all the changes (n=29) for the determinant obstipation the majority was done in T1 with 75% (n=22) and in T2 only 25% (n=7) was done.
Other side effects
For the other side effects weight loss was most prominent in T1 and reason for dietary changes in 25% (n=21) In T2 insufficient growth was seen more often and was reason for dietary changes in 29% (n=8) In short, the category Side effects was more often reason for dietary adjustments in T1 than in T2. The percentage of the total changes decreased from 28% in T1 to 17% in T2. This is a decrease of 11%. Within the category side effects obstipation was a prominent determinant related to the gastrointestinal problems. Weight loss and insufficient growth were also visible determinants
7% 26% 25% 18% 7% 2% 4% 6% 1% 5% 4% 11% 10% 29% 25% 21% 0% 10% 20% 30% 40% 50% 60%
Gastro-intestinal Other side effects Gastro intestinal Other side effects
T1 (n=84) T2 (n=28) N u m b er o f ch an ge s p er d eterm in an t in %
Hypoglycemia Obstipation Vomiting
Nausea Diarrhea Kidney stones
Reflux Gastric acid Cramps
17
3.3.3 Development
In 31% (n=141) of all the cases (n=454) the diet was fine tuned for development. In T1 27% of all the changes (n=78) were made for the category Development. In T2, with 40% (n=63) Development was the largest category as reason for fine-tuning the diet.
Figure 6. Determinants within the category Development
Within the category Development the diet was most often fine-tuned for the determinant growth. In T1 57% (n=45) and in T2 52% (n=33) of the changes were for the determinant growth.
Besides the determinant growth, the determinant development was also often reason for dietary adjustments. In T1 it made up 23% (n=18) of dietary changes in this category. In T2 this was 27% (n=17). Fine-tuning the diet for the determinant prolonging overnight fast was mostly done in T1, 11% (n=9). In T2 it was 5% (n=2) reason for dietary adjustment in this category.
The determinant increased physical activity was only seen in T2 and was in 3% (n=2) of changes reason for dietary adjustments.
To summarize, growth was most often reason to fine-tune the diet within the category Development. Although, the determinant development was also prominent. Prolonging the overnight fast was mostly done in T1. Whilst, increased physical activity was only seen in T2.
7,6% 12,7% 3,2% 11,4% 4,8% 22,8% 27,0% 57,0% 52,4% 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% T1 (n=78) T2 (n=63) N u m b er o f ch an ge s p er d eterm in an t in %
18
3.3.4 Other
In 23% (n=106) of all cases (n=454) the diet was fine-tuned based n reasons that are part of category other. In T1 in 22% (n=65) of the cases and in T2 28% (n=41).
Figure 7. Determinants within the categoryOther
Illness was most often reason for dietary adjustment in this category. In T1, 54% (n=35) of the dietary changes were made for illness and in T2 49% (n=20). Other determinants that stand out are rejecting food, food saturation, hungry and operation.
In 17% (n=11) of the cases the diet was fine-tuned for the determinant rejection of food, in T1. In T2 this was 15% (n=6). In some cases, food was too much, this was scored as food saturation. This led to less to inadequate food intake. In T1 6% (n=4) of the changes for this category were made for food saturation. In T2 it made up 7% (n=5) of dietary changes.
On the other hand, in some cases the diet was fine-tuned because the patients were hungry. In T1, 6% (n=6) of dietary changes in this category were made for the determinant hungry. In T2 this was 7% (n=3). In T1 tube feeding was also reason for dietary changes in 8% (n=5). In some cases, a tube was placed when patients were ill or if they were not able to achieve adequate dietary intake by themselves. In T2 operation was also reason for dietary changes in 12% (n=5) for the category other.
1,5% 1,5% 2,4% 53,8% 48,8% 1,5% 2,4% 12,2% 16,9% 14,6% 1,5% 7,7% 4,9% 6,2% 7,3% 9,2% 7,3% 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% T1 (n=74) T2 (n=44) N u m b er o f ch an ge s p er d eterm in an t in %
Blood levels Coughing (vomiting) Illness Insufficient fluid Operation (OK) Rejecting food Tired (crying) Tube feeding Urine levels too high Food saturation Hungry
19 In summary, in this category the diet was most often fine-tuned for illness. In both T1 and T2 it takes up about 50% of the changes. The diet was fine-tuned for the determinants hungry, tube feeding, food saturation, rejection of food and coughing (vomiting) to maintain adequate dietary intake. Together, in T1 they make up 42% (n=27) and in T2 32% (n=13)
In the following paragraphs the top four determinants will be shown within the different time periods: whole follow-up period, short-term (T1) and long-term (T2). The results of the chi-square test will be shown to answer the question whether there was a significant difference in the number of dietary changes between T1 and T2.
3.3.5 Top four determinants
To specify which specific determinants were most often reason for fine-tuning the diet a top four list was composed. In total 49% (n=223) of all the dietary changes (n=454) were made for the top four determinants: growth (17%), illness (12%), high ketosis (12%) and development (8%). In Figure 8, the rest of the determinants were gathered in the group other (51%).
Figure 8. Top four dietary changes total (n=454)
In T1 approximately 50% of the changes were made for the top four variables (Figure 9): growth (15%), too high ketosis (13%), illness (12%) and obstipation (7%).
In T2 the top four determinants made up 55% of all changes (Figure 10). The first three reasons are the same as in T1: growth (21%), Illness (12%) and too high ketosis (11%). The other determinant was development (11%). The other determinants were gathered in the group other. In T1, this was 53% and in T2, this was 45%.
A total list containing all the determinants for dietary changes and how often these occurred are included in Appendix 4.
Figure 9. Top four dietary changes in T1 (n=295) Figure 10. Top four dietary changes in T2 (n=159)
15% 13% 12% 7% 53% Growth High ketosis Illness Obstipation Other 21% 12% 11% 11% 45% Growth Illness High ketosis Development Other 17% 12% 12% 8% 51% Growth Illness
ketose too high (>5 mmol/L) Development
20 To sum up, the determinants growth, illness and high ketosis are seen throughout the whole follow-up period as most prominent determinants. In both T1 and T2, in 12% illness was reason for fine-tuning of the diet (T1 n=35, T2 n=20). Whilst high ketosis was more often fine-tuned for high ketosis and growth more often in T2. The percentage for the top four determinants was higher in T2 (53%) than in T1 (47%).
3.3.5.1 Chi-square test frequency
Since the first three determinants were the same in T1 and T2. Only the top three determinants were used in the chi-square test to test for difference in frequency. There was no significance in the number of changes between short and long-term for the determinant growth. There were significant differences in the number of changes between short and long-term for the determinants illness and high ketosis. In Table 4 the number of changes and p-values were noted.
Table 4. Number of dietary changes
Short-term (T1) Long-term (T2) Total P-value
Growth 45 33 78 P=0.174
Illness 35 20 55 P=0.043
High ketosis (>5mmol/L)
37 17 54 P=0.006
The previous paragraph has shown that most diet adjustments are done to fine-tune the diet for growth in both short and long-term. The determinants illness and high ketosis were more often reason for fine-tuning the diet during the first period after diet initiation.
In the following paragraph the effectiveness of the dietary changes for the top four determinants, in both short and long-term, will be presented.
21
3.4 Effectiveness dietary changes
Throughout the whole follow-up period 73% (n= 333) of all the changes (n=454) were effective, 24% (n=110) was not effective and 3% (n=11) was unknown.
In T1 (n=295), 71% (n=210) of the changes were effective, 28% (n=83) not effective and 1% (n=2) unknown. Figure 11 shows the effectiveness of the dietary changes for the top five determinants wherefore dietary adjustments were made during T1.
In T2 (n=159), 77%(n=123) of the changes were effective, 17% (n=27) not effective and 6% (n=9) unknown. Figure 12 shows the effectiveness of the dietary adjustments for the top five determinants during T2.
Figure 11. Effectiveness of dietary changes in T1
Figure 12. Effectiveness of dietary changes T2
Effective ; 85% Effective ; 95% Effective ; 35% Effective ; 88% Not effective ; 3% Not effective ; 5% Not effective ; 65% Not effective ; 6% Unknown; 18% Unknown; 6% 0 5 10 15 20 25 30 35
Growth Illness High ketosis Development
N u m b er o f d ie ta ry ch an ge s Effective; 93% Effective ; 94% Effective ; 49% Effective ; 59% Not effective; 7% Not effective ; 6% Not effective ; 37% Not effective ; 41% Unknown;14% 0 5 10 15 20 25 30 35 40 45 50
Growth Illness High ketosis Obstipation
N u m b er o f d ie ta ry ch an ge s
22
3.4.1 Chi-square tests effectiveness
In both short and long-term, the top four determinants were tested for effectiveness. In short-term the top four determinants were growth, illness, high ketosis and obstipation. In long-term the top four determinants were growth, illness, high ketosis and development.
Within three months after diet initiation (T1)
The determinants growth and illness showed a significant difference in effectiveness. The determinants high ketosis and obstipation did not show a significant difference in effectiveness. The number of effective, not effective and unknown changes per determinant and p-values were included in Table 5.
Table 5. Number of dietary changes sorted by effectiveness in T1
Effective Not effective Unknown Total P-value
Growth 42 3 0 45 P=0.00
Illness 33 2 0 35 P=0.00
High ketosis 21 16 0 37 P=0.411
Obstipation 13 9 0 21 P=0.394
Within three to six months after diet initiation (T2)
The determinants growth, illness and development showed a significant difference in effectiveness. The determinant high ketosis did not show a significant difference in effectiveness. The number of effective, not effective and unknown changes per determinant and p-values were included in Table 6.
Table 6. Number of dietary changes sorted by effectiveness in T2
Effective Not effective Unknown Total P-value
Growth 28 1 4 33 P=0.00
Illness 19 1 0 20 P=0.00
High ketosis 6 11 0 17 P=0.225
Development 10 1 1 12 P=0.001
Overall, 74% of all the changes were effective in the whole follow-up period. As shown in Figure 10 and 11 the dietary changes for growth, illness and development were often effective. The chi-square test shows a strong significant difference. The dietary changes for determinants high ketosis and obstipation had limited effectiveness. There was no significant difference in effectiveness for these determinants.
23
3.4.2 Dietary changes
The specific dietary changes will be discussed for the determinants that showed significance for effectiveness.
Growth
First 3 months after diet initiation (T1)
In T1 93% (n=42) of the changes for the determinant growth were effective. An average effective change was calculated for this determinant. This was adding 6% of calories with a protein ratio of 2.3 grams of protein/100 calories. The amount of food was increased depending on body weight. Either grams of formula were increased, or extra complementary food was advised.
In 7% (n=3) changes were not effective. Of these three changes the average change was adding 2.6% of calories and decreasing the grams of protein with 2.9g /100 calories.
3-6 months after diet initiation (T2)
In T2, the diet was adjusted 33 times in total for the determinant growth. Of these 85% (n=28) were effective. The effective changes had an average of adding 7.8% of calories with a protein ratio of 2 grams of protein /100 calories. The specific dietary changes were the same as in T1. Either grams of the diet formula were increased, or extra grams of complementary food was advised.
In 3% (n=1) the dietary adjustment was not effective. The ketodrink was reintroduced into the diet, but was not effective. A ketodrink is an emulsion of Liquigen©© and sugar and fat free yoghurt drink. In 12% (n=4) the effect was unknown. No follow up was available in the patient case file.
Illness
First 3 months after diet initiation (T1)
In T1, 94% (n=33) of the changes for the determinant illness were effective. During illness the emergency plan was implemented. Some patients (n=2) were hospitalized during sickness to ensure close monitoring. When a patient became ill during initiation, the build-up was done slowly (n=3). Patients stayed on the same diet a day longer, instead of increasing the amount of fat daily. All three dietary changes were effective. In one case (n=1) during this study, food was stopped, and an infusion was placed. In only two cases (n=2) the dietary changes were not effective. Of these, one patient had persistent illness. The other patient had experienced long-term behavioral changes. In both cases the emergency plan was used to change the diet.
3-6 months after diet initiation (T2)
In total 20 changes were made in T2 for the determinant illness. Of these 95% (n=19) were effective. Only 5% (n=1) was not effective as palliative care was started for this patient.
In T2 the individualized emergency plan was used most often as dietary change for illness. An example of an emergency plan is included in Appendix 7. Twice patients were hospitalized because of illness. One patient stopped the diet because of illness and one other patient only received extra fluids as dietary change for illness.
Development
In T1, the determinant development was not in the top four and therefore not included in the square test. In T2, the determinant development was part of the top four and included in the chi-square test and showed significance. Therefore, only the specific dietary changes for T2 were discussed.
24 3-6 months after diet initiation (T2)
The diet was adjusted 12 times for the determinant development. Of these, 83% (n=10) were effective, 8% (n=1) was not effective and 8% (n=1) was unknown. The changes for development mostly entailed introducing complementary foods. Complementary foods were vegetable bites, the ketomuffin or pancake and bread. To introduce more carbohydrates into the diet MCT fat was also introduced with these dietary changes. In the one case the change was not effective the patient did not like the vegetables and potatoes. The complementary food was changed to a ketomuffin, which was an effective additive.
25
4. Discussion
In total 45 infants with intractable epilepsy starting the ketogenic diet at the Erasmus MC Sophia’s Children’s Hospital were followed three to six months after diet initiation. This study was set up to determine what determinants were reason for dietary changes, including those already found in literature. The effectiveness of the dietary changes was also included in this study. As stated in the results the top three determinants were growth, illness and high ketosis. The numbers four were obstipation (in T1) and development (in T2). The fact that illness is part of the top three determinants highlights the vulnerability of these young infants (27).
This study has exposed a couple of remarkable findings regarding dietary changes, determinants, effectiveness and study design. These will be further discussed in this chapter.
Findings regarding dietary changes
The dietary changes for growth, illness and development are in line with the ketogenic guidelines for infants. In these guidelines a result of Nation (2014) was mentioned. In Nation (2014) it is suggested that a protein-to-energy ratio of at least 1.5g/100 calories would prevent growth faltering. The findings in this study seem to concur with this suggestion. The effective changes in T1 had an average protein-to-energy ratio of 2.3g of protein/100 calories. Whilst the not effective changes in T1, decreased the grams of protein with 2.9 grams of protein/100 calories. However, the calories were increased with 6% in the effective changes and the not effective changes only increased the calories with 3%.Therefore, it cannot be determined if the effectiveness of the dietary changes is a result of the protein-to-calorie ratio or the increased number of calories.
The dietary changes for development were mostly effective. The diet is slowly expanded with complementary foods, usually starting with a vegetable or fruit bite. Later this is expanded with a ketomuffin or pancake in preparation to start bread. In T1, twice the dietary changes were not effective because the patients were admitted to the hospital for unexpected surgery. This demonstrates the vulnerability of these patients and the importance of the individual emergency plan. The emergency plan is used frequently and is most often effective.
However, the changes for gastro-intestinal side effects were not in line with the ketogenic guidelines. The guidelines mention dietary manipulation as remedy for gastro-intestinal problems. Whilst in practice, medication is mostly used as effective change for gastro-intestinal problems such as obstipation, vomiting, reflux, gastric acid and general gastric problems. For cramps, the medical formula Omneo is used as effective remedy. Omneo is an infant formula specially developed for infants with cramps (28). To increase effectiveness of the fine-tuning for gastro-intestinal problems the start of medication could be implemented as primary treatment.
Remarkable determinants
The finding of the determinant in the top four determinants in short-term is consistent with Hartman and Vinning (2007), whom stated: “Patients on the MCT diet are more likely to experience abdominal bloating and diarrhea than those on the LCT diet, which is believed by some to be less palatable than the MCT diet. Patients on the LCT diet are more prone to obstipation than those consuming a MCT diet.” (17). This statement can be explained by the fact that the diet in infants is started with the Classical version of the ketogenic diet, containing mostly LCT fats. Whilst on the diet, MCT fat is slowly introduced when infants start with complementary food. The amount of MCT fat is increased when complementary foods are expanded. The MCT fats allow for extra carbohydrates in the diet. Because MCT fat is more ketogenic, less fat is needed to maintain the same level of ketosis. However, because MCT fat is reason for more gastro-intestinal problems infants use MCT fat in small quantities. In long-term, 51% of the patients used an average of 29 ml Liquigen©© (a MCT emulsion). Therefore, it is not likely that the gastro-intestinal problems in long-term are caused by the use of MCT fat. This corresponds with the data found in this study. The data shows that the gastro-intestinal problems were seen when patients were teething or when laxatives were instated.
26 Another determinant that stood out was the determinant kidney stones, or actually the lack of this determinant in this study. Eun et al (2006), Nordli et al (2001), Hong et al (2010), Numis et al (2011) and Dressler et al (2015) indicate that kidney stones were a side effect of the ketogenic diet (18,19,20,21,22). In this study the determinant kidney stones was not once reason for dietary adjustments. A possible explanation for this might be that the Erasmus MC Sophia closely monitors fluid intake. When dietary advice concerning daily fluid intake is not achieved, adjustments are made. The implementation of close monitoring specifically for fluid intake, in all centers and hospitals treating with the ketogenic diet, might reduce overall prevalence of kidney stones.
Findings regarding effectiveness
The average number of changes per patient decreased from T1 (6,6) to T2 (3,5). Also, the effectiveness increased in T2 relative to T1. In T1, 71% of the changes were effective, in T2 this was 77%. This result may be explained by the fact that the diet is very complex, and alterations must be made on an individual level. Although dietary changes are based on the international guidelines for infants with intractable epilepsy, this does not necessarily mean that each change will be effective for each patient. However, a tailor-made diet is not feasible in practice. The alternative could be to make a do’s and don’ts list for the dietitians based on the effectiveness of the dietary changes found in this study. There was another remarkable finding which was not included in the results. Patients whom stopped the diet because there was insufficient or no effect (n=14) had an average number of changes of 4.6, in T1. This is less than the average amount of changes in T1, which was 6.6. The patients whom stopped the diet after effective treatment (n=6) had an average number of changes 7.5, in T1. This is higher than the average amount of changes. Dietary changes for the patients whom stopped the diet after effective treatment, were more often not effective. In total 38 changes were made for these 6 patients. Of these 38 only 66% was effective. Compared to the group whom stopped because of insufficient or no effect, wherefore 65 changes were made in total and 89% was effective.
Hence, it could be hypothesized that the number of dietary changes and effectiveness has no influence on the outcome of efficacy. However, the number of dietary changes and effectiveness could influence compliance. The data of this study shows that the three patients whom stopped the diet because there was no compliance had an average number of changes of 8 in T1. In total 24 changes were made, of these 62.5% was effective. By increasing the effectiveness of the dietary changes, the number of dietary changes could be decreased. Overall, the ketogenic diet is a very strict diet and takes a lot of effort to maintain (1). With less dietary changes the diet might be experienced as less burdensome and compliance could increase.
Study design
In retrospective studies it is common to encounter unclear or incomplete data, this is a weakness of this study design. Also, the scoring of the determinants and effectiveness relied on interpretation. There was no set time period for scoring the effectiveness of the dietary changes. This was done to include as much data as possible. At the start of the diet ketose levels are checked regularly, when stable it is checked less frequent. Growth and development are done on each patient’s individual pace. Therefore, setting a certain time period could lead to missing data. Another strength of the study was to check several random case report forms on a regular basis by peer students. This was done to limit misinterpretation and missing data.
27
5. Conclusion and recommendations
The main research question of this thesis was: What determinants were reasons for dietary adjustments and how effective are these adjustments in short (0-3 months) and long-term (3-6 months) after the initiation of the ketogenic diet in infants 0-2 years of age treated for intractable epilepsy in the Erasmus MC Sophia Children’s Hospital?
Growth, illness and high ketosis were the top three determinants wherefore the diet was adjusted in both short and long-term. There was a significant difference in effectiveness for the adjustments growth (p=0.00) and illness (p=0.00). Overall the percentage of effective changes increased between T1 and T2 with 5%. In T1, 72% of the dietary changes were effective and in T2, 77% were effective. Comparing this study to practice, the following outcomes were notable. The outcomes of this study substantiate the expectancy that development, in infants, is often reason for dietary changes. Overall, the category Development was most frequently adjusted for with 31% (n=141). The determinants growth and development stand out in this category, being part of the top four. It was found that a protein-to-calorie ratio of 1.5g protein / 100 calories could be important to prevent growth faltering. Also, it was found that the weaning plan for development as described in the ketogenic guidelines for infants with intractable epilepsy was effective. By implementing the protein-to-calorie ratio more strictly in the future, growth faltering could be reduced. Although, growth and development are part of the top four determinants, it was not assessed in this study whether growth and development were adequate. More research needs to be done if growth and development are adequate whilst on the diet and after stopping the diet.
Another outcome that was notable regards the fine-tuning for gastrointestinal side effects. In short-term the category Side effects was most often reason for dietary adjustments. Instead of dietary changes, medicine was most often used as effective treatment for gastrointestinal problems. Implementing the use of medicine as primary treatment could possibly improve the overall treatment of the diet. Less dietary changes have to be made, which could be beneficial for the patient and reduce stress of the parents. Besides this, time and attention can be dispersed to other aspects of the diet such as efficacy.
The current data highlights the importance of close monitoring and dietary changes to pursue maximum efficacy, with the least amount of side effects and best development possible in this vulnerable group of patients. This study shows that with close monitoring it is possible to meet all the requirements these infants need, despite the complexity of the diet. By further implementing the findings of this study, effectiveness of the diet could be increased. Increased effectiveness of the diet could lead to higher compliance.
28
6. References
1. Marsh E, Freeman J, Kossof E, Vining E, et al. The outcome of children with intractable seizures: a 3- to 6-year follow-up of 67 children who remained on the ketogenic diet less than one year. Epilepsia. 2006 Februari; 47(2):425-30.
2. Robert, M. Anderson Jr. Epilepsy. Practitioner’s Guide to Clinical neuropsychology. 1994. p248-60.
3. Poonam B, Filippi D, Hauser A. The descriptive epidemiology of epilepsy – a review. Epilepsy Res. 2009 July; 85(1):31-45
4. Fisher R, Cross J, French J, Higurashi N, et al. Operational classification of seizure types by the international leage against epilepsy: position paper of the ILEA commission for classification and Terminology. 2017 April;58(4):522-530
5. List of anti-epileptic drugs. Epilepsy society. [datum onbekend].
https://www.epilepsysociety.org.uk/list-anti-epileptic-drugs#.Wbpmc6i0PIU
6. Löscher W. Animal models of intractable epilepsy. Progress in Neurobiology. Elsevier Science 1997 October;53(2):239-58
7. Rho J. How does the ketogenic diet induce anti-seizure effects?. Neuroscience letters. 2017 Januari 10;637:4-10
8. Ketogeen dieet. Epilepsie fonds. [datum onbekend].
https://www.epilepsie.nl/over-epilepsie/pagina/102-2/ketogeen-dieet/
9. Van der Louw E, van den Hurk D, Neal E, Cross H, et al. Ketogenic diet guidelines for infants with refractory epilepsy. 2016
10. Liu Y C, Wang H. Medium-chain triglyceride ketogenic diet an effective treatment for drugresistant epilepsy and a comparison with other ketogenic diets. Biomed. 2013 Januari Februari;36(1):9-15
11. Kossoff E, Zupec-Kania B, Amark P, Bllaban-Gil K, et al. Optimal clinical management of children receiving the ketogenic diet: recommendations of the international ketogenic diet study group. Epilepsia 2009;50(2):304-317
12. Groesbeck DK, Bluml RM, Kossoff EH. Long-term use of the ketogenic diet in treatment of epilepsy. Developmental medicine and child neurology. 2006;48:978-981
13. Kang H, Chung D, Kim D, Kim H. Early- and late-onset complications of the ketogenic diet for intractable epilepsy. ILEA. Epilepsia. 2004 May 9;45(9):1116-23
14. Neal E, Chaffe H, Scwartz R, Lawon M, et al. The ketogenic diet for the treatment of childhood epilepsy: a randomized controlled trial. Lancet neurology. 2008;7:500–06
15. World health organization growth standards. World Health Organization. [datum onbekend]. http://www.who.int/childgrowth/standards/en/
16. Standardized residuals. Statistics How To. 2017.
http://www.statisticshowto.com/what-is-a-standardized-residuals/
17. Hartman A, Vinning E. Clinical aspects of the ketogenic diet. ILEA. Epilepsia. 2007;48(1):31-42 18. Eun S, Kang H, Kim D, Kim H. Ketogenic diet for treatment of infantile spasms. Brain and
Development. 2006;28:566-71
19. Nordli D jr, Kurode M, Caroll J, Koenigsberge D, et al. Experience with the ketogenic diet in infants. Pediatrics. 2001;108:129-33
20. Hong A, et al. Infantile spasms treated with the ketogenic diet: prospective single-center experience in 104 consecutive infants. Epilepsia. 2010;51(8);1403-7
21. Numis A, et al. The relationship of ketosis and growth to the efficacy of the ketogenic diet in infantile spasms. Epilepsy Res. 2011;96(1-2):172-5
29 22. Dressler A, et al. The ketogenic diet in infants – Advantages of early use. Epilepsy res.
2015;116:53-8
23. Neal E, et al. Growth in children on classical and medium-chain triglyceride ketogenic dies. Pediatrics. 2008;122(2):e334-40
24. Kossoff E, et al. A case-control evaluation of the ketogenic diet versus ACTH for new-onset infantile spasms. Epilepsia. 2008;49(9):1504-9
25. Insel P, Ross D, McMahon K, Bernstein M. Nutrition. Fifth edition. Jones and Bartlett learning. 26. Nation J, et al. Linear growth of children on a ketogenic diet: does the protein-to-energy ratio
matter? Journal of child Neurology. 2014;29(11):196-501
27. Infants and Toddlers Are Particularly Vulnerable: Good Child Care and Early Education Can Play a Vital Role in Their Development. Children’s defense fund. 2013.
28. Nutrilon Omneo-Comfort 1. Nutricia. [datum onbekend].
30
7. Appendix
Appendix 1. Fine-tuning measurements
Table 7. Fine-tuning the diet (9)
Note: Adjust the diet only once a week
In general - Check for incompatible medication: (i.e. carbohydrate-containing vitamin D-supplement, multivitamin supplement, antibiotics) - Check calculated recipes
- Check measuring scale is accurate Ketose is too low
(<2 mmol/L)
- Increase ratio
o Stop adding breastmilk or infant formula: increase amount of ketocal formula used with same calories.
o Mix with ketocal 4:1©
▪ 25-75% ▪ 50-50%
o Switch to only ketocal 4:1© (check if all nutrients are
covered)
o Increase amount of fat used in solid food: i.e. butter, margarine, oil, cream, Calogen©, mayonnaise.
- Add 5% energy as MCT to the diet or increase the MCT used by 5% - Decrease 5-10% of calories maintain diet ratio. If this doesn’t help
measure resting energy expenditure.
- Start Carnitine supplementation (start with 10mg/kg, build up to maximum 50mg/kg based on tolerance).
Ketose is too high (>5 mmol/L)
- Decrease ratio
o Add 1-2% protein (using standard infant formula or protein enriched powder)
o Add 1-2% carbohydrate (using carbohydrate polymer or standard infant formula to increase amount of fruit, bread, potato, cereals).
- Decrease 5% energy from MCT used and compensate calories from LCT by increasing amount of LCT emulsion (Calogen©, Carbzero©)
cream, oil and margarine.
- Increase 5-10% of calories, maintaining same diet ratio. If this doesn’t help, measure resting energy expenditure.
31
Appendix 2. Indications and contra-indications for the ketogenic diet in infants
Table 8. Indications/contra-indications for ketogenic diet in infants (9)
Indications for KD treatment: Epilepsy
- Refractory epilepsy after use of 2 AED’s o West syndrome
o Ohtahara syndrome - Adverse effects of AEDs - Waiting for epilepsy surgery Metabolic diseases:
- GLUT-1 deficiency - PDHC- deficiency
Optional:
- Mitochondrial diseases
Contra-indications for KD treatment
Absolute
- Fatty acid oxidation deficiencies (VLCAD, LCHAD, MCAD, OCTN2, CPT1, CPT2) - Pyruvate carboxylase deficiency and other gluconeogenesis defects (fructose 1,6
diphosphatase deficiency)
- Glycogen storage diseases (except type 2) - Ketolysis defects
- Ketogenesis defects - Porphyria
- Prolonged QT syndrome or other cardiac diseases - Liver, kidney or pancreatic insufficiency
- Hyperinsulinism Relative
- Inability to maintain adequate nutrition
- Surgical focus identified by neuroimaging and video EEG monitoring - Parent or caregiver noncompliance
- Growth retardation
32
Appendix 3. Case report form
PID no. Nr in study
Name
Gender 0=male 1=female
Date of birth
Premature 0= No 1 = Yes
Growth retardation 0= No 1= Yes
Ethnicity 1= Caucasian 2= African 3= Asian 4= Indo-Mediterranean 5= Australian-aboriginal-pacific 6= Other 7= Unknown Start of epilepsy Start of KD
Stop of KD/ Last follow-up
Way of feeding 0 = oral
1 = partially tube 2 = fully tube
Reasons for discontinuation 0= insufficient or no effect 1=medical reason
2=adverse effects/illness: 3=no compliance/too restrictive
4=ceasing the diet after successful treatment 5=deceased
6=lost to follow-up 7=other reason: 8=Has not stopped yet
9= epilepsy surgery
Deceased 0=no 1=yes: ..-..-…. Cause:
Amount of seizure types
0= one seizure type 1=multiple seizure types
Underlying type of cause 0=genetic
1=structural/metabolic 2=unknown cause
Cause of metabolic disease 0=no/unknown
1=GLUT-1 def 2=PDHC def 3= mitochondrial 4=other
33
Causal diagnosis 0= Non syndromic
1= Lennon Gastaux syndrome 2= Dravet
3=West 4 = Ohtahara 5 = MEI 6 = Unknown
Determinants present at baseline (T0)
Remarks
Determinants present 0= No 1= Yes
Determinants VP Medication Remarks
Side effects Obstipation Vomiting Nausea Diarrhea Kidney stones Reflux Gastric acid Development Teeth Growth Weight loss Insufficient growth Other Illness Tube feeding Coughing (vomiting) Rejecting food Food saturation Tired (crying) Hungry
34
Dietary changes within 3 months (T1)
Determinants present 0= No 1= Yes
Effect 0 = No effect
1= Effective 2=
Unknown
Determinants VP Effect Medication Remarks
Effectivity
Ketose too low (<2 mmol/L) Increase ratio
Stop adding breastmilk or infant
formula
Mix with Ketocal 4:1©
Switch to Ketocal 4:1 ©
Increase amount of fat used in solid
food
Add 5% energy as MCT to the diet or
increase the MCT used by 5%
Decrease 5-10% of calories
maintaining diet ratio
Start carnitine supplementation
ketose too high (>5 mmol/L)
Decrease ratio
Decrease 5% energy from MCT used and compensate calories from LCT by increasing LCT emulsion, cream, oil,
margarine
Fluctuating ketosis
Optimizing ketosis
35 Determinants present 0= No 1= Yes
Effect 0 = No effect 1 = Effective 2 = Unknown
Determinants VP Effect Medication Remarks
Side effects Hypoglycemia Obstipation Vomiting Nausea Diarrhea Kidney stones Reflux Gastric acid Weight loss Insufficient growth Development Teething
Increased physical activity (higher
energy expenditure)
Development
Growth
Prolonging night rest
Other Illness Tube feeding Coughing (vomiting) Rejecting food Food saturation Tired (crying) Hungry Blood levels Urine levels Insufficient fluids Remarks
36
Dietary changes 3-6 months (T2)
Determinants present 0= No 1= Yes
Effect 0 = No effect 1 = Effective 2 = Unknown
Determinants VP Effect Medication Remarks
Effectivity
Ketose too low (<2 mmol/L) Increase ratio
Stop adding breastmilk or infant
formula
Mix with Ketocal 4:1©
Switch to Ketocal 4:1 ©
Increase amount of fat used in solid
food
Add 5% energy as MCT to the diet or
increase the MCT used by 5%
Decrease 5-10% of calories
maintaining diet ratio
Start carnitine supplementation
ketose too high (>5 mmol/L)
Decrease ratio
Decrease 5% energy from MCT used and compensate calories from LCT by increasing LCT emulsion, cream, oil,
margarine
Fluctuating ketosis
Optimizing ketosis
37 Determinants present 0= No 1= Yes
Effect 0 = No effect 1 = Effective 2 = Unknown
Determinants VP Effect Medication Remarks
Side effects Hypoglycemia Obstipation Vomiting Nausea Diarrhea Kidney stones Reflux Gastric acid Weight loss Insufficient growth Development Teething
Increased physical activity (higher
energy expenditure)
Development
Growth
Prolonging night rest
Other Illness Tube feeding Coughing (vomiting) Rejecting food Food saturation Tired (crying) Hungry Blood levels Urine levels Insufficient fluids Remarks
