Sage Journals: Discover world-class research

Abstract

Background

Several studies propose that brain energy deficit might be partially involved in the pathophysiology of migraine. Previously, studies demonstrated that ketogenic diet causes a substantial reduction in migraine frequency. Since the ketogenic diet is restricting and its adherence is difficult, we proposed to supplement ketone bodies exogenously to provide a prophylactic effect in migraineurs.

Aim

To evaluate the prophylactic effect of exogenous DL-beta-hydroxybutyrate supplementation in episodic migraineurs.

Methods

A double-blind, placebo-controlled, randomised crossover trial was conducted, involving 41 patients with episodic migraine. Patients were randomised 1:1 into placebo or beta-hydroxybutyrate group before entering the first treatment period. Each treatment period was 12 weeks long, followed by four weeks of washout phase and four weeks of run-in phase before entering into the corresponding second treatment period. The primary endpoint was the number of migraine days in the last four weeks of treatment, adjusted for baseline.

Results

We observed no clinically significant amelioration of migraine frequency or intensity under DL-beta-hydroxybutyrate treatment as compared to placebo regarding number of migraine days (mean difference [95% CI]: −1.1[−5.07, 2.85]), migraine intensity (0–10 VAS: 1.5[−0.8, 3.7]).

Conclusion

The selected dose of supplemented exogenous DL-beta-hydroxybutyrate did not demonstrate efficacy in episodic migraineurs.

ClinicalTrials.gov Identifier: NCT03132233

Keywords

Migraine prophylactic therapy randomised clinical trial exogenous ketone bodies ketone salt

Introduction

Migraine is a debilitating disorder characterised by varying symptoms that severely affect a patient’s quality of life. It is reported as one of the leading global disability causes among under-50-year-olds (1). Migraine is prevalent in up to 16% of the global population (2). Whilst a combination of genetic, environmental and lifestyle factors may interact and determine migraine manifestation, its pathophysiology is yet uncertain, and consequently migraine treatment and prevention remain challenging. Certain lines of research suggest brain energy abnormalities are possibly part of migraine pathophysiology (3 –7). Particularly, several reports showed reversible mitochondrial dysfunction in migraineurs (6,8,9). For instance, results of treatment with riboflavin and coenzyme Q10 indicated a protective effect in migraineurs by having a positive effect on energy metabolism (6). Furthermore, increased lactic and pyruvic acid markers were detected in migraine patients, both markers of mitochondrial disease (10). Observation of ³¹P-MRS patterns and COX-negative fibres in some migraineurs are consistent with what is reported in mitochondrial diseases (5). The origin of migraine headache is currently understood as the activation and sensitisation of the trigeminal pain pathway (11).

Current migraine prophylactic agents show substantial side effects making patients cease prescribed therapy. Additionally, most of these agents’ mode of action is not fully understood. Therefore, an innovative prophylactic treatment approach is needed.

The ketogenic diet (KD) has been reported to be an effective preventive treatment for migraine by reducing migraine frequency as well as intensity (12,13). KD consists of high fat, adequate protein and low-carbohydrate intake, shifting metabolism from glycolysis towards ketosis. This diet is reported to be an effective treatment for intractable seizures in children and adolescents (14). Elevated ketone body (KB) levels due to the KD are well tolerated, according to several reports (15,16). Despite promising results with KD treatment in migraineurs, the diet is demanding and long-term adherence to the dietary restriction and side effects is difficult to maintain.

This elicits the question if exogenously ingested KBs would induce a state of mild to medium nutritional ketosis and consequently attenuate migraine frequency. KBs have signalling functions in addition to metabolic actions that may also be of potential relevance in migraine. Based on animal models, there is evidence that KBs reduce the velocity of cortical spreading depression propagation (17) and modulate cortical excitability by heightening GABAergic and diminishing excitatory activities (18,19).

Exogenous KBs intake is well tolerated in humans and is prescribed as treatment in rare metabolic diseases such as multiple acyl-CoA dehydrogenase deficiency (20). The dietary supplementation with KBs, such as beta-hydroxybutyrate, demonstrated that mild ketosis (0.4–2 mmol/l) could be induced, irrespective of blood glucose level (20 –22). This method could be implemented as prophylactic migraine treatment by providing patients with ketogenic powder dissolved in water (consisting of a calcium–magnesium– DL-beta-hydroxybutyrate (βHB) salt ingestion three times a day). For this purpose, we aimed to examine the efficacy and safety of KB mineral salts as prophylactic treatment in episodic migraine patients implemented as a double-blind, randomised, placebo-controlled, trial with a crossover design. This is the first in human study exploring the effect of βHB as a prophylactic agent in episodic migraine patients with the aim to evaluate efficacy and safety.

Methods

Participants and setting

Patients were recruited via the Department of Neurology and the General Medicine Department of the University Hospitals in Basel (USB), Bern, Zurich and St. Gallen. Additionally, announcements were placed at the University Library and on the University of Basel webpages “Marktplatz”, dedicated to research studies, as well as the USB and the University Children’s Hospital websites. The trial was also advertised on local trams and buses. Recruitment started in May 2017 and ended in September 2019. All investigations were performed at the outpatient clinic of the USB, Switzerland.

Eligible patients fulfilled the following inclusion criteria: Age ≥18 and ≤65 years, episodic migraine with or without aura assessed by the international classification of headache disorders version 3 Beta Classification (ICHD-3) criteria, onset of migraine occurred before the age of 50 years; experienced 5 to 14 migraine days per 4 weeks, refrain from initiating/changing the type, dosage/frequency of any prophylactic medications and dietary supplements (such as Q10, riboflavin etc.) against migraine during the study and four months before study start, refrains from making any drastic changes to the diet for the duration of the study, including periods of fasting. Following exclusion criteria were defined: no requirement of oral/injectable steroids, current intake of simple analgesics or non-steroidal anti-inflammatory drugs more than 14 days per 4 weeks; takes triptans over 10 days per 4 weeks for headaches or other body pain; intake of prescription opioids, previous diagnosis of medication overuse headache, which has reverted to episodic migraine within the last 6 months, meets the ICHD-3 Beta Classification criteria for chronic migraine. An elaborate description of exclusion and inclusion criteria is provided in the published trial protocol (23).

Trial design

We conducted an investigator-initiated, double-blind, randomised, placebo-controlled trial with a crossover design. The trial was registered at ClinicalTrials.gov (NCT03132233), approved by the local Ethics Committee (EKNZ 2015-304) and the National Swiss Drug Agency (2016DR2109). All patients provided written informed consent of participation.

The trial consisted of a four week long run-in phase followed by a baseline visit, then a first treatment period of 12 weeks, followed by a washout period of four weeks, and a second run-in phase of four weeks continued with the second treatment period of 12 weeks. See Figure 1.

Figure 1.

CONSORT patient flow.

The trial design, the dose selection and justification of the washout period are detailed in previously published protocol (23).

Randomisation and blinding

Eligible patients were randomised after completing the prospective four-week baseline phase in a 1:1 ratio (verum-placebo/placebo-verum). The computer-generated randomisation list was uploaded into the electronic data capture software SecuTrial. Only pharmacy members of the University Hospital Basel and data managers of the clinical trial unit (CTU) had access to the unblinding list. Before coming to the baseline visit, a randomisation number was assigned automatically via SecuTrial. Both patients and investigators assessing outcomes were blinded to treatment allocation. The study medication was provided in similar-looking sachets. Emergency case reveal was implemented. The CTU provided the medication number of the complementing treatment arm of the first intervention period. This ensured that the trial staff stayed unblinded during the allocation of the second intervention period.

Medication

The investigational medicinal product (IMP) used in this clinical trial was βHB in powdered calcium (Ca²⁺)–magnesium (Mg²⁺)–salt form (Ca-Mg-βHB). The verum group received the daily dose of 9g Ca-βHB containing 7.54g of βHB and 1.47 g Ca²⁺ and was divided into 3 servings containing 2.51g βHB and 0.49g Ca²⁺ respectively. The daily dose of 9g Mg-βHB contained 6.6g βHB and 0.77g Mg²⁺and was also divided into 3 servings containing 2.2g βHB and 0.26g Mg²⁺. The mineral load determined the maximal IMP dose. Both IMPs were provided as a water-soluble powder. The placebo group received sachets containing Mannitol. It is used as the standard placebo substance by the pharmacy of the USB. Mannitol has the same texture and colour, and leads to similar potential side effects like gastrointestinal symptoms, as the IMP. The IMP and placebo were administered with glucose-free syrup to overshadow taste difference. The IMP was obtained from Ergomax (https://www.ergomaxsupplements.com) in bulk powder of GMP quality and packaged at Hänseler AG (Herisau, Switzerland).

Patients were asked to bring back the empty IMP sachets and the unused ones at each visit. All IMPs sachets were counted by study personnel. Numbers of used and unused sachets were registered in a table to control the compliance of patients.

Changes in study protocol

The study was initially designed as a randomised controlled parallel-arm trial. However, due to recruitment difficulties a crossover design with sample size re-estimation was implemented. Secondly, we removed the exclusion criterion history of thyroid disorder, since none of the patients that had finished the trial presented abnormalities of the thyroid hormones TSH, FT4 and FT3. The local ethics committee, the EKNZ and the National Swiss Drug Agency approved both amendments.

Clinical and demographic measures

After inclusion, patients were asked to fill out a daily migraine diary containing information about migraine frequency, intensity (1–10 visual analogue scale [VAS]), and analgesic use for acute pain treatment. At each visit following assessments were conducted: Migraine diary collection, Migraine Disability Test (MIDAS) (24), Headache Impact Test, version 6 (HIT-6) (25) and laboratory parameters. Blood ketone levels were assessed using a portable point-of-care blood ketone meter (precision xtra from Abbot). Further details on data collection are provided in the published study protocol (23).

Outcome measures

The primary endpoint was defined as the number of migraine days in the last four weeks of treatment. The number of migraine days during the respective run-in period was used as a covariate (baseline-adjustment).

Secondary endpoints were assessed during the last four weeks of intervention and were adjusted for baseline. Following outcomes were included: average migraine intensity; consumption of acute migraine medication (number of days with acute medication use); MIDAS score and HIT-6 score.

Sample size estimation

The sample size was estimated to show the superiority of βHB over placebo, assuming a reduction of 2.0 migraine days. A total of 31 evaluable patients would provide 90% power to reject the null hypothesis of no difference at a significance level of 0.05. Anticipating a drop-out rate of 30%, we planned to recruit a total of 45 patients. Further details are provided in the study protocol (23).

Statistical analysis

All statistical analyses were predefined in a statistical analysis plan and conducted using the statistical software package R, Version 4.0.0. All analyses were performed on the intention to treat (ITT) set, and missing values were imputed. For the primary endpoint, two per protocol sensitivity analyses were performed, and missing values were imputed. The primary endpoint was assessed using a linear mixed-effects model. A standard crossover analysis was implemented with treatment (verum-placebo), period (2-1) and their interaction as fixed effects and with subject as random effect. The baseline value –number of migraine days during the run-in period – was included as covariate. Residual distribution was inspected using diagnostic model plots. The estimated regression coefficients and estimated baseline – adjusted differences between treatments are reported with 95% confidence interval (CI) and p-value.

The robustness of the results of the primary analysis was assessed by performing two per-protocol (PP) analyses. The PP analysis sensu stricto included only patients who were compliant during both placebo and verum treatment. For the PP analysis sensu lato all patients who were compliant during the verum treatment – irrespective of compliance during the placebo treatment – were included. Two pre-planned subgroup analyses were performed in order to explore whether the magnitude of the treatment effect depends on migraine with aura (yes/no) or on the baseline frequency of migraine days (medium = 5–9 days/four weeks; high = 10–14 days/four weeks). For each subgroup, the statistical model was complemented with the main effect and the treatment-interaction term for the respective variable. We report the p-value of the interaction term. Treatment effects are reported with 95% CI estimated for each subgroup separately.

Secondary endpoints were analysed for a difference between βHB and placebo using appropriate types of regression models with the ITT set. The models included the corresponding baseline measure as covariate, treatment, period and their interaction as fixed effects and subject as random effect. The estimated baseline-adjusted differences between treatments are reported with 95% CI and p-value.

The number of days with consumption of acute migraine medication and the MIDAS showed a strongly right-skewed distribution and were analysed using negative binomial mixed-effects regression. Estimates are back-transformed and reported as relative effects. Migraine intensity ratings and the HIT-6 score were analysed using linear mixed-effects regression. All secondary analyses were performed on the ITT set, using all data available (complete case analyses).

Results

Fifty-five patients were screened, of these 41 patients were eligible and randomised. Before the baseline visit one patient (randomised into the placebo arm) withdrew consent and was therefore excluded. The ITT set included 40 patients in the first treatment arm during week 1–4. Two patients dropped out before the end of first treatment period due to withdrawal of consent (placebo) and another patient due to intolerable side effects (verum). At the end of the first treatment, 38 patients were enrolled in the study (22 placebo, 16 verum).

Before the start of the second phase four patients discontinued the study (three patients finished the placebo arm and one patient the verum arm). Therefore, 34 patients entered the second treatment period after the wash-out phase. During the first four weeks of the second treatment period, two patients dropped out. No further dropouts occurred thereafter; hence, 32 patients completed both treatment periods: 18 patients with treatment sequence “placebo- verum”, and 14 with “verum-placebo”.

The ITT set includes all patients that dropped out after the first four weeks of the first treatment period. Multiple imputation was performed concerning the primary endpoint for two patients in the first treatment period, who dropped out after the first four weeks and before the end of the first treatment. The primary endpoint for the second treatment period was not imputed for these patients. Patients that dropped out after the end of the first treatment period or during the first four weeks of the second treatment period no imputations were performed due to lack of information on the outcome in the second treatment period.

The per protocol set (PP) sensu stricto includes 32 patients that completed both treatment periods with available primary outcomes in both periods. The verum PP set sensu lato includes 35 patients who completed the verum treatment, irrespective of treatment period and completion of placebo treatment.

The mean age of patients was 35.15 years (10.92 SD) and 37 patients were female (92%), see Table 1. At screening visit 23 (57%) patients confirmed to suffer from aura. Patients reported an average of 17.82 years (10.41 SD) since migraine onset. Patient characteristic at each baseline visit are presented in Table 2.

Table 1.

Summary of patient characteristics assessed at screening visit. Categorical variables are reported as frequencies (%) and numerical variables as mean with SD.

Characteristic	All (n = 40)
Age, y	35.15 (10.92)
Sex, (%) female	37 (92%)
Aura, yes	23 (57%)
Years since onset of migraine	17.82 (10.41)

Table 2.

Summary characteristics assessed at each baseline visit. Migraine days refer to four weeks preceding baseline visit. Weight and BMI: mean (SD), other variables: median [inter-quartile range]. BMI= body mass index.

Characteristics	βHB (n=36)	Placebo (n=38)
Weight (kg)	67.87 (13.68)	68.71 (14.48)
BMI (kg/m²)	23.96 (4.33)	23.40 (3.85)
Migraine intensity average	5.75 [4.07, 6.73]	5.50 [4.20, 7.00]
Migraine days	7.50 [6.00, 10.00]	8.00 [6.00, 9.00]
ALAT (U/l)	16.50 [13.75, 20.25]	17.00 [13.25, 22.00]
ASAT (U/l)	21.00 [19.00, 25.00]	21.50 [19.00, 25.00]
Creatine kinase (U/l)	90.00 [72.00, 115.25]	80.00 [57.50, 113.25]
Cholesterol (mmol/l)	4.31 [3.80, 4.85]	4.50 [3.86, 4.87]
HDL (mmol/l)	1.61 [1.38, 1.78]	1.63 [1.42, 1.80]
Cholesterol/HDL quotient	2.54 [2.27, 3.40]	2.50 [2.27, 3.23]
CRP (mg/l)	0.85 [0.57, 1.97]	0.90 [0.40, 1.88]
HbA1c (%)	5.00 [4.70, 5.32]	4.95 [4.80, 5.18]
Creatinine (µmol/l)	66.00 [60.00, 74.50]	64.50 [62.00, 72.75]
Insulin (mlU/l)	8.35 [6.97, 10.03]	8.55 [6.53, 11.30]
TSH (mlU/l)	1.97 [1.48, 2.31]	1.83 [1.24, 2.57]
βHB before IMP/ Placebo intake (mmol/l)	0.20 [0.10, 0.30]	0.21 [0.10, 0.30]
βHB after 40 min IMP/ Placebo intake (mmol/l)	0.40 [0.30, 0.50]	0.19 [0.10, 0.30]

The estimated, baseline-adjusted mean difference in the number of migraine days during the last four weeks of treatment between βHB and placebo was −1.11 (95% CI [−5.07, 2.85]) days, p = 0.578 (Table 4). Therefore, patients suffered about one migraine day less after taking βHB compared to placebo, which is only half the effect we assumed and considered clinically relevant. However, the between-patient variability of the treatment effect was very large; the resulting 95% CI indicates that a treatment effect ranging between a reduction by five days to an increase by nearly three days is plausible, based on our data.

Table 3.

Summary statistics of outcomes at baseline and end of treatment and the individual changes from baseline to end of treatment for βHB and placebo. Mean (standard deviation) or median [inter-quartile range] are given. MIDAS= Migraine Disability Test, HIT-6 = Headache Impact Test.

Characteristic	βHB			Placebo
	Baseline	End of treatment	Change from baseline	Baseline	End of treatment	Change from baseline
Migraine days	7.5 (2.6)	5.9 (2.8)	−1.7 (3.1)	7.4 (2.5)	6.2 (2.9)	−1.3 (2.9)
Migraine intensity	5.5 (1.9)	5.9 (2.2)	0.4 (1.8)	5.6 (1.8)	5.6 (1.6)	0 (1.4)
Medication use	6 [3.5, 8]	4.5 [2.25, 7]	−0.5 [−2, 1.75]	5 [3, 8]	4 [2, 7]	−1 [−2, 1]
MIDAS score	23 [14.5, 43]	13.5 [6, 22.75]	−10.5 [−18, −0.25]	18 [13, 30]	16.5 [7.5, 25.25]	−5 [−12.5, 5.25]
HIT-6 Score	61.7 (5.2)	58.6 (5.5)	−3.2 (6.2)	61.2 (3.6)	59.4 (5.4)	−1.8 (5)

The sensitivity analyses confirmed the primary result (PP sensu stricto: −1.5 [−5.8, 2.8], p = 0.498; PP sensu lato: −1.9 [−6, 2.1], p = 0.350). The treatment effect was not modified by aura or migraine frequency. The mean difference for patients with aura is −1.4 (95% CI [−6.3, 3.5]) days, and for those without aura −0.7 days (95% CI [−7.5, 6]), p-value for interaction term = 0.748. The mean difference for medium frequency migraineurs was −0.6 days (95% CI [−5.7, 4.5]), and for the high frequency migraineurs −3.6 days (95% CI [−11, 3.9]), p-value for interaction term = 0.488.

Our data provided no statistical evidence for a beneficiary effect of βHB as compared to placebo regarding the secondary outcomes (Table 4). On average, the number of days with intake of acute migraine medication decreased under both βHB and placebo (Table 3) and we found no evidence for a difference between treatments (relative difference: 0.84 [0.30, 2.14], p = 0.709). On average, patients reported an intermediate level of migraine intensity both at baseline and at the end of treatment. The average difference in migraine intensity at the end of treatment was 1.5 on the VAS scale ([−0.8, 3.7], p = 0.207). Both, MIDAS and HIT-6 score decreased under βHB and placebo. Furthermore, we found no evidence for a difference between treatments. (MIDAS: relative difference: 2.8 [0.69, 10.86], p = 0.128; HIT-6: 7.2 score points [−0.6, 15, p = 0.0741).

Table 4.

Estimated treatment effects (mean difference) of βHB versus placebo with 95% confidence interval (CI) and p-value. Analyses were performed on the ITT set (n = 40) using mixed-effects regression models. MIDAS= Migraine Disability Test, HIT-6= Headache Impact Test, * = mean difference given as relative effect.

Characteristics	Mean difference (βHB vs Placebo)	95 % CI	p-value
Migraine days	−1.11	−5.07, 2.85	0.578
Migraine intensity	1.47	−0.8, 3.7	0.21
Medication use in days*	0.84	0.30, 2.14	0.70904
MIDAS score*	2.80	0.69, 10.86	0.1281
HIT-6 Score	7.21	−0.6, 15	0.0741

Safety analyses

No serious adverse events (AE) occurred during the trial. The frequency of AEs between βHB and placebo was not different. The most frequent AE was gastrointestinal upset, followed by upper airway infection (Table 5). AE of the same type recorded multiple times, occurring within the same treatment period is considered only once per patient.

Table 5.

Adverse events observed in patients during verum and placebo including 8 weeks of wash-out, n (%).

AE type	βHB	Placebo
n	46	37
Bacterial pharyngitis	1 (2)	2 (5)
Bronchitis	2 (4)	0 (0)
Ear pain	1 (2)	0 (0)
Eczema	1 (2)	1 (3)
Elective surgery	1 (2)	0 (0)
Elevated T3	0 (0)	1 (3)
Elevated TSH	0 (0)	1 (3)
Erythem	0 (0)	1 (3)
Gastroenteritis	2 (4)	3 (8)
Gastrointestinal upset	19 (41)	16 (43)
Hyponatraemia	1 (2)	0 (0)
Injection site pain	1 (2)	0 (0)
Knee disocation	1 (2)	0 (0)
Musculoskeletal chest pain	0 (0)	1 (3)
Nausea	1 (2)	0 (0)
Constipation	1 (2)	0 (0)
Osteopenia	0 (0)	1 (3)
Otitis media	0 (0)	1 (3)
Paraesthesia	1 (2)	0 (0)
Sore Throat	3 (7)	2 (5)
Upper airway infection	10 (22)	7 (19)

Laboratory values outside of normal ranges were measured under βHB as well as placebo and with equal frequencies. This mainly concerned cholesterol, HbA1c and to a lower extent creatinine kinase, (See supplemental Table 1).

Discussion

Recent research suggests a substantial prophylactic effect of KD in migraineurs. Due to difficulties in implementing such a demanding diet into the daily routine, exogenous KB supplementation was used as a prophylactic intervention in episodic migraine with the aim to create mild ketosis. To the best of our knowledge, this is the first randomised clinical trial (RCT) evaluating the prophylactic effect of exogenous βHB supplementation in medium and high-frequency episodic migraineurs.

The obtained results of 12 weeks of exogenous βHB supplementation in episodic migraineurs demonstrated no reduction in migraine frequency. The expected reduction of two migraine days during βHB treatment compared to placebo was not achieved. We found a smaller effect size (one migraine day reduction), accompanied by substantial interindividual variance. The treatment effect was not modified by differences in migraine frequency (high vs medium) and aura present or absent. The PP analyses resulted in a slightly larger treatment effect (reduction of up to two migraine days). Hence, it is tempting to speculate that despite the primary negative result of this trial, βHB treatment might be an efficient intervention in migraine. Furthermore, the findings demonstrated no reduction in any secondary outcomes of our sample. The supplementation of βHB reported AEs of mild intensity, the most common one being gastrointestinal upset and upper airway infection. KB supplementation is known to be associated with gastrointestinal upset (26). No difference in AE frequency was observed between βHB and placebo treatment. Laboratory parameters did not show any relevant changes.

There are several factors, which might have contributed to the trial not attaining its primary endpoint, apart from the previously mentioned higher than expected variabilities between patients. (I) the low dose of βHB, (II) the selected racemate was not appropriate, and (III) not assessing recordings of migraine frequency prior to study entry may have contributed to the obtained negative outcome.

Since we observed no significant reduction of migraine frequency or intensity, it has to be discussed if the chosen dose of βHB was appropriate. First, in this study the average blood ketone level after 40 minutes of intake of IMP was 0.5 mmol/l. This suggests that the chosen dose did barely breach the anticipated medium therapeutic ketosis level, since the level is assumed to start around 0.4 mmol/l. Furthermore, the selected dose of 18g βHB mineral salt per day was very conservative compared to the production of up to 185g of KBs during fasting (27). The dose in this trial was determined by the mineral load of Mg2+ and Ca2+, which had to be maintained at an acceptable range. The possibility has to be considered that the chosen daily dose of 18g βHB was too low, which may be the reason for obtaining no beneficial effect in our analysis. Future studies should investigate appropriate dose and potentially consider other mineral forms of βHB when aiming for the amelioration of migraine.

The main limitation of this trial is that the chosen KB 3-hydroxybutyrate is a chiral molecule with two enantiomers, D-3-hydroxybutyrate and L-3-hydroxybutyrate. When starting the trial pure enantiomer D-βHB by itself was not available. The use of racemic βHB in mineral form may have been not ideal, since L-3-hydroxybutyrate is utilised more slowly and through different metabolic routes (22,28). Further reports suggest that L-βHB is not significantly metabolised into energy intermediates (22,29). Possibly, the chosen βHB mineral form is not as ketogenic as anticipated. In the future, supplementation with the enantiomer D-βHB should be considered when aiming to implement mild ketosis exogenously.

Furthermore, natural fluctuations are known in migraine and have to be considered as a potential cause of the observed variability. Prospective patients were only included if they suffered periodically from migraine for at least three months. However, this was based on patient recollection and no frequency records were collected three months prior to screening. This could have contributed to the observed individual variability. Hence, longer observation periods should be included to assess migraine periodicity.

Our hypothesis is primarily based on reports of significant reduction of migraine frequency and intensity by adherence to the KD (12,13). A relevant limitation of our trial may be that the KD is not only defined by ketosis but possibly also effective due to other mechanisms such as hypoglycaemia, absence of insulinemic peaks or lowered activity of nutrient-integrating pathways. These mechanisms could have also caused the observed attenuation of migraine symptoms and frequency (30,31). This implication is supported by studies reporting beneficial effects on symptoms like migraine frequency when adhering to a low glycaemic index diet (32). Therefore, along with ketosis, the accompanying metabolic changes and their effect on migraine symptoms must be further examined.

At last, migraineurs are an extremely heterogeneous patient group with only very little current classification. Future metabolic therapeutics should be tested on a more clearly defined migraine subtype that has migraine triggers with a clear link to metabolism, such as fasting, exercise, hypoxia or similar.

Conclusion

This study stands out as the first crossover double-blind, placebo-controlled RCT to specifically evaluate the prophylactic effect of exogenous βHB supplementation in episodic migraineurs. Our results obtained no beneficial effect in migraine frequency or intensity during 12 weeks of βHB treatment. The potential prophylactic effect of exogenous KB supplementation in migraine is not to be discarded considering our sample observed a wide variability in patients, the dosage of βHB supplementation was very conservative and the use of βHB racemate as supplementation.

Clinical implications

Exogenous ketone bodies supplementation is a novel and promising prophylactic treatment against migraine.

12 weeks of DL-beta-hydroxybutyrate supplementation in episodic migraineurs obtained no beneficial treatment effect.

Further research is required to understand the potential role of beta-hydroxybutyrate as prophylactic treatment in migraineurs.

Supplemental Material

sj-pdf-1-cep-10.1177_03331024211043792 - Supplemental material for Efficacy and safety of exogenous beta-hydroxybutyrate for preventive treatment in episodic migraine: A single-centred, randomised, placebo-controlled, double-blind crossover trial

Supplemental material, sj-pdf-1-cep-10.1177_03331024211043792 for Efficacy and safety of exogenous beta-hydroxybutyrate for preventive treatment in episodic migraine: A single-centred, randomised, placebo-controlled, double-blind crossover trial by Niveditha Putananickal, Elena C Gross, Anna-Lena Orsini, Simone Schmidt, Patricia Hafner, Vanya Gocheva, Sara Nagy, Bettina C Henzi, Daniela Rubino, Deborah R Vogt, Sven Cichon, Peter Sandor and Dirk Fischer in Cephalalgia

Footnotes

Acknowledgments

We would like to thank the study nurses of the outpatient clinic and our data manager Constantin Sluka.

Authors’ contributions

NP participated in the organisation, conduct, data acquisition and drafted the manuscript. ECG participated in the organisation, conduct, data acquisition and edited the manuscript. PH, SS, ALO, SN and BCH participated in the conduct of the study. VG and DR participated in the organisation and conduct of the study. DRV covered all statistical aspects of the study and the manuscript. PS, SC, DF participated in the study design, its organisation, and edited the manuscript.

Declaration of conflicting interests

The authors declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: ECG and DF are the inventors of the patent WO 2018/115158 Al held by the UKBB and University of Basel on the use of beta-hydroxybutyrate in migraine prevention. Other authors declare no conflicts of interest.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This trial was funded by the Swiss National Science Foundation, grant number [32003B_173193/1].

ORCID iD

Elena C Gross

References

Steiner

Stovner

Vos

T, et al.

Migraine is first cause of disability in under 50s: will health politicians now take notice?

J Headache Pain 2018; 19: 17.

Stovner

Nichols

Steiner

, et al. Global, regional, and national burden of migraine and tension-type headache, 1990–2016: a systematic analysis for the Global Burden of Disease Study 2016. Lancet 2018; 17: 954–976.

Sandor

Di Clemente

Coppola

, et al. Efficacy of coenzyme Q10 in migraine prophylaxis: a randomized controlled trial. Neurology 2005; 64: 713–715.

Sandor

Dydak

Schoenen

, et al. MR-spectroscopic imaging during visual stimulation in subgroups of migraine with aura. Cephalalgia 2005; 25: 507–518.

Sparaco

Feleppa

Lipton

, et al. Mitochondrial dysfunction and migraine: evidence and hypotheses. Cephalalgia 2006; 26: 361–372.

Colombo

Saraceno

Comi

Riboflavin and migraine: the bridge over troubled mitochondria. Neurol Sci 2014; 35 Suppl 1: 141–144.

Yorns

Jr. Hardison

HH.

Mitochondrial dysfunction in migraine. Semin Pediatr Neurol 2013; 20: 188–193.

Schoenen

Jacquy

Lenaerts

Effectiveness of high-dose riboflavin in migraine prophylaxis. A randomized controlled trial. Neurology 1998; 50: 466–470.

Markley

HG.

CoEnzyme Q10 and riboflavin: the mitochondrial connection. Headache 2012; 52 Suppl 2: 81–87.

10.

Okada

Araga

Takeshima

, et al. Plasma lactic acid and pyruvic acid levels in migraine and tension-type headache. Headache 1998; 38: 39–42.

11.

Pietrobon

Striessnig

Neurobiology of migraine. Nat Rev Neurosci 2003; 4: 386–398.

12.

Di Lorenzo

Coppola

Sirianni

, et al. Migraine improvement during short lasting ketogenesis: a proof‐of‐concept study. Eur J Neurol 2015; 22: 170–177.

13.

Maggioni

Margoni

Zanchin

Ketogenic diet in migraine treatment: a brief but ancient history. Cephalalgia 2011; 31:1150–1151.

14.

Neal

Chaffe

Schwartz

, et al. The ketogenic diet for the treatment of childhood epilepsy: a randomised controlled trial. Lancet 2008; 7: 500–506.

15.

Reid

Mullen

Kim

, et al. Epilepsy, energy deficiency and new therapeutic approaches including diet. Pharmacol Ther 2014; 144: 192–201.

16.

Nei

Ngo

Sirven

, et al. Ketogenic diet in adolescents and adults with epilepsy. Seizure 2014; 23: 439–442.

17.

de Almeida Rabello Oliveira

da Rocha Ataide

de Oliveira

, et al. Effects of short-term and long-term treatment with medium- and long-chain triglycerides ketogenic diet on cortical spreading depression in young rats. Neurosci Lett 2008; 434: 66–70.

18.

Hartman

Gasior

Vining

, et al. The neuropharmacology of the ketogenic diet. Pediatr Neurol 2007; 36: 281–292.

19.

Berg

Yellen

Ketogenic diet metabolites reduce firing in central neurons by opening KATP channels. J Neurosci 2007; 27: 3618–3625.

20.

Gautschi

Weisstanner

Slotboom

, et al. Highly efficient ketone body treatment in multiple acyl-CoA dehydrogenase deficiency-related leukodystrophy. Pediatr Res 2015; 77: 91–98.

21.

Chiolero

Mavrocordatos

Burnier

, et al. Effects of infused sodium acetate, sodium lactate, and sodium beta-hydroxybutyrate on energy expenditure and substrate oxidation rates in lean humans. Am J Clin Nutr 1993; 58: 608–613.

22.

Stubbs

Cox

Evans

, et al. On the metabolism of exogenous ketones in humans. Front Physiol 2017; 8: 848.

23.

Gross

Putananickal

Orsini

, et al. Efficacy and safety of exogenous ketone bodies for preventive treatment of migraine: A study protocol for a single-centred, randomised, placebo-controlled, double-blind crossover trial. Trials 2019; 20: 61.

24.

Fragoso

YD.

MIDAS (Migraine Disability Assessment): a valuable tool for work-site identification of migraine in workers in Brazil. Sao Paulo Med J 2002; 120: 118–121.

25.

Kosinski

Bayliss

Bjorner

, et al. A six-item short-form survey for measuring headache impact: the HIT-6. Qual Life Res 2003; 12: 963–974.

26.

Clarke

Tchabanenko

Pawlosky

, et al. Oral 28-day and developmental toxicity studies of (R)-3-hydroxybutyl (R)-3-hydroxybutyrate. Regul Toxicol Pharmacol 2012; 63: 196–208.

27.

Laffel

Ketone bodies: a review of physiology, pathophysiology and application of monitoring to diabetes. Diabetes Metab Res Rev 1999; 15: 412–426.

28.

Webber

Edmond

Utilization of L(+)-3-hydroxybutyrate, D(-)-3-hydroxybutyrate, acetoacetate, and glucose for respiration and lipid synthesis in the 18-day-old rat.

J Biol Chem 1977; 252: 5222–5226.

29.

Newman

Verdin

Beta-Hydroxybutyrate: A signaling metabolite. Annu Rev Nutr 2017; 37: 51–76.

30.

Di Lorenzo

Coppola

Bracaglia

, et al. Cortical functional correlates of responsiveness to short-lasting preventive intervention with ketogenic diet in migraine: a multimodal evoked potentials study. J Headache Pain 2016; 17: 1–10.

31.

Evcili

Utku

Ogun

, et al. Early and long period follow-up results of low glycemic index diet for migraine prophylaxis. Agri 2018; 30: 8–11.

32.

Finsterer

Frank

Low-glycemic-index diet relieving migraine but inducing muscle cramps. J Neurosci Rural Prac 2019; 10: 552.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.13 MB

Efficacy and safety of exogenous beta-hydroxybutyrate for preventive treatment in episodic migraine: A single-centred,randomised,placebo-controlled,double-blind crossover trial

Abstract

Background

Aim

Methods

Results

Conclusion

Keywords

Introduction

Methods

Participants and setting

Trial design

Randomisation and blinding

Medication

Changes in study protocol

Clinical and demographic measures

Outcome measures

Sample size estimation

Statistical analysis

Results

Safety analyses

Discussion

Conclusion

Clinical implications

Supplemental Material

sj-pdf-1-cep-10.1177_03331024211043792 - Supplemental material for Efficacy and safety of exogenous beta-hydroxybutyrate for preventive treatment in episodic migraine: A single-centred, randomised, placebo-controlled, double-blind crossover trial

Footnotes

Acknowledgments

Authors’ contributions

Declaration of conflicting interests

Funding

ORCID iD

References

Supplementary Material