Article - Very-low-calorie ketogenic diet vs hypocaloric balanced diet in the prevention of high-frequency episodic migraine: the EMIKETO randomized, controlled trial.

Plain Language Summary

Migraine is the second most common cause of disability worldwide and is linked with obesity, especially in women of reproductive age. The aim of this randomised controlled trial was to evaluate the effect of a calorie-restricted ketogenic diet (KD) on migraine frequency. This randomised controlled trial included 57 patients with high-frequency episodic migraine (HFEM) and overweight. Patients randomised to the KD followed a very low-calorie KD (VLCKD) for 8 weeks, followed by a low-calorie diet (LCD) for 4 weeks and a hypocaloric balanced diet (HBD) for another 12 weeks. The control group followed the HBD for 24 weeks. 4 of the 29 patients in the KD group withdrew from the study and 14 of 28 in the HBD group. Migraine frequency reduced in both groups but significantly more so in the KD group. There was no effect on migraine severity in either group. The KD was also more effective than the HBD with regards to weight loss, blood pressure and inflammatory markers. The authors conclude that the VLCKD is an effective therapy for patients with HFEM and overweight or obesity.

Expert Review

Reviewer: Karin Elgar

13th Nov 2024

Conflict of interest

None

Take home message

A VLCKD may help reduce migraine frequency in patients with HFEM and overweight/obesity.

Evidence category

A: Meta-analyses, position-stands, randomized-controlled trials (RCTs)

Summary review

Introduction

The aim of this study was to compare the effects of a very low-calorie ketogenic diet (VLCKD) with a hypocaloric balanced diet (HBD) on migraine prevention in adults with high-frequency episodic migraine (HFEM) and overweight/obesity.

Methods

24-week randomised controlled trial, including 57 adults aged 18-65 years with HFEM and a body mass index between 25-35kg/m2.
Patients on the VLCKD diet plan received a diet containing less than 800 kcal per day (75-105g protein, 30-50g carbohydrate (CHO), 20g fat, 25g fibre) for the first 4 weeks. This was followed by a low-calorie diet for 4 weeks during which CHO were gradually reintroduced. From week 12 the group followed the HBD.
Participants in the control group followed the HBD with a calorie intake of 1500-1600kcal per day (30% fat, 55% CHO, protein approximately 0.8-1.5g/kg ideal bodyweight) for the duration of the study.
The primary outcome measure was monthly migraine days (MMD). Secondary outcomes included anthropometric parameters, blood lipids, glycaemic and inflammatory parameters, electrolytes, liver and kidney function tests.

Results

13.8% of participants on the VLCKD diet and 50% on the HBD dropped out of the study, although no attrition bias was seen in sensitivity analysis.
Patients on the VLCKD had a statistically significant reduction in MMD after 8 weeks (− 6.4±4.8 vs − 2.2±5.0, p=0.008). The reduction remained greater in the VLCDK group after reintroducing CHO and switching to a HBD (week 12: − 7.2±5.42 vs − 3.13±3.58, p=0.007; week 24: − 6.8±6.42 vs − 3.6±3.3, p=0.042).
Weight loss was greater in the VLCKD compared to the control group (week 8: − 8.2±4.5 vs − 4.3±2.9, p=0.002; week 12: − 9.1±6.4 vs − 4.9±2.7, p=0.020; 24 weeks: − 9.1±6.4 vs − 4.3±2.9, no p-value given).
At week 8, adherence to the different dietary protocols was higher in the VLCKD than in the HBD group (p=0.028).
There was a greater reduction in glucose levels in the VLCKD group at week 12 (p=0.003) as well as a greater reduction in AST (p=0.046). There were no significant differences in other blood tests.

Conclusion

The authors conclude that a VLCKD is an effective treatment or adjuvant therapy for patients with HFEM.

Clinical practice applications

A short-term VLCKD can be considered in patients with HFEM and overweight/obesity to help reduce migraine frequency and weight.

Considerations for future research

Clinical trials on a ketogenic diet (not low calorie) in normal weight migraineurs, to establish whether benefits are mediated via weight.
Clinical trials of ketogenic diets in patients with other types of migraines.
Larger studies with consideration for factors that led to high drop-out rates on HBD.

Expert reviews are written by nutrition professionals and academics with advanced qualifications to provide a critical appraisal of the article and implications for practice. Each review is peer-reviewed by a member of the NED Editorial Board. Find out more about our NED Expert Reviewers here.

Abstract

BACKGROUND

Migraine is the second world's cause of disability. Among non-pharmacological treatments, nutritional intervention, particularly ketogenic diet, represents one of the most promising approaches.

METHODS

This a prospective, single center, randomized, controlled study aimed at evaluating the efficacy of a very low-calorie ketogenic diet (VLCKD) compared to a hypocaloric balanced diet (HBD) in migraine prophylaxis in patients affected by high-frequency episodic migraine (HFEM) with a Body Mass Index (BMI) > 27 kg/m. Fifty-seven patients were randomly assigned to a VLCKD (group 1) or HBD (group 2). Group 1 patients followed a VLCKD for 8 weeks, followed by a low calorie diet (LCD, weeks 9-12), and a HBD (weeks 13-24), whereas group 2 patients followed a HBD from week 0 to 24. Anthropometric indexes, urine and blood chemistry were assessed at enrollment, baseline, weeks 4, 8, 12, and 24. Migraine characteristics were evaluated at baseline, weeks 8, 12 and 24. Change in monthly migraine days (MMDs) at weeks 5-8 compared to baseline was the primary endpoint. Secondary endpoints encompassed changes in visual analogue scale (VAS), Headache Impact Test-6 (HIT-6) and Short Form Health Survey-36 (SF-36) scores. We also studied effects on circulating lymphocytes and markers of inflammation, changes in plasma aldosterone and renin levels before and after VLCKD or HBD treatment.

RESULTS

Reduction from baseline in MMDs was greater in VLCKD compared to HBD group at week 8 (p = 0.008), at week 12 (p = 0.007), when ketosis had been interrupted by carbohydrates reintroduction, and at week 24 (p = 0.042), when all patients were following the same dietary regimen. Quality of life scores (SF-36) were improved in VLCKD group at week 8 and 12, and were also improved in HBD group, but only at week 12. Weight-loss was significantly higher in VLCKD group at week 8 (p = 0.002) and week 12 (p = 0.020). At the end of the study weight loss was maintained in VLCKD group whereas a slight weight regain was observed in HBD group. Inflammatory indexes, namely C reactive protein (CRP), neutrophil to lymphocyte ratio (NLR) and total white blood cell count (WBC) were significantly reduced (p < 0.05) in VLCKD group at week 12. Aldosterone plasma level were significantly increased in both groups at week 8, particularly in VLCKD group. However, electrolytes and renin plasma levels were never altered throughout the study in both groups.

CONCLUSIONS

VLCKD is more effective than HBD in reducing MMD in patients with HFEM and represents an effective prophylaxis in patients with overweight/obesity. Trial registration ClinicalTrials.gov identifier: NCT04360148.