The impact of ketogenic diet on the onset and progression of multiple sclerosis

Abstract

Multiple sclerosis (MS) is a chronic autoimmune disease of the central nervous system (CNS) characterized by inflammation and neurodegeneration. Current research suggests that diet may influence disease course, severity of symptoms, and quality of life in MS patients. The ketogenic diet (KD) has been used for more than a century as a therapeutic approach for various medical conditions. It was originally developed in the 1920s as a treatment option for epilepsy, and especially in the last 30 years, has gained popularity for its potential benefits in a variety of neurological conditions other than epilepsy. This prompted us to perform a literature survey regarding the effect of KD on the onset and progression of MS. The here reviewed 15 original research articles including in vitro, preclinical, and clinical studies provide evidence for the safety and feasibility of the KD in MS, showing potential neuroprotective effects and positive impacts on cellular metabolism and disease outcome. Since the literature is limited and most studies were conducted with low numbers of MS patients and rather exploratory in nature, further studies with larger cohorts are needed to gain a better understanding of the mechanisms by which the improvements of the MS disease course are achieved.

Introduction

Multiple sclerosis

Multiple sclerosis (MS) is the most common chronic inflammatory and neurodegenerative disease of the central nervous system (CNS) with typical hallmarks of inflammation and demyelination, associated with gliosis and neurodegeneration [1]. It is estimated that approximately 2.8 million people worldwide are currently living with MS (35.9 per 100,000 individuals), while the prevalence has increased all around the globe since 2013. MS is diagnosed at the mean age of 32 years, with females suffering twice as frequently as males [1]. The disease development in MS patients is variable and comprises symptoms such as sensory loss, impaired muscle function, visual impairment or cognitive dysfunction [2]. Whereas both, the time course and the severities of developing symptoms are unpredictable, three clinical courses are currently distinguished, namely primary progressive MS, secondary progressive MS, and relapsing-remitting MS. In addition, the so-called clinically isolated syndrome can progress to MS when clinical symptoms are accompanied by CNS lesions [3].

Ketogenic diet

The ketogenic diet (KD) is characterized by its high-fat content and low levels of carbohydrates and proteins. The typical ratio of fats to carbohydrates and protein (all measured in grams) is around 3:1 or 4:1 [4]. While the KD focuses on reducing the carbohydrate intake, it introduces an alternative energy source to the brain called ketones. The KD has been used since the early 1920s to control seizures in epilepsy patients who did not respond well to medication but has been somewhat forgotten thereafter due to advanced developments in anticonvulsive drugs [5]. During the last 30 years, however, the KD has increasingly gained research attention, and studies provided emerging evidence of the promising therapeutic potential of the KD for a broad variety of morbidities including obesity, metabolic syndrome, and type 2 diabetes mellitus as well as inflammatory intestinal diseases, and cancer [6]. In addition, there have been reports of potential benefits of the KD in neurological diseases besides epilepsy like Parkinson's disease [7] and Alzheimer's disease [8], for instance.

Aim

This literature survey aims to review and summarize the results obtained from in vitro, preclinical, and clinical studies addressing the effects of the KD on the onset, progression, and outcome of MS.

Methods

Search strategy, data collection, inclusion, and exclusion criteria

A structured literary survey was conducted from June 7th to June 15th, 2023, by applying the meta database “PubMed” of the U.S. National Institute of Health with access to biomedical and medical studies, including MEDLINE, PubMed Central, and MeSH databases. First, a rather broad search approach was performed in the PubMed database by the term “multiple sclerosis“. Then, the MeSH term “demyelinating autoimmune diseases, CNS“ was added to the search in order to include all potentially relevant studies that addressed MS without including this in the title or abstract. The third part of the search contained the keyword “ketogenic diet”. Finally, the first and second searches were combined with the operator “OR” and finally added to the third search by using the operator “AND”. By this approach, 61 studies could be found in total. After careful reading of all abstracts, 20 studies were excluded since they addressed other topics. Of the remaining 41 studies, 26 papers were excluded of which 25 publications were review articles, and one paper was a protocol of an ongoing study. The remaining 15 original research articles were included in our literature review.

Results

Preclinical studies

In a subset of the reviewed studies, the impact of KD in MS was investigated in preclinical studies applying in vivo MS models. Two mouse models are regarded as suitable to study the immunopathogenesis and potential treatment options for autoimmune diseases of the CNS including MS, namely the experimental autoimmune encephalomyelitis (EAE) and the cuprizone (CPZ) mouse models [9]. The two models differ in the mechanisms of demyelination. In the EAE model, demyelination is the result of autoimmune cell responses particularly exerted by activated T lymphocytes accumulating in the CNS that attack the myelin of the nerve fibers. In the CPZ model, however, demyelination occurs due to direct toxicity of the compound to oligodendrocytes, resulting in the loss of myelin sheaths [9, 10].

Choi et al. [9] utilized the EAE model to analyze the benefits of various dietary regimens during disease development. The authors found that KD was able to alleviate disease severity when initiated around the clinical onset of EAE. When compared to a fasting-mimicking diet, however, the biological effect turned out to be rather minor given that the mean severity scores (EAE score range from 0 to 5) were reduced by approximately 2 and 1 following KD and fasting-mimicking diet, respectively [9].

Since the clinical manifestations of MS are heterogeneous, several preclinical studies exist on diverse symptoms of MS and how these are affected by a KD. Zyla-Jackson and coworkers applied a KD regimen that was enriched in fiber and contained medium-chain triglycerides, caprylic acid (C8), and capric acid (C10) along with flaxseed oil and canola oil as fat sources [10]. The study addressed whether the modified KD could preserve motor and visual functions in male and female EAE mice. The researchers evaluated both, a preventive regimen prior EAE induction and an interventional regimen starting after the onset of symptoms. The results revealed that both KD regimens could robustly protect against EAE mediated motor and vision loss concomitant with reducing immune cell infiltration and preserving myelination of the optic nerve [10].

In line, Kim et al. [11] reported that EAE mice, that received a KD showed improved motor performance when compared to mice on a standard diet. In their study, the KD diet was initiated 7 days prior EAE induction and consisted of a 3:1 ratio of fats:(carbohydrates + protein). The changes in motor function were monitored daily using a motor disability scale with higher scores reflecting a more pronounced impairment. In KD-fed EAE mice, the disability scores at both, the peak stage (14–19 days post-induction) and mild recovery stage (25–35 days post-induction) were significantly lower as compared to EAE control mice on standard diet [11]. In addition, the motor activity was assessed by measuring the swim speed. On day 5 post-induction, both, EAE mice on a standard diet and KD-fed EAE mice swam slower than naive mice without EAE. However, EAE mice on a KD showed a recovery of swim speed on day 17 days post-induction. Remarkably, there was no difference in swim speeds between naive mice and KD-fed EAE mice on 25 days after EAE induction. Consistently, increased latencies to find a hidden platform in a Morris Water Maze Test (MWMT) were observed in standard diet-fed EAE mice as compared to EAE mice on KD or the standard diet-fed naive control mice (without EAE). The authors also investigated the effect of the KD on memory function in EAE mice. This was done by measuring CA1 hippocampal synaptic plasticity (long-term potentiation) and spatial learning and memory (assessed with the MWMT). In an initial immunization period before the EAE onset, KD-fed mice showed improved results in the MWMT suggesting that KD treatment prevented a decline in behavioral spatial learning and memory [11].

Duking et al. [12] monitored the effects of KD on neuronal energy metabolism. The authors addressed whether the induction of peripheral ketosis could resolve the lack of circulating nutrients due to adipose tissue depletion in EAE mice. The study revealed a mitigation of clinical symptoms in EAE mice as indicated by lower clinical scores in EAE mice when therapeutically fed a KD as compared to a standard diet. The improved clinical conditions following KD treatment were accompanied by attenuated densities of pro-inflammatory CD4⁺, CD8⁺, and CD45⁺ immune cell populations in the spinal cord and reduced mean lesion size in EAE mice from the KD versus standard diet cohort.

Two studies investigated KD mediated changes of the clinical symptoms in MS applying the CPZ disease model [13, 14]. Both studies employed different KD related regimens but reported that in behavioral assessments, mice treated with the KD alongside CPZ exhibited improved learning, memory, anxiety-like behavior, and motor performance as compared to those treated with CPZ alone. Since KD has been shown to increase ketone bodies including β-hydroxy-butyrate, Sun et al. [13] applied β-hydroxy-butyrate intraperitoneally to mice starting either 5 days prior (prophylactic regimen) or upon CPZ administration (therapeutic regimen) for consecutive 35 days until the end of the experiments, and compared the results of behavioral tests to those of two control groups on a standard diet, with and without CPZ administration. The results of the open field test suggested that both, the prophylactic and therapeutic butyrate regimens decreased the anxiety-like behavior in CPZ mice, as the total distance and time spent in the central area of the field were both increased in mice from either butyrate intervention cohort if compared to the control groups. In the MWMT, however, no differences in average travel speeds between the butyrate intervention cohorts and the control groups could be assessed. However, the results in time improvement to find a hidden platform in butyrate treated mice suggested that butyrate application can improve the spatial learning and memory abilities in CPZ-fed mice [13].

Liu et al. [14] evaluated the effects of a KD (fat/non-fat ratio of 3:1) on behavioral, cognitive, biochemical, and histopathological parameters in CPZ mice. In the open field test, KD-treated mice showed increased exploration and decreased anxiety compared to the CPZ mice on normal diet. In the Rotarod Test, CPZ application led to impaired motor coordination in mice, as indicated by a shorter latency to fall off the rotarod apparatus. However, KD treatment significantly improved motor coordination, as KD-treated mice demonstrated longer stays on the rotarod compared to the normal diet-fed CPZ animals. During the MWMT, CPZ mice displayed deficits in learning and memory, evidenced by longer distances traveled to reach the target, decreased platform crossing times, and reduced residence duration in the target quadrant. KD treatment, however, reversed these deficits, resulting in improved performances when compared to the CPZ mice fed a normal diet [14].

Several studies also addressed the impact of KD on the neuropathological conditions in MS. In the CPZ model, a positive effect of KD, and of β-hydroxy-butyrate in particular, on hippocampal myelination was shown [13, 14]. The authors also found a reduced activation of microglia and reactive astrocytes, as well as an enhanced expression of mature oligodendrocytes. Sun et al. [13] reported that β-hydroxy-butyrate treatment efficiently supported the differentiation of oligodendrocyte progenitor cells (OPCs) and decreased the apoptosis in oligodendrocytes in CPZ mice, at least in part by down-regulating TRPA1 and PARP expression, which was associated with an inhibited p38-MAPK/JNK/JUN pathway and conversely, in an activation of PI3K/AKT/mTOR signaling [13]. In support, Liu et al. reported protection from demyelination in CPZ mice upon KD intervention through activation of P-Akt/mTOR pathways by sirtulin-1 (SIRT1) [14].

Whereas in EAE mice increased concentrations of pro-inflammatory cytokines and chemokines could be observed, two studies reported decreased secretion of pro-inflammatory interleukins in mice that were fed a KD before EAE induction [10, 11].

In their study, Duking and colleagues performed gene set enrichment analyses and showed an enhanced abundance of ketolytic enzymes, including the rate-limiting succinyl-CoA:3-ketoacid CoA transferase (SCOT), as well as an upregulated mitochondrial respiratory chain/oxidative phosphorylation [12]. The authors concluded that feeding KD to EAE mice provided the required energy substrate for neuronal ketolysis, which in turn contributed to disease amelioration [12]. Furthermore, a reduction of reactive oxygen species (ROS) levels in the brain after administration of a KD to EAE mice was shown by Kim et al. in 2012 [11]. These data were consistent with the results obtained by Duking et al. [12] showing that KD application resulted in increased expression of ROS scavengers such as catalase in EAE mice.

Clinical studies

Four open-label clinical trials have been published to date, including two randomized controlled trials, one phase II study, and one pilot study. Further, there is one case report and several studies that addressed specific aspects of KD and MS, including extension studies of the clinical trial by Bock et al. (clinicaltrials.gov identifier: NCT01538355). In 2019, Nathan et al. [15] published a case report of a 60-year-old patient with secondary progressive MS who was treated with a KD (fat:(carbohydrate + protein) ratio of 2.2:1) and subsequently showed improvement in several physical tests, as well as stabilization of the Expanded Disability Status Scale (EDSS) score. After the patient had stopped the KD, the symptoms worsened as did the EDSS score increase by 1.5 points. The patient then restarted the KD, which again caused a partial improvement and stabilization of the results in physical tests and an improved EDSS score returning to baseline level [15].

In 2019, Brenton et al. [16] performed a 6-month, single-arm, uncontrolled, open-label pilot study. Subjects were included if they were between 15 and 50 years old, diagnosed with relapsing-remitting MS, maintained disease-modifying therapy, and exhibited disease stability for at least 12 months. The 20 included patients were instructed to follow a KD modified Atkins diet with restricted carbohydrate intake (<20 g per day) and increased fat intake to induce ketosis. A study dietitian provided guidance throughout the 6-month diet trial. Compliance with the diet was assessed using urine ketone test strips, and subjects were required to provide daily-dated pictures of their test strips as evidence of ketosis. The results of the study showed that the KD modified Atkins diet was well-tolerated and safe for subjects with relapsing-remitting MS. Adherence to the diet was 75% after 6 months. No new or enlarging lesions were observed on magnetic resonance imaging (MRI) scans, indicating stability in disease progression. Subject-reported outcomes demonstrated improvements in fatigue and depression scores. The study also revealed a significant improvement in EDSS scores for subjects on diet. This improvement was primarily due to enhancements in bowel/bladder and sensory functions. Furthermore, there was no deterioration in measures of lower limb functioning, but a significant improvement at 6 months with the non-dominant hand in the 9-Hole-Peg-Test, where subjects place nine pegs (one by one) onto a block as quickly as possible, before removing them again [17], demonstrating an improved upper limb functioning [16].

Since their pilot study published in 2019 [16], Brenton et al. expanded their research in a phase II trial in order to demonstrate the tolerability of a monitored KD over 6 months [18]. Furthermore, the impact of the KD on the clinical outcomes and pro-inflammatory leptin levels were assessed. Sixty-five subjects aged 12–55 years with relapsing-remitting MS were enrolled. For this study, two pediatric subjects were included (15 and 17 years of age at the time of enrollment), representing 3% of the total study cohort. Participants in the study had to meet specific criteria, including clinical stability under their current therapy and an EDSS score of ≤6.0. The study included data from 20 subjects of the pilot study, with additional data from 45 subjects added for this phase II trial. The participants received education on initiating the KD and were instructed to restrict carbohydrates to <20 g/day and increase the “healthy” fat intake. A dietician provided ongoing support throughout the study. The subjects underwent various assessments, including EDSS evaluation, MS Functional Composite Testing, Symbol Digit Modality Test, low contrast visual acuity, 6-min walk, and accelerometer-based measurement of physical activity. Patient-reported outcomes such as depression and fatigue assessments were also conducted. Fasting blood tests for insulin, lipids, carnitine, vitamin D, and adipokines including leptin and adiponectin were performed to assess metabolic markers. The patients’ adherence to the KD was monitored using daily urine ketone test strips. The participants were required to provide photographic evidence of their daily ketone tests, and adherence was defined as >85% adherent days during the 6-month study period. 83% of subjects successfully adhered to the KD for 6 months. The final analysis included data from 64 subjects who started the KD modified Atkins diet and provided at least one dated keto test strip photograph. The study found significant reductions in the body mass index (BMI), waist circumference, fat mass, and resting metabolic rate after the intervention as well as decreased fat mass percentage in included patients. Self-reported fatigue impact, fatigue severity, and depression scores showed a nearly 50% decline. The authors noted that the magnitude of improvement in fatigue upon KD modified Atkins diet was comparable with improvements noted in MS clinical drug trials with amantadine, modafinil, and methylphenidate, for instance. Also, the quality of life-scores for physical and mental health improved. Neurological disability, as measured by EDSS, showed a reduction in scores. In addition, there was an improvement in the 9-Hole-Peg-Test and walking speed/endurance on the 6-min walk test. However, no significant changes in the other measures of cognitive function or daily activity described above could be assessed [18]. Laboratory monitoring revealed that some subjects on the KD exhibited a secondary carnitine deficiency, which was subsequently treated with carnitine supplementation. Fasting insulin, hemoglobin A1c, and triglyceride serum levels showed a significant decrease upon the diet. Serum leptin concentrations were significantly reduced at 3 and 6 months, also after accounting for the change in BMI to separate the effects of diet and BMI. Adiponectin, an anti-inflammatory adipokine, showed a non-significant trend to increase from baseline to 3 months, after accounting for changes in the BMI [18].

A randomized controlled trial was performed by Choi et al. in 2016 in order to assess the safety and feasibility of KD treatment on health-related quality of life in relapsing-remitting MS [9]. Therefore, 60 patients were randomly assigned to either a control diet, a KD for 6 months or a single cycle of a modified fasting-mimicking diet for 7 days followed by a Mediterranean diet for 6 months. Twenty patients between 18 and 67 years of age were included in each group. Patients in the KD group were recommended an average daily intake of <50 g carbohydrates, >160 g fat, and <100 g protein, received detailed information about nutritional facts including glycemic load from a nutritional coach and learned how to handle carbohydrates. The KD cohort displayed clinically meaningful improvements in the health-related quality of life questionnaire (MS-54) at 3 months, which included the overall quality of life, change in health, a physical health composite, and a mental health composite. However, in the patients on fasting-mimicking diet even better results could be obtained in most of the quality of life-scores.

Adverse events (most commonly airway infections) were reported in 92%, 78%, and 78% of patients from the control diet, fasting-mimicking diet, and KD cohort, respectively, whereas severe adverse events such as lower urinary tract infections occurred in 8%, 16%, and 11% of included patients, respectively. Overall, the interventions were tolerated well, as evidenced by high adherence rates (control diet, 60%; fasting-mimicking diet, 100%; KD, 90%). In addition to increased β-hydroxy-butyrate plasma concentrations in the KD group, the researchers detected a mild median reduction of 0.5 points in the EDSS scores in the KD group at 6 months, but not in the cohort receiving a fasting-mimicking diet. During the 6-month study period, the authors observed a total of eight relapses, namely four in the control diet group, one in the KD cohort, and three in the patients on fasting-mimicking diet [9].

In 2021, Lee et al. [19] conducted a randomized controlled study that included 15 patients aged 36–63 years and compared the effects of KD in participants with MS who were randomized into three groups: Modified medium-chain triglyceride (MCT)-based ketogenic diet (a ketogenic version of the Wahl's Paleolithic diet), modified Paleolithic diet, and a usual control diet group. Diet counseling was provided to participants in the study diet groups by registered dietitians. The participants were asked to keep food logs, to record daily pain and fatigue levels, and to complete weighed food records. Motor outcome measures and quality of life assessments were also conducted. The initial inclusion criteria were only primary progressive MS, but was expanded to relapsing-remitting, and secondary progressive MS later due to low enrollment. The Paleolithic diet revealed more disease-alleviating results than the KD in terms of fatigue and quality of life. However, it is important to note that the Paleolithic diet group contained the most disabled and fatigued individuals and displayed the lowest overall quality of life-scores at baseline. While no group showed significant improvement in terms of EDSS scores and the 9-Hole-Peg-Test, which were measured at weeks 4, 8, and 12 following start of respectively diet, the KD group achieved and maintained nutritional ketosis throughout the study. At 4, 8, and 12 weeks, the mean plasma ß-hydroxy-butyrate concentration, as a marker of ketosis, was significantly higher in the KD group if compared to the Paleolithic diet and the control groups. Furthermore, the patients from the KD cohort displayed lower blood glucose concentrations at 4 and 12 weeks compared to their baseline levels. These lower glucose levels were also significantly lower as compared to those measured in the Paleolithic diet group at the 4-week time point. Furthermore, the KD group showed lower plasma insulin concentrations at 4 and 12 weeks compared to baseline values, whereas the patients from the control and the Paleolithic diet groups did exhibit significant differences in insulin concentrations measured at any time point [19].

In 2017, Svidinski et al. [20] investigated 14 healthy volunteers and 25 patients with relapsing-remitting MS for the composition of their colonic microbiome by using ribosomal RNA-based fluorescence in situ hybridization (FISH) probes. After a baseline assessment, MS patients received a KD for 6 months. The KD was designed to achieve modest ketosis (i.e., ≥500 μmol L⁻¹ ß-hydroxy-butyrate in the blood and ≥500 μmol L⁻¹ acetoacetate in the urine). The instructions and assistance criteria for implementing the KD were comparable to those published by Choi et al. [9]. After analyzing the microbiome in stool samples of 10 randomly selected MS patients and comparing them to those of healthy controls, the authors found that the diversity of all analysed substantial commensal bacterial groups in fecal samples taken from the MS patients was reduced by 36%, and the mean total concentrations of respective bacterial groups were reduced by 24% [20]. Within two weeks on KD, however, the concentrations of nearly all analyzed fecal bacterial phyla had decreased, but started to increase as early as 12 weeks following the start of the KD. At weeks 23 and 24, respective commensal bacterial phyla reached fecal concentrations comparable to those obtained in healthy controls, and were significantly higher than baseline values measured in MS patients before initiation of the diet. Of note, fecal Akkermansia concentrations decreased during the KD [20].

In 2018, Bock and colleagues conducted a randomized three-armed clinical trial, examined distinct genes that are coding for pro-inflammatory (ALOX5, COX1, COX2) and anti-inflammatory (ALOX15) eicosanoids in relapsing-remitting MS patients on KD [21]. The three cohorts were either subjected to an adapted KD [9] for 6 months, a fasting-mimicking diet (7-day fasting followed by a normal diet), or a normal control diet [21]. A nutritional coach assisted the total number of 24 patients, of which 11 were on the adapted KD, 5 were on FMD, and 8 were controls.

The results of the adapted KD and the fasting-mimicking diet groups were pooled for analysis given the small sample sizes (due to patient drop-out and loss of blood samples). The expression levels of all target genes under investigation showed a non-significant trend for a positive correlation with the EDSS score. At 6 months, ALOX5 as well as COX1 expression levels were inversely correlated with MSQOL-54, a multi-dimensional health-related quality of life measure that combines both, generic and MS-specific items. Of note, no significant effects on fasting blood sugar and insulin concentrations were reported upon respective dietary regimens [21].

In 2022, Bock et al. [22] expanded their research and retrospectively evaluated the impact of a KD on the serum neurofilament light chain (NfL), a marker of neuroaxonal damage. A total of 40 participants (mean age of 43.6 years) was enrolled in a three-armed parallel-group, single-center, controlled, and randomized clinical trial and subjected to dietary interventions reported in earlier studies conducted in 2016 [9] and 2018 [21]. At baseline, the serum NfL concentrations did not differ between the three groups, indicating that the initial dietary habits of the patients had no impact on the neuroaxonal damage marker. Also at baseline, the authors found a significant association between serum NfL and age, disease duration, and the Multiple Sclerosis Functional Composite assessing leg function/ambulation, arm/hand function, and cognitive function. The majority of subjects in this study's KD group was untreated (29%) or on low-efficacy disease-modifying therapy (48%). Regarding the effects of the dietary intervention on serum NfL, the results showed that patients of the adapted KD group displayed a significant decrease in serum NfL concentrations after 6 months on diet if compared to baseline. In contrast, the caloric restriction group and control diet cohort did not show significant changes in their serum NfL concentrations throughout the study period. Furthermore, when comparing the adapted KD group with the control diet cohort, the researchers found a significant decline in serum NfL concentrations in the former versus the latter after 6 months of the dietary intervention [22].

The impact of a KD on serum NfL was also addressed by Oh et al. [23]. Sixty-five subjects with relapsing-remitting MS were enrolled in the study with inclusion and exclusion criteria as outlined previously [18]. Importantly, in comparison to the study by Bock et al. [22], patients were either untreated or had to exhibit clinical- and neuroimaging-stability on their current disease-modifying therapy, for at least 6 months before enrollment. The authors did neither find differences in mean serum NfL concentrations at 6 months following KD initiation if compared to baseline values, nor did the individual changes in serum NfL levels correlate with changes in the clinical outcome of the respective patient. These results indicate that the serum NfL concentrations were not affected by the KD in the enrolled relapsing-remitting MS patient cohort. Subjects that exhibited rather pronounced ketosis as indicated by serum ß-hydroxy-butyrate concentrations of ≥1.0 mmol L⁻¹ at 3 and 6 months on KD, however, displayed a more distinct decline in serum NfL concentrations at 6 months when compared to patients with a ß-hydroxy-butyrate level of <1.0 mmol L⁻¹ at the 3-months and 6-months time points [23]. The authors hypothesized that the degree of ketosis might be considered as the main mediator between KD and neuroprotection.

Discussion

Main findings of the search

Most of the studies reviewed here provide evidence for the beneficial effects of a KD on various parameters assessing the clinical course of MS. In murine EAE and CPZ induced MS, improvements in cognitive [13, 14, 19] and motor [10, 11, 14] symptoms were accompanied by attenuated pro-inflammatory cytokine secretion [10, 11, 13] and enhanced oligodendrocyte maturation and survival [13, 14]. Several clinical studies in patients also reported favorable effects of the KD on MS symptoms such as improved EDSS scores [9, 15, 16, 18] and self-reported quality of life [9, 18], as well as of less pronounced depression [16, 18] and fatigue [16, 18]. One study [19] assessing the effects of KD on fatigue, quality of life, and EDSS scores in MS patients, however, could not find significant changes in the clinical outcome of these categories after 12 weeks on KD. Importantly, the KD proved to be safe and feasible in MS patients, as evidenced by a low number of serious adverse events and high diet adherence.

When assessing the neuro-cellular modulating effects of KD, two studies [13, 14] found a reduced activation of microglia and reactive astrocytes following KD, as well as an enhanced differentiation and maturation of oligodendrocytes. The mechanisms by which ketone bodies could modulate microglial cells have been addressed very recently [24], but require further research.

To date, information regarding the effects of KD on the gut microbiome in MS patients are scarce. However, one study addressed this topic [20] and found that except for Akkermansia, virtually all main commensal bacterial phyla increased during KD in MS patients. Since increased abundance of Akkermansia in the gut microbiome of MS patients has recently been reported [25, 26], the Akkermansia-lowering effects of KD might open interesting insights into the bacteria-host interactions during MS immunopathogenesis and could open novel treatment approaches, but require further investigations in in vitro, murine, and human intervention studies.

To study the beneficial effects of the KD for MS patients in more detail, larger cohorts are crucial. In the protocol of a three-armed randomized controlled trial by Bahr et al. [27], for instance, the authors calculated that an enrollment of 111 participants would be necessary to attain a statistical power of 80%. This calculation takes an assumed patient drop-out rate of 10% into account.

Limitations

Our literature review has been limited by the following points. Firstly, only a confined number of studies could be found addressing our initially raised research question. Secondly, the relatively small sample sizes in most studies further hamper the understanding of the effects of KD on the immunopathogenesis and clinical course of MS. Thirdly, differences in experimental set-ups and patient characteristics including EDSS score, disease course, variations in the composition of the diet, and duration of treatments, for instance, make it difficult to draw definite conclusions across studies. Fourthly, research mistakes cannot be ruled out given that data collection and analyses were performed by a single investigator and relevant publications might have been missed.

Conclusions and outlook

The KD has been used as anti-convulsive treatment measure for long. In recent years, various reports of the anti-inflammatory and potentially neuroprotective effects of the KD in neurological diseases such as Parkinson's disease and Alzheimer's disease but also of MS have emerged. The existing literature supports the safety and feasibility of the KD in patients living with MS. In consistency with preclinical studies in animal MS models, clinical trials provide evidence for disease including disability-alleviating, neuroprotective, and cell metabolism-preserving effects. Since the data from the literature to date is limited and most studies were conducted with low numbers of MS patients and were rather exploratory in nature, further studies with larger cohorts are needed to gain a better understanding of the mechanisms by which the improvements of the MS disease course upon KD are achieved. The interaction of KD with the gut microbiome or the modulating effects of ketone bodies on microglial cells might be of particular interest to develop novel treatment concepts in the combat of MS.

Ethics statement

Not applicable (literature survey).

Conflict of interests

SB and MMH are Editorial Board members. Therefore, the submission was handled by a different member of the editorial board, and they did not take part in the review process in any capacity.

Funding

None.

Authors' contributions

JDB conceived and designed the survey, wrote the paper. SB provided critical advice in design of the survey, edited paper. MMH supervised the survey, co-wrote the paper.

List of abbreviations

BMI

body mass index

CNS

central nervous system

CPZ

cuprizone

EAE

experimental autoimmune encephalomyelitis

EDSS

Expanded Disability Status Scale

FISH

Fluorescent in situ hybridization

KD

ketogenic diet

MRI

magnetic resonance imaging

MS

Multiple Sclerosis

MWMT

Morris Water Maze Test

ROS

reactive oxygen species

SCOT

succinyl-CoA:3-ketoacid CoA transferase

SIRT1

sirtuin 1

sNFL

serum neurofilament light chain