Alternative Diets for Drug-Resistant Epilepsy Beyond Ketogenic Diet and Fish Oil: Early Clinical Evidence Review

Epilepsy affects more than 50 million people across the globe including both children and adults [1]. While antiepileptic drugs (AEDs) are available, about 30–40% of epilepsy turns into drug-resistant epilepsy (DRE) [2]. DRE poses a significant threat to quality of life and creates a heavy economic and social load on families and society. This clinical challenge has driven scientists to explore novel therapeutic avenues, including dietary interventions [3,4,5,6,7,8,9,10,11,12].

Since the 1920s, the ketogenic diet (KD) has developed and has quickly become a cornerstone dietary approach to manage DRE [3,4,5,6,10,13,14,15,16]. The KD, a high-fat, low-carbohydrate, and controlled protein diet, is thought to exert its anticonvulsant effects by, among other mechanisms, boosting the action of GABA nerve signals through reducing available oxaloacetate for the aspartate aminotransferase reaction [13,17,18,19]. GABA is a known brain chemical that slows nerve activity and helps stop seizures [20,21]. While the classic ketogenic diet (KD) has served as a therapeutic cornerstone since the 1920s, its long-term clinical utility is often hampered by poor palatability and restrictive protocols that lead to high attrition and low compliance [14]. Similarly, while omega-3 supplementation (fish oil) has shown anti-inflammatory potential, clinical results remain variable across different seizure types [3,7,22,23]. Clinical trials have shown the promise of KD and fish oil supplementation, two of the most popular dietary interventions for epilepsy [3,6,7,15,22,23,24]. The effectiveness and quality of evidence of these diets have been updated in recent systematic meta-analyses [9,15,25]. Consequently, there is an acute clinical need to investigate dietary solutions that are more tolerable and target novel pathophysiological pathways, such as the gut-microbiota-brain axis and metabolic fuel shifts.

These emerging interventions, often supported by mechanistic insights or strong animal data, have advanced into human clinical trials [11,24,26,27,28,29,30,31]. Notably, probiotics and symbiotics have been investigated in managing epilepsy by targeting the gut microbiota via the gut–brain axis [11,26,32,33]. Scientists have explored supplementation of branched amino acids (BCAAs) as an adjunct therapy for epilepsy [31]. Animal studies showed that BCAAs can promote brain GABA production by promoting glutamate dehydrogenase activity [34,35,36,37]. Additionally, alternative diets have been explored in animal and human studies [24,28,29,30]. Notably, the prevalence of celiac disease in patients with epilepsy is estimated to be as high as 2.5%, compared to approximately 1% in the general population [29]. This provides a strong clinical rationale for investigating gluten-free diets (GFD) in this sub-population, where epilepsy may manifest as a neurological symptom of systemic gluten sensitivity [38,39].

A critical gap in the literature exists regarding these emerging strategies. While the effectiveness and quality of evidence for standard KD, its carbohydrate-modified versions (e.g., Modified Atkins Diet), and fish oil supplementation have been extensively updated in recent systematic reviews and meta-analyses, the clinical effectiveness and safety of these other novel dietary strategies have not been systematically evaluated. Although these novel dietary strategies show promise and are supported by mechanistic insights and animal studies, their supporting evidence in humans consists of disparate studies with varied designs. Currently, there are no comprehensive reviews available to examine these specific human studies. This review is therefore crucial to synthesize the available evidence, rigorously assess its quality and certainty, and provide clear guidance for clinical practice and future research. This manuscript systematically reviews the preliminary clinical evidence for these emerging strategies, replacing broad categorical summaries with a deep, individual study analysis to evaluate their potential role in DRE management.

The central finding of this systematic review is the marked conflict between the plausible mechanisms of alternative diets and the current lack of high-quality clinical evidence. interventions such as probiotic supplementation targeting the gut–brain axis or BCAA additions targeting GABA synthesis are supported by robust animal models [32]. In a 2009 single-arm pre-post trial [31], BCAAs plus KD appeared promising, over 50% of participants (8 out of 17) experienced a ≥50% reduction in seizure frequency, and 18% achieved complete seizure freedom following 24-month intervention. However, we remain cautious when interpreting the findings due to small sample size and lack of a control group, limiting the study’s strength. Previous studies suggest that BCAAs could upregulate glutamate decarboxylase and subsequently increase GABA [42]. BCAAs could also modulate serotonin and dopamine synthesis and further influence brain excitatory/inhibitory balance by reducing aromatic amino acid uptake [43,44]. However, current evidence remains preliminary and limited to support BCAAs for epilepsy.

A recent case report showed that dietary supplementation of MCTs remarkably reduce seizure [45], suggesting that more ketogenic MCTs may offer more benefits than traditional vegetable oils used in KD. Two trials followed this direction and examined the effects of MCTs on epilepsy [28,30]. One RCT found that 12-month MCT-based KD was not superior to classical vegetable oil-based KD, though both diets significantly reduced seizure in children with DRE [30]. The other trial investigated the effects of MCTs supplementation on a regular diet and reported a 42% reduction in seizure frequency [28]. However, this trial was at a high risk of bias due to its a small sample size (n = 9) and a lack of a control group (pre-post design). Given this limitation, we find the evidence supporting MCTs for epilepsy is very week and inconclusive.

In addition to MCTs, olive oil, a component of Mediterranean diet has been investigated for its potential in epilepsy management [24]. Olive oil is rich in mono glyceride and has been widely studies for its antioxidant and anti-inflammatory activities [46,47]. The Guzel trial showed a very high respondent rate to olive-oil based KD: 83% of participants (389 children) experienced a >50% reduction in seizure frequency after 12-month dietary intervention. While the findings results are promising, the study design presents a high risk of bias, and further validation through a rigorous RCT is necessary to establish the efficacy of an olive oil-based KD diet in epilepsy management. Other than alternative modifications of KD, dietary supplementation of bioactive components have also been explored. Both Gómez-Eguílaz 2018 [11] and Shariatmadari 2024 [26] trials demonstrated that supplementation of probiotics and synbiotic significantly reduced seizure frequency and improved quality of life of epilepsy patients [11,26]. The probiotic approach targets gut microbiota which is believed to play a role in brain function and neurological disorders including epilepsy [32,48]. Certain gut bacteria have been known for their capability to produce or modulate neurotransmitters such as GABA, glutamate, and serotonin, which are involved in neuronal excitability and seizure thresholds [49]. Though both the trials suggest a potential of probiotics or synbiotic for epilepsy management, the current evidence is preliminary and with high risk of bias as both trials lacked placebo control. Furthermore, the trials used different probiotic/synbiotic formulations, making direct comparisons and definitive conclusions challenging.

The progression of dietary interventions for epilepsy beyond KD is marked by a shift towards exploring modifications to improve KD, and investigating alternative dietary strategies based on emerging understandings of epilepsy pathophysiology (MCT, probiotics, gut–brain axis, glutamate, etc.). Our synthesis reveals that while preliminary signals exist, most studies are small, open-label, and uncontrolled, which significantly limits their certainty. A significant methodological concern identified across the 8 studies is the high risk of bias in NRSIs. In the field of epilepsy, the “placebo effect” can account for up to a 20–30% reduction in seizure frequency in clinical trials. Therefore, the responder rates reported in uncontrolled studies must be interpreted with extreme caution until validated against a placebo in a blinded RCT. The only RCT in this review that compared a restrictive diet (low glutamate) to a control group yielded a negative result, highlighting the risk of overstating efficacy based on uncontrolled pilot data. Furthermore, there is a clear imbalance in the evidence. The largest study (Guzel 2019 [24]) focuses on a modification of the established ketogenic diet, where the sample synbiotics are represented by very small cohorts (n < 45). This disparity makes broad clinical conclusions difficult. Future research should prioritize rigorous RCTs with well-defined protocols and strong mechanistic rationales to determine the true clinical value of these alternative dietary strategies.

The findings of this review indicate a significant gap between the strong mechanistic rationale for alternative diets—such as the gut–brain axis for probiotics or GABA modulation for BCAAs—and the actual clinical evidence available to date. The high responder rates seen in uncontrolled trials (notably olive oil-based KDs) likely reflect a combination of true biological signal and the substantial placebo effect characteristic of epilepsy research. Consequently, these interventions should be interpreted as promising experimental targets rather than established clinical tools. The negative result from the only blinded RCT on glutamate restriction serves as a critical caution against adopting alternative diets based solely on mechanistic theory.

The clinical evidence for emerging dietary interventions beyond standard KD and fish oil supplementation is preliminary and of low certainty. While pilot data for olive oil-based KDs and probiotics are encouraging, they are currently insufficient to support routine clinical use. The negative result from the only RCT on glutamate restriction serves as a reminder that mechanistic theory requires rigorous validation. Future research must prioritize blinded RCTs that focus on specific patient sub-populations.

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/neurolint18010009/s1↗, Table S1: PRISMA 2020 Checklist.

Conceptualization, X.M. and K.Z.; methodology, X.M. and K.Z.; validation, X.M. and K.Z.; formal analysis, X.M. and K.Z.; data curation, X.M. and K.Z.; writing—original draft preparation, X.M.; writing—review and editing, X.M. and K.Z.; funding acquisition, X.M. All authors have read and agreed to the published version of the manuscript.

Data sharing is not applicable to this article as no new data were created or analyzed in this study.

There are no relevant financial or non-financial competing interests to report.

This study was funded by the Stable Support Project of Shenzhen (Project no. 20231122135121001) and Multidisciplinary epilepsy diagnosis and treatment team of Prof. Wang Yuping from Xuanwu Hospital Capital Medical University (SZSM202003006). The funders had no role in study design, data collection and analysis, and decision to publish or preparation of the manuscript.