Bipolar disorder is a common mood disorder characterised by periods of depression along with episodes of elevated mood, known as mania or hypomania. People with bipolar disorder face not only mood instability but also an elevated risk of obesity, type 2 diabetes, and cardiovascular and respiratory disease. Deaths due to physical illnesses in people with a diagnosis of bipolar may be twice as high as those found in the general population. Several first-line medications for bipolar disorder have significant metabolic side-effects and additional treatment options are urgently needed.
Enter the ketogenic diet: a high-fat, low-carbohydrate intervention which has been shown to be beneficial in reducing seizures in those with epilepsy. Epilepsy and bipolar disorder appear to share some overlapping pathophysiological mechanisms and many of the mood-stabilisers used in bipolar disorder are also anti-epileptic medications. Hence, the interest in the potential of a ketogenic diet for improving symptoms of bipolar disorder.
Ketones are a type of acid that your body produces when it breaks down fat for energy, especially when glucose (sugar) is not readily available. Normally, the liver produces ketones, and the body uses them for fuel, but excessive levels can be harmful. A ketogenic diet forces the body to burn fat for energy instead of carbohydrates, leading to the production of ketones in the liver. This metabolic state is called ketosis.
Emerging research into the effects of a ketogenic diet have reported neuroprotective and anti-inflammatory mechanisms along with a positive metabolic effect. A number of case reports, observational studies and pilot data suggest this diet may help stabilise mood and reduce metabolic dysfunction in bipolar disorder. However, robust clinical evidence is still in its infancy.
This recent pilot study (Campbell et al., 2025) aimed to assess the clinical and metabolic effects of a ketogenic diet in people with bipolar disorder who were clinically stable (euthymic). They also studied changes in brain chemistry using magnetic resonance spectroscopy (MRS). While exploratory in nature, this study offers early insights into whether ketogenic interventions could one day become part of the psychiatric toolkit.
Emerging research into the effects of a ketogenic diet have reported neuroprotective and anti-inflammatory mechanisms along with a positive metabolic effect.
Methods
This was a 6 to 8 week, single-arm, open-label pilot study involving 27 euthymic adults with a diagnosis of bipolar disorder. Participants received individualised ketogenic diet plans, aiming for nutritional ketosis (blood ketone levels of 1–4 mmol/L), guided by a study dietitian. Clinical assessments, metabolic blood tests, and brain MRS scans were conducted before and after the intervention. Daily ecological momentary assessments (EMAs) were used to track mood, energy, impulsivity, and anxiety. Actigraphy devices monitored sleep and activity.
The primary aim of the study was to evaluate feasibility, safety and acceptability (previously published; Needham et al., 2023), but this report focused on secondary clinical and metabolic outcomes to inform future randomised controlled trials.
Participants received individualised diet plans.
Results
Of the 27 participants recruited, 26 started the diet and 20 completed the full 6–8-week programme. Among completers, the diet was well-tolerated and all patients were euthymic at follow-up. There were no significant changes in the median scores on validated mood scales (Affective Lability Scale-18 (ALS-18), Beck Depression Inventory (BDI), Young Mania Rating Scale (YMRS)) indicating no major shifts in depressive or manic symptoms. One patient did report an episode of hypomania during the study period.
Real-time daily symptom tracking with EMA revealed interesting findings. In the 14 participants who provided reliable data, higher daily ketone levels correlated with improved self-rated mood (r = 0.21), higher energy (r = 0.19), and reduced impulsivity (r = −0.30) and anxiety (r = −0.19), all statistically significant (P
Metabolic benefits were also notable. Participants lost an average of 4.2 kg (P by 7.4 mmHg (P = 0.041). Changes in other metabolic parameters (diastolic blood pressure, fasting insulin, fasting glucose, HBA1c, blood lipids) were not observed. Importantly, there were no adverse shifts in cholesterol or triglyceride levels. This is encouraging given concerns about a potential adverse impact of the ketogenic diet on lipid profile.
Blood serum analysis showed expected increases in ketone bodies and a significant reduction in glucose and in lactate: a marker previously linked to mitochondrial dysfunction in bipolar disorder. Brain MRS scans found decreases in glutamate and glutamine levels in the anterior (−11.6%) and posterior cingulate cortices (−13.6%), regions implicated in mood regulation. Reduced myo-inositol and choline were also observed. These neurometabolic changes are consistent with prior brain MRS findings of elevated glutamate in bipolar disorder and with observations that changes in glutamate metabolism may be an important mechanism of action shared between anti-seizure and mood stabilising medications.
Notably, one-third of completers chose to continue the ketogenic diet beyond the study period, citing perceived mental health and energy benefits and also suggesting it was an acceptable alternative diet. While the actigraphy data was incomplete, it could be included in future research to explore impacts on circadian rhythms.
A range of benefits were observed including weight loss of on average 4.2kg.
Conclusions
This pilot study provides preliminary evidence that a ketogenic diet is safe, well-tolerated, and potentially beneficial for mood symptoms and metabolic health in people with bipolar disorder. While no significant changes in standard clinical mood scales were observed, correlations between ketone levels and momentary daily mood, energy, impulsivity and anxiety ratings, along with improvements in weight and blood pressure, are promising. Decreases in glutamate and other brain metabolites suggest possible mechanisms.
However, given the small sample size and lack of a control group, these findings should be interpreted cautiously. Randomised controlled trials are now warranted to confirm these early signals and explore this novel therapeutic approach.
Given the small sample size and lack of a control group, these findings should be interpreted with caution.
Strengths and limitations
Strengths
- Novel study, exploring the impact of a ketogenic diet on a comprehensive variety of parameters in bipolar disorder. Authors have previously published safety and feasibility data (Needham et al., 2023) and a process evaluation of the study (Rigby et al., 2025) along with the current paper on clinical and metabolic outcomes.
- Adherence to a ketogenic diet was high: 91% of daily readings indicated ketosis (Needham et al., 2023), likely reflecting the comprehensive, individualised support from experienced dieticians. Additionally, one third of patients chose to continue the diet following the study, indicating satisfaction with the outcomes.
- The study’s use of multimodal assessments: including clinical scales, daily mood and symptom tracking, blood biomarkers, metabolomics, and brain MRS imaging, provides a comprehensive picture of both behavioural and biological outcomes.
- Real-time ecological momentary assessment (EMA) allowed for nuanced understanding of how ketone levels may influence mood and other symptoms on a day-to-day basis.
Limitations
- As a pilot, the study was not powered to detect significant clinical effects, and all findings should be considered exploratory.
- The open-label, single-arm design without a control group limits the ability to attribute observed changes to the ketogenic diet alone.
- Participants were self-selected, largely recruited through a charity, Bipolar Scotland, and may have been more motivated or health-conscious than typical clinical populations.
- The relatively short duration (6–8 weeks including an initial adaptation phase), may not be sufficient to fully evaluate long-term mental health or metabolic impacts.
- Actigraphy data had poor compliance and could not be meaningfully interpreted.
- EMA was only available for a subset of participants (n=14), due to early variability in instructions leading to confusion and precluding meaningful analysis. However, this is an important information for informing a future randomised controlled trial.
In summary, while the study design limits causal inference, its methodological diversity and encouraging findings suggest a strong rationale for conducting a larger, randomised controlled trial. These early findings are intriguing enough to merit cautious optimism, but replication in randomised controlled trials is necessary.
Findings suggest a strong rationale for conducting a larger, randomised controlled trial.
Implications for practice
This study points toward a potential paradigm shift in the management of bipolar disorder. Traditionally treated almost exclusively with pharmacological and psychological interventions, bipolar disorder may benefit from the emerging field of nutritional and metabolic psychiatry. We have previously described how gut-based interventions such as probiotics have previously been shown to hold promise for bipolar disorder and nutritional and metabolic approaches are also of major interest in depression. If replicated, the findings from this pilot suggest that the ketogenic diet could serve as a novel, adjunctive strategy targeting both mood stability and physical health.
As a psychiatrist who routinely manages patients with BD and comorbid metabolic issues, this study raises several important considerations for my clinical practice.
- Dual Targeting: The ketogenic diet may improve both mental and physical health; a vital step in a population with high rates of obesity, diabetes, and cardiovascular disease where first-line medications often have adverse metabolic effects.
- Patient Engagement: The relatively high adherence rate and willingness of participants to continue the diet post-study indicate that, when supported, people with bipolar disorder are capable of making substantial dietary changes.
- Mood Tracking Tools: Traditional rating scales failed to detect change, but daily EMA showed mood, energy, and anxiety improvements correlated with ketone levels. This suggests more dynamic, real-time assessment tools may be useful to capture dietary intervention effects in clinical practice. Mood tracking phone applications are often used by patients with BD.
- Caution on Lipids: There are concerns that a ketogenic diet could adversely affect lipid profile, in particular LDL cholesterol. While lipid profiles didn’t worsen in this sample, it would be wise for clinicians to monitor lipid markers closely and consider dietary fat sources (favouring unsaturated fats).
- Practical Implementation: Rolling out ketogenic diets in routine psychiatric care would require significant dietetic input, monitoring protocols, and training. It is not an intervention that most psychiatrists are likely to be comfortable with, without specific training.
Ultimately, while it’s too early to recommend ketogenic diets as standard treatment for bipolar disorder, this study suggests that metabolic interventions should be part of the research and clinical conversation. For now, it may be suitable as an adjunctive, carefully monitored intervention for motivated individuals; ideally under research conditions or with specific clinical training and multidisciplinary support.
Could nutrition play a role in managing bipolar disorder? Early findings suggest the ketogenic diet might support mood and metabolic health, but more research and clinical caution are needed.
Links
Primary paper
Campbell IH, Needham N, Grossi H, et al. A pilot study of a ketogenic diet in bipolar disorder: clinical, metabolic and magnetic resonance spectroscopy findings. BJPsych Open. 2025;11(2):e34. doi:10.1192/bjo.2024.841
Other references
P. Rigby, B., Needham, N., Grossi, H. et al. Pilot study of a ketogenic diet in bipolar disorder: a process evaluation. BMC Psychiatry 25, 63 (2025). https://doi.org/10.1186/s12888-025-06479-y
Campbell, I. H., Needham, N., Grossi, H., Kamenska, I., Luz, S., Sheehan, S.,…Smith, D. J. (2025). A pilot study of a ketogenic diet in bipolar disorder: clinical, metabolic and magnetic resonance spectroscopy findings. BJPsych Open, 11(2), e34, Article e34. https://doi.org/10.1192/bjo.2024.841
Needham, N., Campbell, I. H., Grossi, H., Kamenska, I., Rigby, B. P., Simpson, S. A.,…Smith, D. J. (2023). Pilot study of a ketogenic diet in bipolar disorder. BJPsych Open, 9(6), e176. https://doi.org/10.1192/bjo.2023.568
