As ketogenic diets grow in popularity, a new review sheds light on the short-term side effects of keto-induction and offers science-backed solutions to help manage the dreaded “keto flu.”

Symptoms during initiation of a ketogenic diet: a scoping review of occurrence rates, mechanisms and relief strategies. Image Credit: eldar nurkovic / Shutterstock

In a recent study published in the journal Frontiers in Nutrition, researchers at the University of Bergen, Norway, evaluated symptoms during the transition to a ketogenic diet (KD), i.e., keto-induction.

KD is characterized by a higher intake of fats, moderate intake of proteins, and lower consumption of carbohydrates. The low dietary carbohydrate maintains the predominant oxidation of fatty acids, mimicking the state of starvation that involves ketogenesis in the liver. Ketogenesis involves the synthesis of ketone bodies, such as acetone, β-hydroxybutyrate, and acetoacetate, which serve as an alternative fuel source for cells.

KDs have been extensively explored as therapeutic options for cancer, obesity, diabetes, dementia, and cardiovascular disease. For more than a century, these diets have been used to treat intractable epilepsy in adults and children. Research also shows the neuroprotective effects of KD, with implications for Parkinson’s disease, Alzheimer’s disease, glioma, and migraine, among others.

However, a potential obstacle to KD adoption is the experience of adverse symptoms, referred to as keto-flu, in the first weeks. These symptoms include fatigue, brain fog, lethargy, lightheadedness, headache, mood changes, halitosis, muscle cramps, reduced exercise capacity, diarrhea, and constipation. These arise due to physiological shifts such as electrolyte loss (e.g., sodium and potassium), reduced glucose availability during metabolic adaptation, and transient alterations in gut microbiota composition. Documenting these symptoms during keto-induction and understanding the underlying mechanisms may help prevent or treat these symptoms.

The Study and Findings

In the present study, researchers reviewed the literature on keto-flu regarding its underlying physiological symptoms and evidence for symptom alleviation and treatment. First, they searched the Web of Science, Medline, and Embase databases for relevant studies. Eligible studies were those in which one of the study groups followed a KD and those exploring the adverse effects, mechanisms, and interventions for relief of symptoms of keto-induction.

Non-English studies, animal studies, studies exclusively exploring long-term benefits and side effects of KD, and those involving high-fat diets, not low in carbohydrates, were excluded. Overall, the researchers identified more than 5,700 records from database searches. Following deduplication and title/abstract screening, 115 articles were assessed for eligibility, 89 of which were included in the study.

These studies described keto-induction as involving a range of transient symptoms. These symptoms occur within two to three days of KD initiation and resolve within two to four weeks, with minimal or no intervention, driven by mechanisms such as natriuresis (excessive sodium excretion), hypovolemia (low blood volume), reduced energy substrate availability during the metabolic shift from glucose to ketones, and changes in gut microbiome diversity. Rarely were symptoms severe enough to warrant KD discontinuation. Symptoms included muscle cramps, fatigue, mood changes, lightheadedness, constipation, halitosis, and diarrhea, among others.

Macronutrient ratios used in some standard diet formulations [calculated by mass, not energy percentage of total energy].

Some studies reported other symptoms, including hypoglycemia, kidney stones, acidosis, prurigo pigmentosa (a rare skin rash linked to ketosis), emesis, and nausea. In some studies, halitosis was reported as a common side effect of keto-induction. It was reported to occur 38% of the time in people on KD compared to 8% for those on a low-fat diet. Gastrointestinal adverse effects were among the most frequent keto-induction symptoms.

Constipation was reported to occur 15% to 63% of the time in pediatric populations, while it was described as a common side effect in healthy adult populations. Vomiting and nausea were also reported as common side effects in children. Further, dizziness and lightheadedness were common symptoms in adults but not in children. Hypoglycemia was reported to mainly affect the pediatric population, with no direct reports of occurrence in adults, though symptoms like fatigue and dizziness in adults may reflect mild glucose fluctuations.

The occurrence of kidney stones was 2.5% to 4% and ≤25% in studies of long-term adherence to KD. However, short-term studies did not report kidney stones as a side effect. Notably, some populations appeared to be more prone to adverse symptoms. Age was a significant predictor of adverse effects in a study involving epileptic children. Consistently, another study also concluded that children under three years had a higher burden of adverse events than those aged over three years.

A classic KD has a 4:1 ratio of fats to non-fats based on molecular mass. However, some studies reported compositions as energy percent. As such, direct comparisons of KDs between studies should not be made without examining their definitions. Some studies also noted transient increases in LDL cholesterol during keto-induction, though these typically stabilized over time. Further, traditional KD initiation involves an initial fasting period, followed by a gradual increase in calories.

Notably, a fasted KD initiation has a significantly shorter time to ketosis than a gradual transition. However, children experience a greater burden of lethargy and hypoglycemia under this regimen. Other initiation strategies involve initiation at full energy content with the intended fat-to-protein and carbohydrate ratio or a gradual increase in this ratio.

Medium-chain triglycerides (MCTs) have become a popular supplement to KDs as they allow for a higher intake of protein and carbohydrates while maintaining ketone levels and efficacy in epilepsy treatment. Notably, MCTs have been proposed as a possible intervention to alleviate keto-induction symptoms, though they may cause gastrointestinal distress in some individuals if not titrated gradually. Available data suggest the benefits of titrated MCT supplementation with elevated ketone synthesis as an energy substrate in low-glucose conditions.

Conclusions

Taken together, transient symptoms related to keto-induction are common, and various mechanisms have been proposed for these symptoms. The reviewed populations were highly heterogeneous in terms of health status, age, diet strictness, initiation method, KD formulation, ad libitum feeding, food choices, and caloric restriction.

Approaches to reduce keto-induction symptoms include supplementation of ketone salts and MCTs, targeted electrolyte replacement (e.g., sodium/potassium), and avoidance of conventional fasted initiation of KD. While there is physiological rationale for ketone ester and electrolyte supplementation, clinical studies are lacking to confirm their efficacy. Short-term lipid profile changes, such as elevated LDL cholesterol, were reported in some studies but generally resolved with sustained adherence. Further research is required to fill the existing knowledge gaps and minimize the hurdles preventing KD adoption.