Tom Wade MD

*Sodium-glucose cotransporter 2 (SGLT2) inhibitors

6.2. Type 1 diabetes mellitus (T1DM)

Type 1 diabetes mellitus (T1DM) is pathophysiologically distinct from type 2 diabetes mellitus (T2DM). In T2DM, insulin is produced, but target cells exhibit resistance to its action, whereas in T1DM, the primary issue is insulin deficiency resulting from the autoimmune destruction of pancreatic β-cells [185].

Patients with T1DM (who are therefore also taking insulin) who consider switching to the ketogenic diet should exercise particular caution [19]. According to the scientific consensus of the Society of Metabolic Health Practitioners (SMHP), and supported by the emerging scientific evidence, low-carbohydrate diets show promising results in T1DM. The authors emphasise the need for education, improved access to comprehensive information, and support from the healthcare team for patients diagnosed with type 1 diabetes in the context of low-carbohydrate diets [186]. Since dietary carbohydrate supply is very low on the KD, blood glucose levels do not rise nearly as much after meals, and often do not rise at all [71]. The ketogenic diet therefore naturally reduces the need for insulin, one purpose of which is to lower blood glucose levels. Close monitoring and appropriate adjustment of insulin doses by a specialist are therefore necessary to reduce risk of hypoglycaemia. On the other hand, there are concerns that people with T1DM who follow a KD are at increased risk for diabetic ketoacidosis (DKA) [187], and a small number of case studies have documented that this can occur [45,188]. However, as a general rule, risk for diabetic ketoacidosis appears to be low in most cases [189]. For example, one study reported no DKA episodes in a T1DM patient who had followed KD for 10 years [190]. The authors of another paper highlighted that the results (including stable glycemia, absence of health complications, no hypoglycemic symptoms despite low glucose levels, no deterioration in physical or mental performance, and no risk of diabetic ketoacidosis) could be reassuring for clinicians who consider switching their T1DM patients to the ketogenic lifestyle [191]. Among T1DM patients using KD, episodes of DKA are clearly sporadic (as they are observed in only a small subset of T1DM patients following a ketogenic diet, most likely as a consequence of improper dietary implementation) [19], but nevertheless should be kept in mind.

6.3. Hypertensive patients taking antihypertensive drugs

Hypertension is one of the most prevalent health conditions globally, affecting approximately 1.28 billion individuals aged 30–79 years. It is also the leading cause of premature death worldwide [192], with more than half (51.2%) of individuals with hypertension taking antihypertensive medication to lower blood pressure [193].

A number of scientific papers have demonstrated that the KD lowers blood pressure [25,145] even more effectively than the DASH (Dietary Approaches to Stop Hypertension) diet, including a 2023 randomised controlled trial [194]. The situation is therefore similar to patients taking hypoglycaemic drugs: the high efficacy of the KD may necessitate adjusting (lowering) the dosage of antihypertensive drugs or weaning them off altogether, if clinically indicated. Otherwise, a potential risk of hypotension arises [195]. This is reflected in some preliminary studies demonstrating that reducing or discontinuing antihypertensive medication in some patients is not only possible, but necessary [196]. However, it is worth noting that the relationship between the KD and antihypertensive drugs has not yet been thoroughly researched. Nevertheless, it has been observed that when combined with antihypertensive treatment, the ketogenic diet can significantly lower blood pressure [197]. Therefore, caution is required in hypertensive patients taking antihypertensive medication.

6.4. Cholelithiasis

Cholelithiasis is a clinical condition in which crystals composed mainly of cholesterol, bilirubin and bile components, form in the gallbladder or the bile ducts, and it is one of the most common causes of gastrointestinal dysfunction worldwide. Although it is often asymptomatic, when symptoms do occur, they can be acute, chronic, or episodic [198]. For example, biliary colic is the result of mobile gallstones migrating towards the cystic duct and blocking the flow of bile. When the cystic duct is blocked for several hours, cholecystitis occurs. When the gallstones reach the common bile duct, biliary obstruction may develop, leading to jaundice, cholangitis and even pancreatitis, among other complications [199]. Risk factors for cholelithiasis include metabolic disorders, which are estimated to account for approximately 75% of cholesterol stone cases in Western countries. These metabolic disorders encompass conditions such as diabetes, insulin resistance, obesity and hypertriglyceridemia. Risk for cholelithiasis significantly increases when gallbladder contraction frequency is reduced, because this promotes bile stasis, which in turn elevates the risk of deposit formation [198]. This mechanism may be both dependent on and independent from the factors listed above.

According to some sources, based on observations of children with epilepsy who followed the ketogenic diet, this dietary pattern is sometimes considered to increase the risk of cholelithiasis [200]. In fact, an accurate determination of the actual risk in this case remains difficult due to the concomitant use of medication, the presence of comorbidities or disorders, and the varied nature of the ketogenic diet. The state of ketosis is merely one aspect of the KD; the quality of the food consumed may also significantly affect the risks and benefits of this dietary approach. Additionally, the mere presence of symptoms of lithiasis (the risk of which is in fact increased during KD) does not prove causality, as discussed below. However, there are logical arguments, based on the pathogenesis of cholelithiasis and confirmed by human studies, that low-fat diets can promote the development of this disease. This stands to reason, as bile stasis is one of the primary factors in its aetiology and it is fatty foods that initiate gallbladder contraction and prevent excessive bile stasis [201]. Studies in people following low-fat and low-calorie diets for weight loss have shown that in these individuals, the risk of developing cholelithiasis is higher, precisely as a result of bile stasis [202]. It is also noted that even on low-calorie diets, high fat intake can prevent gallstone formation, precisely by inducing gallbladder contractions [203]. At the same time, higher carbohydrate intake, as well as higher glycaemic index and high glycaemic load are known to increase the risk of symptomatic cholelithiasis in men. Some authors emphasise that low-fat and high-carbohydrate diets may not be the optimal dietary recommendation for the prevention of this disease [204].

Paradoxically, given the mechanisms described above, people with cholelithiasis should be particularly cautious when switching to KD. While the risk of developing new stones is probably much lower, because the gallbladder is more strongly stimulated to contract and expel bile on a high-fat KD, existing stones are more likely to migrate into the cystic duct and common bile duct. This can lead to biliary colic, and/or a number of other related symptoms, as described above. The risks associated with the presence of gallstones aside, the increased likelihood of their migration may help empty the gallbladder, which may be beneficial in certain clinical situations. Low-fat diets also carry some risk, as by reducing the chances of migration of existing stones and favouring the development of new ones, they make gallbladder removal surgery more likely. Further research is necessary to better understand the potential risks and benefits of these different dietary patterns.

6.5. History of cholecystectomy

Cholecystectomy (gallbladder removal surgery) is currently performed by laparoscopy. According to the literature, some of the indications for this procedure include acute and chronic cholecystitis, biliary dyskinesia, acalculous cholecystitis, gallstone pancreatitis, gallbladder polyps, and symptomatic cholelithiasis, a condition described in the previous section [205]. After cholecystectomy, gallbladder functions related to bile accumulation and ejection into the duodenum are taken over by the liver. Whereas gallbladder contractions release bile infrequently but abundantly, the liver secretes bile continuously but in small amounts [206,207], which is simultaneously associated with an increased risk for a number of liver diseases [208].

In people with no gallbladder, the ketogenic diet is possible, although care must be taken. There have been few studies on people with a history of cholecystectomy who used the ketogenic diet. However, one study described the case of a 5-year-old boy who used this diet for epilepsy and was able to continue this dietary approach after cholecystectomy without further complications [209].

As regards the bile secretion mechanisms after cholecystectomy, several key aspects should be considered by people who consider switching to the ketogenic diet. Firstly, given the continuous but less abundant bile secretion by the liver after cholecystectomy [206], it seems logical to introduce fat gradually and to determine the body’s tolerance. A symptom known as fatty diarrhoea (steatorrhea) is often closely associated with dietary fat intake in these individuals [210], which can be used to help determine an individual’s tolerance for dietary fat. It also makes sense to spread daily fat intake over 3–4 smaller meals per day, as slow-flowing bile from the liver may have more difficulty digesting a daily portion of fat in 1 or 2 meals per day. Additionally, increasing the proportion of medium-chain triglycerides (MCTs) in the diet seems like a good idea too, as MCTs do not require bile for digestion [211,212].

6.6. Electrolyte deficiency

Electrolytes are essential minerals for normal bodily functions. They include sodium, potassium, magnesium, calcium, chloride, phosphate, and bicarbonates. Excessively low (as well as excessively high) electrolyte levels interfere with normal body functions and can sometimes cause life-threatening complications [213]. Among all electrolyte disorders, sodium deficiency (blood sodium level below 135 mmol/l), known as hyponatraemia, is the most common [214], and is associated with increased morbidity and even increased risk of mortality [215]. Another common electrolyte disorder is hypokalaemia, or potassium deficiency (defined as a blood potassium level below 3.5 mmol/l). Severe hypokalaemia can be life-threatening, as it can cause fatal arrhythmias or respiratory muscle paralysis. However, in most cases hypokalaemia is fairly mild, with symptoms including fatigue, muscle cramps, weakness, palpitations or constipation [216]. Magnesium deficiency (hypomagnesaemia), defined as a blood magnesium level below 1.46 mg/dl, can cause a wide range of both mild and severe clinical symptoms, from weakness and mild muscle tremor to cardiac ischaemia and even death [217].

6.7. Cardiac arrhythmias

Cardiac arrhythmias comprise a broad spectrum of heart rhythm disturbances and abnormalities. Based on heart rate, a distinction is made between bradyarrhythmias (heart rate below 60 beats per minute) and tachyarrhythmias (heart rate above 100 beats per minute); and these are both further subcategorized by origin or associated syndromes, among other factors [219]. Arrhythmias can involve a wide range of symptoms, such as palpitations, dizziness, breathlessness, fainting, chest discomfort, fatigue, anxiety, among others. Some arrhythmias may even be asymptomatic and thus go unnoticed [220].

People with cardiac arrhythmias wishing to switch to the ketogenic diet should take precautions, especially in the initial stages when the body typically loses more water and electrolytes (for details, see section 4.6), because deficiency of electrolytes such as potassium [216], sodium [221] or magnesium [222], can lead to or exacerbate arrhythmias. Consequently, in individuals with abnormal heart rhythms, the negative effects of electrolyte disorders may be more severe [223,224]. Therefore, when switching to KD, people with cardiac arrhythmias should ensure an adequate supply of electrolytes, especially sodium, potassium, and magnesium.

6.8. Pregnancy

It is known that maternal nutrition during pregnancy affects the development and health of the child after birth and in adulthood. It is also associated with epigenetic changes in the foetus [225–227]. The ketogenic diet, for example in patients with polycystic ovary syndrome (PCOS), increases the chance of pregnancy, as a 2024 study shows [29]. It has also been established that during pregnancy, ketone bodies easily cross the placenta, reaching the same levels in maternal and foetal circulation [228,229]. The number of studies investigating KD in pregnant women is still too low to conclusively determine the diet’s effect on the course of pregnancy. Another difficulty is that studies on animal models have been inconclusive [225], and irrespective of their findings, results from animal studies cannot be directly extrapolated to the human population. However, a few papers studying pregnant women using the KD have been published. A 2021 study described the case of a 19-year-old woman who had become pregnant while treated with the ketogenic diet for glucose transporter type 1 deficiency syndrome (Glut1DS). Importantly, her child also suffered from Glut1DS and so was placed on the ketogenic diet at birth. The authors highlighted that, at the time of publication, the child was 5 years old and thriving, which in this case suggests no negative effect of the ketogenic diet on prenatal or postnatal development [230]. Other authors published a series of 2 case studies of women with epilepsy who followed the ketogenic diet during pregnancy. The first of these was the case of a 27 year-old woman whose pregnancy progressed without any complications: a healthy baby boy was born with normal Apgar scores, and his neurological development was normal. The woman continued the ketogenic diet after birth, and during breastfeeding. The other was the case of a 36 year-old woman who had been treated with epilepsy medication and who used the KD during pregnancy. She gave birth to a healthy boy with bilateral auricular deformities, whose neonatal hearing screening results were normal. (It has been suggested that the auricular deformities may not be linked to the KD.) The authors concluded that non-pharmacological therapies for epilepsy such as the KD and the modified Atkins diet (MAD) may be effective during pregnancy, but that their safety in pregnancy still needs to be determined [231]. It should however also be noted that the risk of congenital heart defects is known to increase by 8 per cent for every 10 mg/dl of glucose above normal in early pregnancy in women without gestational diabetes [232]. Importantly, the KD effectively reduces fasting glucose and glycated haemoglobin [19], and is not associated with postprandial glucose spikes [71].

It is interesting to contemplate the example of the Inuit, whose diet was historically composed primarily of fatty animal foods, and was therefore naturally extremely low in carbohydrate. Inuit women continued their usual diet even during pregnancy and, in fact, the consumption of seal or caribou meat has long been seen as important for pregnant women’s wellbeing [233]. Although there is still not enough data to conclusively determine the effect of KD on pregnancy [225], available information suggests that a well-formulated KD may be safe in pregnant women. However, switching to the KD during pregnancy can be problematic due to the well-known physiological stress of keto-adaptation [218]. Therefore, unless medically indicated, pregnancy may not be a good time to implement this particular dietary approach. Switching to the KD well in advance of pregnancy would be much safer, as this timing would not expose mother or the developing baby to the stress of the keto-adaptation period. Given these important considerations and the limited clinical data available in this area, the KD should only be used in pregnancy with extreme care, adequate precautions, and clinical supervision.

6.9. Breastfeeding

The importance of breastfeeding is indisputable as it offers a number of benefits not only for the child (including a lower risk of metabolic diseases such as diabetes, overweight or obesity, and higher scores on IQ tests in the future), but also for the mother (including a lower risk of metabolic diseases, cardiovascular diseases and even many cancers) [234]. It is estimated that increasing breastfeeding rates to near-universal levels could prevent up to 823,000 deaths annually among children under the age of five [235].

The question of whether the ketogenic diet is safe during breastfeeding will require more research to draw firm conclusions. Marshall et al. recommend avoiding low-carbohydrate diets altogether, with particular emphasis on the ketogenic diet [236]. However, interesting insights can be found in a 2024 systematic review which collected data on carbohydrate restriction (mainly different variants of the ketogenic diet and other low-carbohydrate diets) during lactation. In the studies evaluated, although nursing mothers experienced negative symptoms such as vomiting, nausea, abdominal pain, muscle weakness, fatigue and malaise, no risks to their babies were reported [237]. Furthermore, these outcomes should not be extrapolated to the general population, because the analysed individuals (with two exceptions) were using the KD without professional supervision, and mainly for weight loss, therefore calorie deficits may have contributed to the negative experiences. It should be emphasised that restrictive low-calorie diets are absolutely contraindicated during pregnancy and breastfeeding [238]. A survey of women who had followed a KD while breastfeeding found that most of them appreciated this dietary strategy. However, attention was drawn to the risk of lactation ketoacidosis, which was reported in 1 out of 21 women. Still, this condition is probably not caused by KD alone, but also due to other variables, such as a child’s illness leading to more frequent breastfeeding combined with extremely low food intake by the mother. Some women also noted reduced milk supply associated with the diet which seemed to resolve relatively quickly when additional calories and/or carbohydrates were added [239]. Returning to the example of the Inuit, it is known that Inuit women successfully breastfeed their children for centuries despite very limited access to carbohydrates. This historical example strongly suggests that the human body is able to adapt to a low-carbohydrate dietary pattern, even during lactation [233]. However, it should be noted that the diet and lifestyle of indigenous Inuit women cannot be directly extrapolated to most of the world’s female population.

No firm conclusions can be made on the basis of this limited information. However, it is reasonable to state that a well-formulated ketogenic diet may be safe during breastfeeding, as long as extreme caution is taken and the clinical condition of mother and child is regularly monitored.

6.10. Underweight

Most individuals (up to 77%) use the ketogenic diet for weight loss, which is also how it is most commonly perceived [240–243]. This is very likely because the diet has been shown to be superior to high-carbohydrate weight loss diets. KD outperforms other diets in many respects, including appetite and hunger suppression, greater initial weight loss, superior anti-inflammatory effects, better regulation of glycaemia and insulinaemia, and psychological benefits [21]. Some of these effects, although desirable in the context of weight loss, can be highly problematic for underweight individuals who need to gain weight. For example, it is known that individuals wishing to gain weight should eat more, but the KD effectively reduces appetite and hunger [244–246], and the initial rapid weight loss (resulting from the loss of glycogen and water) [21] that occurs with the KD is also undesirable in underweight individuals.

Paradoxically, preliminary findings suggest that the KD may hold benefits for people with anorexia nervosa (AN). For example, a retrospective case series of 3 adults with severe, underweight AN that had been refractory to standard therapy documented that after switching to a ‘carnivore’ KD (comprised entirely of animal protein and fat), complete, sustained remission of AN was achieved, with all 3 individuals gaining more than 20 kg of body weight. This dietary intervention was also associated with reduced anxiety symptoms and improved overall psychological wellbeing [247]. There are also 6 prospective cases of women with weight-restored anorexia nervosa whose residual eating-disordered thinking and other psychological symptoms of AN partially improved in response to a standard KD followed by ketamine therapy [248].

A 2024 paper discussed the potential mechanisms underlying the reversal of AN symptoms in response to KD therapy. Among other things, the authors point to the diet’s effects on the central nervous system, which may attenuate the pathological desire to lose weight and mitigate other symptoms characteristic of eating disorders [249]. Norwitz et al. proposed classifying AN as a ‘metabolic-psychiatric’ condition, indicating that it may respond to treatment with metabolic interventions that exhibit neuromodulatory properties [247].

With the above in mind, a distinction must be made between cases of slight underweight in people simply wishing to gain a few pounds, and AN-a life-threatening metabolic-psychiatric disease. Given the diet’s neuromodulatory properties and other metabolic benefits, the KD may prove an effective tool in the management of AN, much like in other conditions studied in the growing field of metabolic psychiatry [9,250–252]. Conversely, for reasons mentioned in the first paragraph, in cases of simple underweight unrelated to AN, the KD may not be the most appropriate or effective weight normalisation strategy.

6.11. Intensive physical activity

The ketogenic diet appears to be well suited to lower-intensity physical activity requiring a balanced or moderate energy demand, but there is no consensus regarding its effect on more intense exercise [253]. One paper reports that for endurance athletes, the literature supports the use of the KD as an effective strategy for reducing body weight and fat mass, suggesting potential improvements in performance at submaximal intensities (∼60%) [254]. The authors of another paper suggest that KD is unlikely to be optimal for improving performance in high-intensity endurance events or activities requiring rapid bursts of carbohydrate-derived energy [255]. However, it is worth noting that there is some disagreement regarding the effect of KD on athletic performance. One of the key concerns is the adaptation to this dietary approach [34,256]. It is estimated that a low-carbohydrate diet must be adhered to for a minimum of four weeks in order to achieve the required energy homeostasis and improve exercise performance. This is yet another observation suggesting the importance of the aforementioned adaptation period [257]. While this particular aspect may be decisive, some studies have looked at athletic performance in non-adapted athletes. A 2025 paper examining the impact of low-carbohydrate diets on athlete performance shed a new light on the officially accepted narrative. According to its authors, their study challenges the key role of carbohydrates in athletic performance and the belief that muscle glycogen (or an adequate supply of exogenous carbohydrates) is key and/or indispensable in maintaining long-term exercise performance. They pointed out that a key factor affecting exercise performance is hypoglycaemia, eliminated by consuming just 10 g of carbohydrate per hour (which improved performance by 22% despite low glycogen availability).

At the same time, water and electrolyte losses are known to occur during exercise (especially if intense), both of which are essential for muscle and nervous system function, among other things [213,258,259]. The ketogenic diet increases both electrolyte and water excretion during the adaptation period [218], which can be detrimental to athletes. This should be kept in mind by everyone switching to the KD in a period of physical activity (especially intense physical exercise).

C

6.12. Periods of intense stress

For several reasons, a period of intense stress is not a good time to switch to the ketogenic diet. Firstly, the period of adaptation to KD is itself stressful for the body, if only because of a number of changes that are triggered by switching to a new energy source (from glucose to ketones), including changes in water-electrolyte balance [218]. Indeed, it is known that in the short-term (<3 weeks), the KD can increase levels of cortisol [260], the stress hormone, and increased cortisol levels are accompanied by psychological stress. As a result, the burden on the body may become excessive [261]. Stress is also often associated with impaired urinary function [262] which plays a particularly important role during keto- adaptation [218]. Cortisol also affects the regulation of ketogenesis, the metabolic pathway through which ketone bodies are produced [39]. At the same time, cortisol also stimulates gluconeogenesis – the conversion of non-sugar precursors into glucose [263] – which in turn may interfere with the normal adaptation to the state of ketosis (depending on the type, intensity and duration of stress), which may once again be linked to the inhibition of ketosis by insulin. The literature even suggests that type 2 diabetes may be triggered by psychological and physical stress. Indeed, it is known that stress can increase glucose levels (hyperglycaemia) and insulin requirements, at the same time promoting insulin resistance [264]. The process of ketogenesis and adaptation to the state of ketosis may therefore be inhibited by elevated levels of glucose. Given anti-ketogenic properties of cortisol, periods of intense stress are not the best times to start the ketogenic diet. However, this question needs to be addressed in a more nuanced way; since obesity itself can be a significant stressor [265], weight reduction that accompanies the KD may reduce stress symptoms [266]. It is important to note, however, that these benefits are primarily observed in chronic stress rather than acute, intense stress.

6.13. Postoperative recovery period

Post-operative recovery is a period when the body undergoes a process of tissue regeneration and metabolic adaptation, and returns to physiological homeostasis to restore functional capacity and overall health. During this period, the patient should rest, avoid any activity that could interfere with recovery, and maintain proper hydration and nutrition [267]. Surgery itself can be psychologically stressful, but can also disturb metabolic and physiological processes, impinging on the metabolic pathways of nutrient absorption and digestion that ultimately lead to energy production [268]. Switching to the ketogenic diet at this time is not optimal, as the process of keto-adaptation is in itself a stressor for the body. Combined with the psychological and metabolic stress after the surgery, it carries the risk of excessive burden on the body, as described in section 4.12.

This is also true for patients after bariatric surgery and other surgeries that involve the gastrointestinal tract (e.g. oesophageal, pancreatic, small and large bowel surgeries). In these cases, the issues described above (recovery and postoperative stress) are accompanied (in the long term) by the impact of changes in the GI tract structure. While it is known that a very low-calorie ketogenic diet can be safe and effective for weight restoration in patients after bariatric surgery [269,270], the feasibility of the KD in such patients will depend on the type of surgery and their individual tolerance. Therefore, special caution should be exercised in all such cases.

Situations in which particular caution is required when using the ketogenic diet are illustrated graphically in Figure 1

Figure 1.

Situations in which special caution should be exercised when following a ketogenic diet Created in BioRender https://BioRender.com/v8tm6qb

7. KD and drug interactions

The ketogenic diet can interact with certain drugs both indirectly – through modulation of common metabolic or physiological pathways – and directly, at the molecular level, affecting drugs’ pharmacokinetics and pharmacodynamics. Some interactions are beneficial, such as enhancing the effect of medications and thus supporting eventual dose reductions. However, adverse interactions are also possible. These may weaken drug efficacy or exacerbate side effects, requiring careful monitoring and individualised therapy. Selected interactions are presented below, but the authors wish to emphasise that, given the multitude of drugs and their mechanisms of action, those described below are certainly not the only possible interactions between KD and medication to be aware of.

7.1. Sodium-glucose cotransporter-2 (SGLT-2) inhibitors

Sodium-glucose cotransporter-2 (SGLT-2) inhibitors are a class of drugs commonly used to improve blood glucose control in adults with type 2 diabetes. SGLT2s are proteins involved in the absorption of filtered glucose from the tubular lumen of the kidneys. Thus, inhibitors of these proteins will decrease the absorption of filtered glucose, lower the renal threshold for glucose, and increase urinary glucose excretion [271]. It appears that similar effects can be achieved by restricting carbohydrates in the diet, with a low risk of side effects, as described by the authors of one paper [272]. In contrast, the KD combined with SGLT-2 inhibitors may offer some benefits as well as create some risks.

Firstly, both the KD and SGLT-2 inhibitors promote lower blood glucose levels [19,271], which is beneficial, provided treatment is monitored to avoid hypoglycaemia.

Secondly, both SGLT-2 inhibitors and the initial phase of the KD (keto-adaptation) may disrupt water and electrolyte balance by increasing the loss of water and electrolytes from the body [273]. Importantly, some studies do not clearly confirm the impact of SGLT-2 inhibitors on electrolyte balance, suggesting that the observed effects may be selective and element-specific. For example, one study demonstrated that SGLT-2 inhibitors increased serum magnesium and phosphate levels, with no significant effect on other electrolytes [274]. One report even suggested the risk of hyperkalaemia [271].

Thirdly, some reports point to a possibly greater risk of euglycemic diabetic ketoacidosis (euDKA) when combining KD with SGLT-2 inhibitors [275]. One patient following a carbohydrate-restricted diet developed euDKA after just the first dose of empagliflozin, while another patient who was taking an SGLT2 inhibitor developed euDKA after only 1 week on the ketogenic diet [276].

7.2. Metformin

Metformin is a drug used in patients with type 2 diabetes, and in some patients with pre-diabetes. The drug lowers blood glucose concentrations by reducing glucose production in the liver and its absorption in the intestine. In addition, it improves insulin sensitivity. Recently, an increasing number of studies have been published exploringits potential anti-cancer, anti-ageing and neuroprotective properties [277].

The ketogenic diet may act synergistically with metformin, as it also effectively lowers blood glucose concentrations and improves insulin sensitivity [19,277]. The first preclinical study demonstrated a possible beneficial synergy of KD and metformin with respect to anti-cancer effects, specifically in the treatment of triple-negative breast cancer (TNBC). Indeed, it is known that cancer cells require abnormally high concentrations of glucose, so it would be logical to reduce glucose intake and blood glucose levels. The study found that by lowering blood glucose concentration, the integrated actions of KD and metformin significantly inhibited tumour proliferation and improved overall survival [278]. Importantly, the combination of KD and metformin was shown to have a better effect in this context than either of these strategies alone.

7.3. Glucagon-like peptide-1 (GLP-1) agonists

GLP-1 (glucagon-like peptide-1) is an endogenous incretin hormone synthesised in the L cells of the small intestine in response to food intake. It has a multifaceted physiological role: on the one hand, it modulates central appetite control mechanisms. Its anorexigenic effect is achieved by binding to GLP-1 receptors in the brain and slowing gastric emptying, which promotes feelings of satiety. On the other hand, it influences glucose homeostasis, lowering glucose concentrations in both fasting and postprandial periods, mainly by stimulating glucose-dependent insulin secretion and inhibiting glucagon secretion. GLP-1 agonists mimic the action of GLP-1 by binding to the GLP-1 receptor, thereby activating it much like the natural hormone [279–281].

The efficacy of these drugs in clinical trials often exceeds their efficacy in the real world. The discrepancy is due to a number of reasons, including non-adherence (up to 70.1% of patients discontinue therapy after 24 months) and an unrepresentative selection of patient groups [282]. In contrast, a 2025 meta-analysis found that, despite their efficacy in reducing body weight in those taking certain GLP-1 agonists, much of the weight loss was attributable to a reduction in lean body mass, which may be a disadvantage [283].

One randomised controlled trial [284] demonstrated that GLP-1 levels were higher in people on the ketogenic diet than in control groups. At the same time, the KD is known to be effective in suppressing hunger, enhancing satiety, reducing blood glucose levels, and improving glycaemia management [21]. It is therefore not surprising that both the ketogenic diet and GLP-1 agonists show similar weight loss effects [285]. Both are also effective in reducing blood glucose levels, which implies that their effects can be synergistic. Therefore, in individuals simultaneously using the KD and GLP-1 analogues, the drug dosage can probably be reduced without compromising the therapeutic efficacy. In fact, it is known that GLP-1 analogues may be associated with hypoglycaemia [286], which should be kept in mind when combining with the KD. Low-carbohydrate diets have been shown to help patients maintain glycaemic and weight control even after discontinuing these drugs [287,288].

7.4. Insulin and sulphonylurea derivatives

Both insulin and sulphonylurea derivatives are diabetes drugs used primarily to lower serum glucose concentrations. At the same time, it is known that KD lowers blood glucose concentrations so effectively that it can lead to hypoglycaemia when combined with these drugs. Therefore, their dosage may need to be adjusted (reduced), or discontinued altogether in KD users [177,179–183]. This issue is also addressed in section 4.1, as the use of hypoglycaemic drugs in combination with KD is one of the situations that requires special caution. This is because under conditions of reduced carbohydrate intake, the efficacy of these drugs may potentially be too strong, which may increase the risk of hypoglycaemia and hence requires careful monitoring and possible dose modification.

7.5. Antiepileptic drugs

Antiepileptic drugs are commonly used in treating epilepsy, and some are also often used in psychiatry for the treatment of mental illnesses such as bipolar disorder. These drugs can be divided into those that induce hepatic enzymes (e.g. phenytoin, phenobarbital and carbamazepine) and those that do not (e.g. topiramate, zonisamide, sodium valproate, lamotrigine, pregabalin or levetiracetam) [197]. These drugs do not always produce the clinically expected results, for example, in cases of drug-resistant epilepsy, where a ketogenic diet may be more effective. For this reason, the KD and antiepileptic drugs are often used concomitantly, raising the legitimate question of potential drug-KD interactions. Possible interactions of KD with selected antiepileptic drugs are listed below. Please note that other potential interactions are certainly possible.

Topiramate is mainly used as monotherapy and adjunctive therapy in the treatment of epilepsy, as well as in psychiatric disorders and in the treatment of migraine disorders [289]. Zonisamide is mainly used as adjunctive therapy in epilepsy in patients over 16 years of age [290]. Both drugs inhibit carbonic anhydrase, which can lead to metabolic acidosis due to reduced serum bicarbonate levels. Indications are that the risk of metabolic acidosis may be higher in the early phase of adaptation to the ketogenic diet. This was demonstrated in one study in which the KD was introduced in patients who were already on topiramate therapy. Serum bicarbonate levels decreased, albeit mainly in the adaptation period. The authors concluded that the combination therapy of topiramate with the ketogenic diet requires careful monitoring of bicarbonate levels. If any negative symptoms are observed, bicarbonates should be supplemented [291]. This issue was also addressed in a more recent paper pointing out not only the risk of metabolic acidosis developing as a result of the interaction of KD with topiramate or zonisamide, but also suggesting that the combination of KD with zonisamide (but not topiramate) may increase the risk of nephrolithiasis [292]. However, the authors emphasise that there is no need to avoid the concomitant use of KD with these drugs. It is possible that, due to their overlapping mechanisms of action, the combination of the ketogenic diet and zonisamide may have a potential additive effect, potentially enhancing therapeutic efficacy without the need to intensify pharmacological treatment.

In addition, KD may increase the activity of cytochrome P450 enzymes and thus potentially interact with anti-epileptic drugs metabolised by these enzymes. One of the results will be a reduction in the blood concentrations of these drugs. The literature reports reductions in the concentrations of lamotrigine, topiramate and lacosamide, as well as significant reductions in the concentrations of carbamazepine, valproic acid, and clobazam (which was also shown to be reduced by 42% in another study [293], but no change in the concentrations of oxcarbazepine, zonisamide, and levetiracetam after 4–12 weeks of MAD [294].

Valproic acid is itself a fatty acid that can be burned by cells as an energy source, and this process may be enhanced during KD, when fat metabolism accelerates. This may be one reason why valproate blood levels tend to decrease on a KD. Monitoring of serum valproate levels is recommended as dosage adjustments may be clinically indicated when valproate and the KD are used together [295,296].

Lithium is a naturally occurring mineral with mood-stabilizing properties that is often used in the management of bipolar disorder. As discussed earlier, the ketogenic diet changes the balance of fluid and electrolytes in the body, and this is largely due to increased renal excretion of sodium and water. Lithium is filtered and excreted by the kidneys, which handle lithium in a similar way to sodium [297,298]. The combination of lithium and the KD may therefore result in changes to blood lithium levels, particularly during the initial keto-adaptation period. Since lithium has a narrow therapeutic window,monitoring of lithium levels before and during KD therapy is recommended.

7.6. Diuretics

Diuretics are drugs that increase diuresis (urine excretion). They are used, for example, in treating oedema, heart failure, hypertension, and ascites. They increase urine production and volume and thus intensify excretion of water and electrolytes (with some exceptions) from the body [299]. Given that the period of adaptation to the ketogenic diet produces similar effects (increased diuresis and loss of electrolytes) [218], initiating the KD in patients taking diuretic drugs may potentiate these effects, thereby potentially disrupting water and electrolyte balance. The pharmacodynamics of thiazide diuretics may also make it more difficult to achieve the state of ketosis, by increasing blood glucose levels [197].

7.7. Lipophilic drugs

Lipophilic drugs are fat-soluble compounds with low water solubility that can penetrate biological barriers (e.g. the blood-brain barrier) [300]. This class of drugs includes a broad spectrum of different types of drugs [197]: olanzapine and clozapine (classified as antipsychotics) [301]; amitriptyline, nortriptyline and doxepin (classified as antidepressants) [302–304]; diazepam and midazolam (categorised as benzodiazepines) [305]; zolpidem and zopiclone (categorised as sedatives) [306,307]; phenytoin, carbamazepine, valproic acid, gabapentin and pregabalin (categorised as antiepileptics) [308,309]; simvastatin, fluvastatin, lovastatin, pitavastatin, and atorvastatin (classified as hypolipemic drugs) [310]; propranolol and metoprolol (classified as beta-blockers) [311]; amiodarone (classified as an antiarrhythmic) [312]; and a number of others, such as chloroquine and mefloquine (antimalarial drugs); ketoconazole and itraconazole (antifungal drugs); spironolactone (a diuretic); rifampicin (an antituberculosis drug); ivermectin (an antiparasitic drug); cetirizine and loratadine (antihistamines); methadone (an opioid); ritonavir and saquinavir (antivirals); and tacrolimus (an immunosuppressant) [197].

Highly lipophilic drugs may show increased bioavailability when administered in combination with high-fat diets [313]. Therefore, the KD may increase their absorption, which could increase their serum concentrations; this could increase their therapeutic efficacy but also increases the risk of adverse effects. In such cases, the physician must use their clinical judgment to make adjustments in drug dosages as needed. with the aim of achieving maximal efficacy with minimal adverse effects [197].

7.8. Corticosteroids

Corticosteroids are synthetic analogues of the natural steroid hormones produced by the adrenal cortex and have applications in a wide range of medical situations. This class of drugs is divided into glucocorticoids (which have immunosuppressive, anti-inflammatory and vasoconstrictive effects) and mineralocorticoids (which control the body’s water and electrolyte balance) [314].

These drugs may interact with the KD, as common side effects of corticosteroids are increased blood glucose concentrations (due to increased hepatic gluconeogenesis), decreased insulin sensitivity, and decreased glucose utilisation by adipose and muscle tissue [315]. In this case as well, elevated insulin levels play a role, as insulin is anti-ketogenic. High blood glucose is one of the main factors hindering the optimal state of nutritional ketosis. Namely, it is known that the higher the glucose to ketone ratio, the higher the glucose ketone index (GKI), which in turn adversely affects the depth of ketosis [316]. Moreover, the KD can increase cortisol levels [Whittaker et al. 2022] and activate the hypothalamic-pituitary-adrenal (HPA) axis [317], particularly in the short term, while systemic corticosteroids can lead to hypercortisolemia and inhibition of the HPA axis, another potential interaction of these drugs with KD [197,318]. Additionally, some findings point out to a possibly higher risk of water-electrolyte imbalance [197].

8. Possible side effects of KD

Potential side effects of the ketogenic diet are most often related to the process of the body adapting to the state of ketosis in the initial period of the diet. According to the literature, these side effects are typically observed in the first 2–3 days of starting the KD, most or all of them resolve within 2 to 4 weeks, and they are rarely severe enough to require discontinuation of KD [218]. In a retrospective study of 226 people who had followed a KD, reported side effects included dizziness (mild, 39.8%; moderate, 27.4%; severe, 11.5%), polyuria (72.1% overall), lethargy (69.7%), and nausea (mild, 29.2%; moderate, 16.4%; severe, 5.8%). The authors remarked that 90.3% of these subjects were satisfied with the KD, and 81.9% would recommend this dietary approach to anyone wishing to reduce body weight. Thus, despite some transient side effects, the overall perception of KD was very positive [319]. Another study evaluated personal experiences with KD shared by 300 online forum users. The most common side effects (in order of descending frequency) were: flu-like symptoms (44.6%), headache (24.8%), fatigue (17.8%), nausea (15.8%), dizziness (14.9%), brain fog (10.9%), gastrointestinal discomfort (10.9%), decreased energy (9.9%), feeling faint (7.9%), heartbeat alterations (5.9%), sore throat (5.9%), decreased appetite (5.0%), shaking (5.0%), body aches (4.0%), cravings (4.0%), cramping (3.0%), hunger (3.0%), increased thirst (3.0%), insomnia (3.0%), irritability (3.0%), muscular soreness (3.0%), muscular weakness (3.0%), panic (3.0%), sluggishness (3.0%), anxiety (2.0%), bloating (2.0%), dehydration (2.0%), diarrhoea (2.0%), dry mouth (2.0%), fever (2.0%), joint pain (2.0%), lethargy (2.0%), listlessness (2.0%), weak fingernails (2.0%), acne (1.0%), asthma (1.0%), chills (1.0%), constipation (1.0%), coughing (1.0%), decreased motivation (1.0%) and a few others, such as migraine, ears blocked, muscle tightness, vomiting, and phlegm in throat. Side effect reports peaked in the first week and decreased after four weeks [320]. However, this was a study based on an analysis of subjective opinions published on online forums, which limits its scientific value. More authoritative data was presented by Cervenka et al. who analysed a survey among an international panel of experts with clinical experience in the use of the ketogenic diet in adults, which allowed for a less biased and more standardised assessment of the diet’s effects. However, the results of this study cannot be extrapolated to the general population, as it was mainly concerned with KD side effects in patients with neurological disorders. For example, weight loss in that context was listed as a side effect, while it is the main goal (and therefore viewed as a benefit) of the KD in the general population. The study by Cervenka et al. estimated that the most common side effects of KD included (in descending order) constipation, weight loss, hyperlipidaemia, leg cramps, irregular menstruation (in women of childbearing age), nausea, and vomiting [321]. A systematic review published in 2025 analysed the available scientific literature on the prevalence of symptoms, as well as possible mechanisms and proposed interventions to alleviate these symptoms of the keto adaptation period. Children and adult populations were analysed separately. The most common symptoms in adults and children were constipation (1–68% in adults and 15–63% in children), nausea (8–16% and 27–42%), vomiting (1% and 5–36%), diarrhoea (2–23% and 3.6–20%), feelings of decreased energy (18–25% and 17–28%) and mood changes (1% and 6.7–10%), respectively. Symptoms identified in adults only were halitosis (38%), liver function abnormalities (24%), light-headedness/dizziness (15–21%), headache (8–25%), brain fog/reduced cognitive performance (10%) and skin rash (13–21%). Conversely, the side effects observed in children only included hypoglycaemia (6–28%), ketoacidosis (2–4%) and kidney stones (2.5–4%). It is known, however, that adequate preparation for the KD helps avoid a large proportion of symptoms, and that most of these symptoms resolve within 2–4 weeks anyway [218]. It is also worth noting from clinical practice that it is extremely rare for a patient on a ketogenic diet to experience dramatic weight loss accompanied by a rising HbA1c level – an atypical scenario, as HbA1c would typically be expected to decrease. Such a pattern may suggest underlying insulin deficiency or, in some cases, pancreatic cancer.

Figure 2 The potential adverse effects of the ketogenic diet are presented in detail in Table 2, while the most common of them are illustrated in a more general, visual manner in Figure 2.

Figure 2.

Possible side effects of the ketogenic diet. Created in BioRender https://BioRender.com/4rkpsp0

Table 2.

Possible side effect of ketogenic diet.

Possible side effect of ketogenic diet
Populations	Comments	Frequency of all reported side effects	Sources
Adults and children	A scoping review based on 89 studies	Both adults and children: constipation (1–68% in adults / 15–63% in children), nausea (8–16% / 27–42%), vomiting (1% / 5–36%), diarrhoea (2–23% / 3.6–20%), decreased energy/fatigue (18–25% / 17–28%) and mood changes (1% / 6.7–10%). Adults only: halitosis (38%), abnormal liver function tests (24%), dizziness/light-headedness (15–21%), headache (8–25%), brain fog/reduced cognitive performance (10%) and skin rash (13–21%). Children only: hypoglycaemia (6–28%), ketoacidosis (2–4%) and kidney stones (2.5–4%).	[218]
Adults	Retrospective observational study involving 226 individuals (n = 226)	Dizziness (mild, 39.8%; moderate, 27.4%; severe, 11.5%), polyuria (72.1% overall), lethargy (69.7%), and nausea (mild, 29.2%; moderate, 16.4%; severe, 5.8%)	[319]
Adults	Qualitative content analysis of online forums (43 forums, 300 unique users (n = 300))	Flu-like symptoms (44.6%), headache (24.8%), fatigue (17.8%), nausea (15.8%), dizziness (14.9%), brain fog (10.9%), gastrointestinal discomfort (10.9%), decreased energy (9.9%), feeling faint (7.9%), heartbeat alterations (5.9%), sore throat (5.9%), decreased appetite (5.0%), shaking (5.0%), body aches (4.0%), cravings (4.0%), cramping (3.0%), hunger (3.0%), increased thirst (3.0%), insomnia (3.0%), irritability (3.0%), muscular soreness (3.0%), muscular weakness (3.0%), panic (3.0%), sluggishness (3.0%), anxiety (2.0%), bloating (2.0%), dehydration (2.0%), diarrhoea (2.0%), dry mouth (2.0%), fever (2.0%), joint pain (2.0%), lethargy (2.0%), listlessness (2.0%), weak fingernails (2.0%), and other symptoms reported by single users (1.0% each): acne, asthma, depressed mood, derealization, ears blocked, mental confusion, migraine, muscle tightness, phlegm in throat, photophobia, restlessness, run down, runny nose, sinus pressure, tingling in head, vomiting.	[320]
Adults with epilepsy or other neurologic disorders	International multicenter expert survey (20 centers treating >2000 adults)	Side effects (in descending order of frequency): constipation, weight loss, hyperlipidaemia, leg cramps, irregular menstruation (women of childbearing age), nausea, vomiting (quantitative data limited; only hyperlipidaemia reported in ∼1/3 of adults)	[321]

9. Potential problems with the KD in clinical practice

Obviously, to achieve the therapeutic effects of KD, patients must adhere to its principles, which can be challenging. Possible practical problems initiating and maintaining the ketogenic diet may include psychological, emotional and sociological factors. For example, the prospect of imagining a dinner plate with a minimum amount of carbohydrates can seem overwhelming to some; others struggle with quitting ultra-processed foods (especially if their relationship with those foods is addictive. Also, the fear of failure or standing out can lead to anxiety. In turn, some patients with bipolar disorder may experience temporary exacerbation of mood symptoms that precede sustained improvement [322]. In a social context, adherence to the KD often involves the need to forgo traditional meals at family gatherings, which can be a significant challenge, especially in cultures where such meals play an important social role. In addition, people who prefer to eat out may have problems finding restaurants that offer suitable meals on their menus. For adolescents and adults who seek to conform to the dietary norms prevailing in their peer groups, a sense of exclusion may arise. People on the KD may find that avoiding carbohydrates leads to tense and stressful interactions with family members and friends, which can manifest as intrusive questions, disapproval, and even warnings from those who are less knowledgeable about the KD. The root cause of such reactions is that the principles of the KD conflict with the common belief that balanced diets are healthiest [322–324].

10. Strengths and limitations

This narrative review provides a comprehensive overview of the available literature, including meta-analyses, systematic reviews, randomized controlled trials, clinical studies, case reports, mechanistic evidence, reference databases (e.g. Orphanet), and clinical guidelines. The search strategy allowed identification of evidence on specific contraindications and safety concerns that may not be captured by broader keyword searches. To our knowledge, no other publication has addressed this topic so broadly from multiple perspectives, making this review a uniquely comprehensive contribution to the field.

As this is a narrative review, the selection of evidence does not follow a predefined systematic protocol; therefore, the review may be susceptible to selection bias. Additionally, the heterogeneity of included studies and reliance on selected databases, guidelines, and keywords may limit the generalizability of the findings, and evidence on rare conditions or specific drug interactions remains scarce.

11. Conclusions

The ketogenic diet is a promising tool for both prevention and treatment of not only certain neurological disorders, but also a number of metabolic disorders and diseases such as obesity, overweight, type 2 diabetes, and MAFLD. It has also gained recognition in the growing field of metabolic psychiatry. However, in certain disorders, diseases, or clinical situations, the ketogenic diet is contraindicated, whether absolutely or relatively. Furthermore, in some settings, the application of this dietary approach must be accompanied by special precautions.

Most absolute contraindications to the KD are specific rare genetic predispositions, mainly involving difficulties in converting fats into energy. These absolute contraindications include pyruvate carboxylase (PC) deficiency; disorders of fatty acid beta-oxidation (i.e. primary carnitine deficiency or PCD), carnitine palmitoyltransferase deficiency, carnitine-acylcarnitine translocase (CACT) deficiency, 3-hydroxyacyl-coenzyme A dehydrogenase (HADH) deficiency, medium-chain 3-hydroxyacyl-CoA dehydrogenase deficiency (MHADD), long-chain 3-hydroxyacyl-CoA dehydrogenase deficiency, medium-chain acyl-CoA dehydrogenase deficiency (MCADD), very-long-chain acyl-CoA dehydrogenase deficiency (VLCADD); and porphyria.

Relative contraindications are based on some literature findings and/or indirect mechanisms that make KD potentially risky. Relative contraindications include a number of situations, such as acute pancreatitis (AP), acute liver failure (ALF), advanced stages of chronic kidney disease (CKD), use of propofol, and familial hypercholesterolaemia (FH). What makes these different from absolute contradictions is that the KD can potentially be a valid and potentially benefical option in these relatively contraindicated situations, but relevant research and clinical experience is often lacking. As a precaution, it may be safe to consider these situations absolute contraindications until they have been thoroughly researched and any outstanding safety questions resolved.

Situations requiring special caution, while not necessarily absolute contraindications to the KD, call for a tailored and carefully planned approach. This includes, but is not limited to, an in-depth knowledge of potential interactions and risks, close monitoring of the patient’s health status, increased clinical supervision, adjustment – and in some cases discontinuation – of medication doses, an appropriately designed nutrition plan, and possibly increasing the proportion of selected nutrients in the diet and/or targeted supplementation. It is likely that in some of these cases the KD could fail, but in others (e.g. in patients with type 2 diabetes taking hypoglycaemic drugs) it may offer a number of benefits, provided that medication doses are monitored and adjusted (or discontinued) by a specialist. Situations in which special care and attention should be exercised include (but are not limited to) the following: T2DM treated with hypoglycaemic medication (not metformin or GLP-1 drug), T1DM, hypertension treated with antihypertensive medication, cholelithiasis, gallbladder removal, electrolyte deficiency, cardiac arrhythmia, pregnancy, breastfeeding, underweight, intense physical activity, periods of intense stress, and postoperative recovery.

The ketogenic diet may interact with certain medications, which can lead to a range of both benefits and risks. Interacting substances include SGLT-2, metformin, GLP-1 agonists, insulin, sulphonylurea derivatives, antiepileptics, diuretics, lipophilic drugs, and corticosteroids.

Adverse effects associated with the KD are generally transient; they most often occur during the initial period of the body’s adaptation to the new metabolic state, and typically resolve within a few weeks. The most commonly reported side effects include the so-called ‘keto flu’, fatigue, nausea, lethargy, headaches, dizziness, gastrointestinal complaints (such as constipation or diarrhoea), lack of energy, polyuria,and hypoglycaemia. Importantly, these can often be prevented if switching to KD is preceded by adequate preparation.

This review seeks to serve as a useful reference for clinical practice by presenting a detailed consideration of clinical and metabolic safety concerns related to the KD, its possible side effects, and potential interactions between the KD and commonly prescribed medications. At the same time, it should be noted that the contraindications we mention do not exhaust the full spectrum of possible clinical issues that may arise with the ketogenic diet. Please also note that the assignment of contraindications to absolute or relative categories sometimes varies between literature sources, often based on authors’ subjective clinical interpretation and experience; therefore, clinicians are advised to use their clinical judgment when considering the use of the KD in these clinical scenarios and tailor their treatment plan to the specific needs of their patients.