The ketogenic diet is unique and distinct from other low-carbohydrate diets such as the Paleo or Atkins diet in that subjects are encouraged to forget nearly all carbohydrates, avoid excess protein, and consume high levels of fat (which generally exceed 70% of total calories consumed), resulting in the production of ketones ^[1][88].

The classic ketogenic diet is a high-fat, very-low-carbohydrate diet, which restricts carbohydrate intake to 5–10% of total daily energy intake and replaces the remaining with high amounts of dietary fat (70–80%) and moderate amounts of protein (10–20%) ^[2][89]. Depending on the exact weight ratio of dietary fat to combined protein and carbohydrate, it can be further classified into 4:1 and 3:1, and should be always medically supervised. Beyond the classic ketogenic diet, there are also other variants of low-carbohydrate diets which allow more protein or carbohydrate ^[3][90], as summarized in Table 13. Ketogenic diets are different from the so-called low-carbohydrate diets. The latter refer to a carbohydrate intake below the recommended dietary allowance of 130 g/day ^[4][17], which is usually not sufficiently low to induce nutritional ketosis, whereas in ketogenic diets, carbohydrate intake is restricted to less than 50 g/day.

^][99,100].

Table 24 summarizes the major meta-analyses of RCTs in the field of ketogenic diets in humans. In a meta-analysis of 13 RCTs assessing the long-term effects of ketogenic diets (intervention duration longer than one year), it was found that ketogenic diets were associated with less than a kilogram (0.9 kg) of additional weight loss as compared to high-carbohydrate low-fat strategies ^[4][17]. Although this difference was found to be statistically significant, it is doubtful whether it is also clinically significant and translates into a meaningful clinical effect. In the same meta-analysis, no significant difference was found in glycemic control (HbA1c) between ketogenic and low-fat diets in patients with T2DM, and there was also no difference in weight loss at 2 years ^[4][17].

The major mechanism explaining the weight loss effects of ketogenic diets is associated with reduced insulin levels, which redirects lipid metabolism from storage to oxidation, promotes the use of ketone bodies as alternative fuels, and induces a metabolically beneficial state of nutritional ketosis, which mimics metabolic starvation in the human body ^[5][91]. Other possible mechanisms implicated in the weight loss effects of ketogenic diets include: (i) appetite suppression mediated by the higher satiety effect of proteins ^[6][7][92,93], the modulation of appetite-regulating hormones ^[8][94], and the direct anorexigenic effect of circulating ketone bodies ^[9][95]; (ii) reduced respiratory quotient ratio and greater metabolic efficiency in the direction of fat utilization ^[10][96]; (iii) increased metabolic costs of gluconeogenesis and protein-induced thermogenesis ^[11][12][97,98]; and (iv) increased energy expenditure as shown in short-term studies using state-of-the-art technology such as doubly labelled water ^[13][14

Whether ketogenic diets are better than low-fat diets in terms of body composition changes promoting preferential fat mass loss remains questionable. In a metabolic ward study in 17 overweight or obese men, weight loss accelerated but fat mass loss slowed down for 2 weeks after switching from the baseline to the ketogenic diet ^[13][99]. A subsequent metabolic ward study by the same research group tested the effects of an animal-based ketogenic diet compared to a plant-based, carbohydrate-rich, low-fat diet in 20 overweight young adults ^[14][100]. Participants were randomized to each diet, which they consumed for 2 weeks before crossing over to the other diet. Total energy intake was lower by nearly 700 kcal/day in the low-fat diet group. The reported hunger and dietary satisfaction scores were similar between groups. Both diets induced weight loss. However, most of the weight lost with the ketogenic diet came from LBM, whereas the low-fat diet resulted in significant losses of body fat after both the first and the second week ^[14][100]. The study concluded that low-fat, plant-based diets may control appetite and satiety even better than ketogenic diets contrary to what is commonly believed, and furthermore, that the rapid initial weight loss observed with ketogenic diets is predominantly due to loss of LBM comprising water, glycogen and muscle proteins ^[14][100]. These data fuel even more the skepticism regarding the superiority of ketogenic diets over the traditional low-fat hypocaloric dietary approaches.

Subgroup analyses have revealed more pronounced effects of ketogenic diets in overweight/obese and diabetic participants. A meta-analysis of 14 RCTs, comparing ketogenic diets with low-fat diets in subjects with and without T2DM, produced in summary the following findings: (i) ketogenic diets followed for 3–12 months are more effective than low-fat diets for improving glycemic control and insulin sensitivity in patients with T2DM, but their glycemic effects are comparable with those of low-fat diets in non-diabetic individuals; (ii) ketogenic diets followed for 1–12 months are more effective than low-fat diets for weight reduction in overweight and obese subjects, whether they have T2DM or not; (iii) ketogenic diets followed for 4 days up to 2 years have a more beneficial effect on lipid profile compared to low-fat diets only in diabetic patients, by reducing triglycerides and increasing HDL; but (iv) the lipidemic effect of ketogenic diets is adverse in non-diabetic subjects, by increasing total and LDL cholesterol ^[20][14]. The effects of carbohydrate-restricted diets on glycemic control in patients with T2DM appear to be most significant within the first 3–6 months and typically wane thereafter ^{[18][21][22][23]}[15,101,102,103]. These effects are mainly attributable to weight loss ^[22][23][101,102]. Of note, there is too little evidence to support that ketogenic diets improve glucose intolerance independently of the concurrent weight loss. This is not the case with other dietary approaches such as low-fat diets, in which glycemic control is improved despite the consumption of healthy carbohydrate-rich foods, such as legumes, fruits and whole-grain products, even in the absence of weight loss.

In adults with T1DM, both beneficial and adverse outcomes have been reported. In a small study of 11 adults with T1DM, a ketogenic diet improved glycemic control, but triggered more frequent hypoglycemic episodes and promoted dyslipidemia ^[24][104]. In pathophysiological terms, sustained ketosis should be strongly discouraged in patients with T1DM, since it has been associated with oxidative stress, inflammation and insulin resistance ^[25][105]. In patients with T2DM participating in clinical trials with ketogenic diets, a common finding is that glucose-lowering medications are frequently reduced or even eliminated ^{[26][27][28][29][30][31][32][33]}[106,107,108,109,110,111,112,113]. With regard to long-term glycemic effects, one open-label, non-randomized, controlled study of the ketogenic diet in patients with T2DM showed a reduction in HbA1c by 1.3% at one year and 0.9% at 2 years, but it should be noted that this group received intensive technological and behavioral support (digitally monitored continuous care intervention), which was not provided to the control group, and this is not always feasible in everyday clinical practice ^[32][34][112,114].

Safety issues and adherence: The potential adverse effects of ketogenic diets range from the relatively benign but unpleasant “keto flu”, which is an induction period of fatigue, weakness and gastrointestinal disturbances, to the less common but more dangerous occurrence of cardiac arrhythmias due to selenium deficiency ^[1][88]. Other adverse effects may include nephrolithiasis, constipation, muscle cramps, headaches, bone fractures, pancreatitis, and multiple vitamin and mineral deficiencies ^[1][88]. In the absence of multivitamin supplementation, individuals on ketogenic diets are at risk of frank nutritional deficiencies ^[35][115]. Even when consuming nutrient-dense foods, the classic ketogenic diet 4:1 can lead to multiple micronutrient shortfalls, often lacking in vitamin K, linolenic acid and water-soluble vitamins ^[36][116].

The most important risk, related to ketogenic diets, is the exclusion of fiber and unrefined carbohydrates from the diet, considering that fruits, legumes and whole grains have indispensable effects for human health. Extreme carbohydrate restrictions may profoundly affect diet quality, eliminating fiber sources and increasing the consumption of animal products being rich in cholesterol and saturated fat. In this context, evidence suggests that LDL cholesterol and apo-B-containing lipoprotein levels may fail to improve, or even significantly increase, with a ketogenic diet despite weight loss in healthy adults ^[37][117]. In a randomized, controlled, parallel-designed study in 30 young lean adults, a severely carbohydrate-restricted diet with less than 20 g carbohydrate per day for 3 weeks was associated with a significant increase in LDL cholesterol levels by 44% compared to the control habitual diet, and there was a greatly variable individual response ^[37][117]. Interestingly, carbohydrate-restricted diets can be linked to either increased or decreased mortality, depending on the quality of carbohydrates consumed and whether they rely more on animal proteins and saturated fat or plant-derived proteins and unsaturated fat, respectively ^[38][118]. These diets may also increase the risk of chronic diseases, since dietary components typically increased in ketogenic diets (red processed meat, saturated fat) are linked to an elevated risk of chronic kidney disease, cardiovascular disease, cancer, diabetes, and Alzheimer’s disease ^[39][13].

Urinary ketone levels are often used as an indicator of dietary adherence ^[40][35]. Ketogenic diets may have poor long-term tolerability and sustainability for many individuals ^[22][23][101,102].

Taking all evidence into consideration, it can be concluded that the risks of ketogenic diets may actually outweigh the potential benefits, since ketogenic diets confer only temporary improvements, they have unfavorable effects on diet quality, and there are currently inadequate data confirming their long-term safety and sustainability ^[39][13].