The Keto Diet: Rules, Benefits, Foods to Eat, and Honest Risks
Diet guide
Rules, foods & sample plan
The ketogenic diet is a very-high-fat, very-low-carbohydrate eating pattern that induces ketosis; it produces meaningful short-term weight loss and has proven clinical utility for epilepsy and type 2 diabetes management, but long-term adherence is low, saturated fat intake is often high, and most benefits disappear when the diet is not strictly maintained.
The ketogenic diet shifts your body's primary fuel source from glucose to fat by restricting carbohydrates to fewer than 20–50 grams of net carbs per day. Originally developed in the 1920s for pediatric epilepsy, it has gained widespread use for weight loss and type 2 diabetes management. While the short-term results are real, long-term adherence is the diet's Achilles heel, and the rise in LDL cholesterol seen in some individuals requires monitoring, not dismissal.
This is general nutrition and wellness information, not medical advice. If you're on a weight-loss medication or managing a health condition, confirm specifics with your clinician.
How the Ketogenic Diet diet works
The ketogenic diet was developed in the 1920s at Mayo Clinic for pediatric epilepsy and has since been applied broadly to weight loss. It restricts carbohydrates to typically fewer than 20–50 g of net carbs per day (net carbs = total carbs minus fiber), which depletes liver glycogen within 1–3 days and forces the liver to produce ketone bodies (beta-hydroxybutyrate, acetoacetate, acetone) from fatty acids — the state known as nutritional ketosis. In ketosis, the body relies primarily on fat (dietary and stored) rather than glucose for fuel. Macros typically run approximately 70–80% fat, 10–20% protein, 5–10% carbohydrate by calories. Protein is kept moderate (not high) in classical keto to avoid gluconeogenesis disrupting ketosis. Commercial variations include 'lazy keto' (only tracking carbs, not full macros), 'dirty keto' (any fats including processed foods), and 'clean keto' (emphasis on whole food fat sources).
The rules of the Ketogenic Diet diet
Net carbohydrates ≤20–50 g/day (the lower end induces ketosis more reliably; the upper end may work for some individuals)
Macros approximately 70–80% fat, 10–20% protein, 5–10% carbohydrate by calories
Protein is kept moderate (not unrestricted) to avoid gluconeogenesis disrupting ketosis — typically ~0.6–0.8 g per lb of lean body mass
Fat sources are the primary calories: oils (olive, coconut, avocado), butter, full-fat dairy, fatty meat, nuts, seeds, avocado
Carb sources strictly limited to non-starchy vegetables (leafy greens, broccoli, cauliflower, zucchini, peppers) with careful portion control even on allowed vegetables
No sugar, no grains (bread, rice, pasta, oats), no legumes, no most fruit, no starchy vegetables (potato, corn, peas, carrots in quantity)
Electrolyte management (sodium, potassium, magnesium) is important during adaptation — electrolyte loss accompanies the water weight lost during glycogen depletion
Foods to eat & avoid
Eat freely
- Fatty meats: beef, lamb, pork, bacon, sausage (preferably minimally processed)
- Poultry: chicken thighs, wings, skin on
- Fatty fish: salmon, mackerel, sardines, herring
- Eggs
- Full-fat dairy: butter, hard cheese, cream, full-fat Greek yogurt in small amounts
- Avocado
- Nuts and seeds: almonds, walnuts, pecans, macadamia, chia, flaxseed (carb counts vary, portions controlled)
- Low-carb vegetables: leafy greens, broccoli, cauliflower, zucchini, bell pepper, asparagus, mushrooms
- Oils: olive, avocado, coconut
- Berries in small portions: raspberries, blackberries (lower carb than other fruits)
Limit or skip
- All grains and grain products: bread, rice, pasta, oats, cereals, crackers
- All legumes: beans, lentils, chickpeas, peanuts
- Most fruit: bananas, apples, oranges, grapes, mango (too high in carbs)
- Starchy vegetables: potatoes, sweet potatoes, corn, peas, beets
- Milk and sweetened dairy
- Sugar and all sweeteners except stevia and erythritol
- Alcohol (most, except dry wine and spirits in moderation — each raises blood glucose and disrupts ketosis)
- Processed foods with added sugar or starch
Benefits & honest risks
Every diet has upsides and catches. Here's the balanced picture before you commit.
- Rapid initial weight loss (largely water and glycogen — 3–10 lb in first 1–2 weeks, followed by slower fat loss)
- Short-term fat loss superior to low-fat diets in some but not all RCTs at 6 months; at 12+ months results converge
- Clinically proven for drug-resistant pediatric epilepsy: approximately 50% of patients see ≥50% reduction in seizures
- Meaningful improvements in fasting glucose, HbA1c, and insulin sensitivity in type 2 diabetes in the short-to-medium term
- Triglyceride reduction and HDL increase are commonly observed on keto
- Strong appetite suppression beyond the calorie deficit — ketones and high fat content produce significant satiety
- 'Keto flu' during the adaptation phase (first 1–2 weeks): fatigue, headache, irritability, brain fog, muscle cramps — caused by glycogen depletion, electrolyte loss, and adaptation
- LDL cholesterol often rises on keto, particularly if saturated fat intake is high; response varies significantly by individual and LDL particle size — some individuals are 'hyper-responders' with significant LDL increases
- Long-term adherence is poor: most RCTs show adherence advantage for low-fat over low-carb diets at 12 months
- Constipation from low fiber intake (the low-carb restriction also limits high-fiber legumes and grains)
- Risk of kidney stones — elevated uric acid and calcium oxalate excretion on keto
- Not appropriate without medical supervision for: type 1 diabetes, people on SGLT2 inhibitors (euglycemic DKA risk), kidney disease, liver disease, pregnancy
- Nutrient gaps: restricted legumes, grains, and fruit reduce fiber, folate, B vitamins, and some minerals — supplementation often needed
What the research says
Stronger than most commercial diets for specific outcomes. The A TO Z study (Gardner et al., JAMA 2007) found Atkins (low-carb/keto-adjacent) produced greater 6-month weight loss than Zone, LEARN, and Ornish, but differences diminished at 12 months. Multiple RCTs show short-term superiority for fat loss vs. low-fat diets, with convergence at 1 year. Epilepsy evidence is robust (Cochrane review level). Type 2 diabetes evidence is good in the short-to-medium term (Virva Health two-year trial). Long-term (>2 year) cardiometabolic safety is less well established, and LDL effects vary significantly by individual.
Your sample day
3 scrambled eggs cooked in butter with spinach and 2 strips bacon; black coffee
Large salad with grilled salmon, avocado, cucumber, olive oil and lemon dressing; full-fat cheese
A handful of macadamia nuts; a few raspberries
Ribeye steak, roasted broccoli and zucchini with butter; small side of leafy greens
Small portion of full-fat Greek yogurt (plain) with a few blackberries
What Is the Keto Diet and How Does Ketosis Work?
The ketogenic diet isn't a brand name or a fad — it's a precisely defined metabolic intervention that shifts your body's primary fuel source from glucose to fat. Developed in the 1920s at Mayo Clinic by Dr. Russell Wilder for pediatric epilepsy, the diet has since been applied broadly to weight loss and metabolic health.
The mechanism is straightforward: by restricting carbohydrates to fewer than 20–50 grams of net carbs per day (net carbs = total carbohydrates minus fiber), you deplete liver glycogen within 1–3 days. Once glycogen stores are exhausted, your liver begins converting fatty acids into ketone bodies — primarily beta-hydroxybutyrate, acetoacetate, and acetone — which your brain and muscles can use for fuel. This metabolic state is called nutritional ketosis.
Blood ketone levels between 0.5 and 3.0 mmol/L of beta-hydroxybutyrate indicate nutritional ketosis. Below 0.5 mmol/L, you're not in ketosis; above 3.0 mmol/L, you're approaching the range of therapeutic ketosis used for epilepsy management, which requires medical supervision. The standard macronutrient breakdown for achieving this state is approximately 70–80% of calories from fat, 10–20% from protein, and 5–10% from carbohydrates.
Protein is deliberately kept moderate rather than high because excess protein can be converted to glucose through gluconeogenesis, potentially disrupting ketosis. The typical guidance is approximately 0.6–0.8 grams of protein per pound of lean body mass. Several commercial variations exist: 'lazy keto' tracks only carbohydrates without monitoring full macros; 'dirty keto' permits any fat sources including processed foods as long as carb limits are met; 'clean keto' emphasizes whole-food fat sources like avocado, olive oil, and fatty fish; 'targeted keto' allows small amounts of carbs around workouts; and 'cyclical keto' incorporates periodic higher-carb days.
Bottom line
Ketosis is a real and precisely defined metabolic state — not a brand name. Understanding what actually triggers and maintains ketosis (carb restriction below the individual threshold, typically 20–50 g net carbs/day) clarifies why the diet requires strict carb tracking and why 'cheating' ends ketosis quickly. The diet's mechanism is specific; its execution demands specificity.
Keto Food List — What You Can Eat, What to Avoid, and Where People Go Wrong
The most common reason people 'try keto and don't lose weight' is hidden carbohydrates — in sauces, dressings, processed meats, dairy products, and nuts consumed in unrestricted quantities. Successful keto requires tracking net carbs accurately for at least the first few weeks until you understand your personal thresholds.
The foods that surprise people most: nearly all fruit is too high in carbohydrates — a single medium apple contains roughly 25 grams of net carbs, which is an entire day's allowance on strict keto. Legumes are excluded entirely; half a cup of black beans contains approximately 13 grams of net carbs. Even some nuts require portion control — cashews contain roughly 8 grams of net carbs per ounce, while macadamia nuts and pecans are much lower at 1–2 grams per ounce.
Foods people overeat on keto thinking they're unrestricted: cheese and nuts both have meaningful carbohydrate counts in large portions. An ounce of cheddar has less than 1 gram of carbs, but four ounces — easy to consume as a snack — pushes 4 grams. The 'dirty keto' versus 'clean keto' distinction matters here: dirty keto permits any fat source within carb limits, including processed meats and fast food, while clean keto emphasizes whole-food fat sources like avocado, fatty fish, olive oil, and nuts.
Electrolyte management is essential. The water weight lost during glycogen depletion also flushes out sodium, potassium, and magnesium. Prioritize sodium from salt and broth, potassium from avocado and leafy greens, and magnesium from nuts and seeds. Without adequate electrolytes, keto flu symptoms — fatigue, headache, muscle cramps — are significantly worse and can derail adherence in the first week.
- Fatty meats: beef, lamb, pork, bacon, sausage (preferably minimally processed)
- Poultry: chicken thighs, wings, skin on
- Fatty fish: salmon, mackerel, sardines, herring
- Eggs: whole eggs, any preparation
- Full-fat dairy: butter, hard cheese, cream, full-fat Greek yogurt in small amounts
- Avocado: whole or as oil
- Nuts and seeds: almonds, walnuts, pecans, macadamia, chia, flaxseed (portions controlled)
- Low-carb vegetables: leafy greens, broccoli, cauliflower, zucchini, bell pepper, asparagus, mushrooms
- Oils: olive, avocado, coconut
- Berries in small portions: raspberries, blackberries
Bottom line
The most common reason people 'try keto and don't lose weight' is hidden carbs — in sauces, dressings, processed meats, dairy products, and nuts consumed in unrestricted quantities. Successful keto requires tracking net carbs accurately for at least the first few weeks until personal thresholds are understood.
Does Keto Work? What the Research Says — and at What Time Scale
Keto is one of the most evidence-supported commercial diets for short-to-medium-term fat loss, but its long-term picture is more complicated than most summaries admit. The landmark A TO Z study, published in JAMA in 2007, compared Atkins (a low-carb, keto-adjacent diet) to Zone, LEARN, and Ornish diets. At six months, the Atkins group had lost significantly more weight — but by 12 months, the differences between groups had largely converged. This pattern repeats across multiple randomized controlled trials: keto produces superior short-term weight loss, but the advantage diminishes as adherence drops.
The epilepsy evidence is robust and uncontroversial. A Cochrane review-level body of research shows that approximately 50% of children with drug-resistant epilepsy experience a 50% or greater reduction in seizure frequency on a medically supervised ketogenic diet. This is the diet's original and strongest indication. For type 2 diabetes, the evidence is promising but shorter-term. The Virta Health two-year trial demonstrated meaningful HbA1c reduction and medication de-escalation in patients following a ketogenic diet with remote medical supervision.
The LDL cholesterol question is where keto coverage often becomes dishonest. LDL rises in many keto dieters, and the response varies significantly by individual. 'Lean mass hyperresponders' — typically lean, metabolically healthy individuals — can see dramatic LDL increases on keto. This is not a harmless variation; it warrants a follow-up lipid panel and physician interpretation. Triglycerides typically decrease and HDL typically increases, which are favorable changes, but the net cardiovascular risk depends on LDL particle size and individual context.
The honest long-term adherence picture: the diet that produces the most weight loss is the one you maintain. Most studies show poor adherence to keto at 12 months. The strictest dietary pattern most people will ever attempt outside of clinical supervision is also the one most people abandon within a year. This doesn't mean keto doesn't work — it means its real-world effectiveness is constrained by the same adherence challenges that affect all dietary interventions.
Bottom line
Keto is one of the most evidence-supported commercial diets for short-to-medium-term fat loss and has uniquely strong evidence for epilepsy and promising evidence for type 2 diabetes management. Its Achilles heel is long-term adherence — the strictest dietary pattern most people will ever attempt outside of clinical supervision is also the one most people abandon within a year.
Keto Risks — 'Keto Flu,' LDL, and Who Should Not Use It
Keto's risks are not trivial and are not fully disclosed in most popular summaries. Keto flu — the fatigue, headache, irritability, and muscle cramps that occur during the first 1–2 weeks — is caused by glycogen depletion, electrolyte loss, and the metabolic adaptation to fat-burning. It is manageable with aggressive electrolyte supplementation (sodium, potassium, magnesium) and adequate hydration, but it is unpleasant enough to cause many people to quit before reaching ketosis.
The LDL cholesterol rise is the most under-discussed risk. Some individuals — particularly lean mass hyperresponders — see LDL increases of 50–100 mg/dL or more on keto. This requires a baseline lipid panel before starting and a follow-up panel after 3–6 months. The clinical significance depends on LDL particle size, ApoB levels, and overall cardiovascular risk profile — but dismissing it as 'pattern A versus pattern B' without actual lab work is irresponsible. Anyone with a family history of cardiovascular disease or elevated baseline LDL should discuss keto with their physician.
Constipation is common due to low fiber intake when legumes, whole grains, and most fruits are eliminated. Strategies include emphasizing non-starchy vegetables, chia seeds, flaxseed, and adequate hydration. Kidney stone risk is elevated on keto due to increased uric acid and calcium oxalate excretion — another reason hydration and electrolyte balance matter. The absolute contraindications include type 1 diabetes, pregnancy, kidney disease, liver disease, and pancreatitis.
The SGLT2 inhibitor contraindication is a real, documented clinical hazard that the FDA has warned about. Euglycemic diabetic ketoacidosis occurs when ketosis from carbohydrate restriction combines with SGLT2 inhibitor therapy, producing DKA without the typically elevated blood glucose that would trigger clinical suspicion. Any patient on an SGLT2 inhibitor who is also following a very-low-carbohydrate diet needs active physician awareness. People with type 2 diabetes on any glucose-lowering medication should only attempt keto under medical supervision.
Bottom line
Keto's risks are not trivial and are not fully disclosed in most popular summaries. The SGLT2 inhibitor contraindication (euglycemic DKA) is a real, documented clinical hazard that the FDA has warned about. LDL rise in susceptible individuals requires monitoring. These are manageable with supervision; they are dangerous without it.
Keto and GLP-1 Medications — Synergy and the Protein Problem
The combination of keto and GLP-1 medications like semaglutide or tirzepatide produces powerful appetite suppression through two distinct mechanisms: ketone satiety signals from keto, and GLP-1 receptor agonism from the medication slowing gastric emptying and suppressing hunger signals. The theoretical appeal is obvious — two appetite-control systems working together. The practical result is often very low total food intake, which drives aggressive fat loss but also significant lean-mass loss.
The lean-mass data demands attention. In the semaglutide STEP-1 trial, 45.2% of weight lost was lean mass. This is a substantial proportion, and keto's traditional protein moderation — 10–20% of calories to preserve ketosis — may make this worse, not better. Protecting lean mass on a GLP-1 medication requires a higher protein target, typically around 1.6 grams per kilogram of body weight per day, which exceeds classical keto's protein ceiling.
This creates a practical conflict: the ketosis requirement demands moderate protein to avoid gluconeogenesis disrupting ketone production, while the lean-mass protection requirement demands higher protein. Clinicians often resolve this by allowing a 'modified keto' with higher protein, accepting that occasional gluconeogenesis may occur but prioritizing lean-mass preservation. The trade-off is worth it for most patients — maintaining muscle mass matters more for long-term metabolic health than perfect ketosis.
Lipid monitoring becomes doubly important on this combination. Keto can raise LDL, and GLP-1 medications have their own lipid effects. Anyone combining the two needs a baseline lipid panel and follow-up monitoring. The SGLT2 inhibitor contraindication is especially relevant here: a patient on a GLP-1 medication who is also taking an SGLT2 inhibitor and following keto requires active prescriber awareness — this combination demands medical supervision.
Bottom line
Keto and GLP-1 medications have real synergy for appetite suppression, but the combination requires active management of two specific risks: protein adequacy (keto's protein moderation conflicts with GLP-1's lean-mass loss risk) and LDL monitoring (keto can raise LDL; GLP-1 medications have their own lipid effects). Any patient combining keto with a GLP-1 drug needs prescriber awareness and lipid monitoring.
What most pages leave out
The two most commonly understated points in keto coverage: (1) At 12 months in controlled trials, keto's weight loss advantage over low-fat diets largely disappears — the initial benefit is real, but the long-term result is determined by adherence, not by ketosis per se. (2) LDL cholesterol rises meaningfully in a significant subset of keto dieters — particularly 'lean mass hyperresponders' — and this is not a harmless variation; it warrants physician monitoring and lipid panel assessment, not dismissal.
We flag this so you can make an informed choice — not to scare you off.
❓Frequently Asked Questions
A very-high-fat, very-low-carbohydrate eating pattern that forces the liver to produce ketone bodies from fat, shifting the body's primary fuel source from glucose to fat — a metabolic state called nutritional ketosis. The diet typically restricts net carbohydrates to fewer than 20–50 grams per day.
Most protocols target fewer than 20–50 grams of net carbs per day (total carbs minus fiber). Twenty grams per day is the strictest threshold and induces ketosis most reliably; some individuals can maintain ketosis at the higher end, but personal tolerance varies.
Keto flu is the adaptation phase when glycogen depletes and ketone production ramps up, typically lasting days 3–7. Symptoms include fatigue, headache, irritability, and muscle cramps. Electrolyte supplementation with sodium, potassium, and magnesium significantly reduces symptom severity.
LDL cholesterol rises in many keto dieters, particularly in 'lean mass hyperresponders' — individuals who are lean and metabolically healthy at baseline. Triglycerides typically decrease and HDL typically increases. Total cardiovascular risk depends on LDL particle size and individual lipid response, so a follow-up lipid panel is recommended after starting the diet.
With physician supervision, keto can significantly improve blood glucose and HbA1c in type 2 diabetes. However, people on SGLT2 inhibitors must not follow keto without physician awareness due to euglycemic DKA risk. GLP-1 users should discuss keto with their prescriber to manage protein intake and lipid monitoring.
Typically 1–3 days of keeping net carbs below 20–50 grams per day. Glycogen depletion happens first, followed by ketone production as the liver begins converting fatty acids into ketone bodies. Exercise can accelerate the timeline by depleting glycogen stores faster.
Bread, rice, pasta, oats, and all grains; legumes including beans and lentils; most fruit; starchy vegetables like potatoes, corn, and peas; sugar and sweetened products; milk; and most processed foods — anything that would push net carbohydrates above the 20–50 gram daily threshold.
With prescriber awareness, yes — but the combination requires active management of protein intake to offset lean-mass loss risk and LDL monitoring. Anyone on an SGLT2 inhibitor must not combine it with keto without physician supervision due to euglycemic DKA risk.
Related Articles
One Meal A Day (OMAD) Diet: Rules, Benefits, and Foods to Eat
OMAD (One Meal A Day) is an extreme 23:1 fasting pattern. Learn the rules, what to eat, the weak evidence, and why it's risky on GLP-1 medications.
Read moreOptavia Diet: Rules, Benefits, Foods to Eat, and an Honest Review
Honest Optavia diet review covering the 5 & 1 plan rules, foods list, MLM coaching risks, and why the ~800–1,000 calorie level demands medical supervision.
Read moreMediterranean Diet Guide: Rules, Benefits, and Foods to Eat
Evidence-backed guide to the Mediterranean diet: PREDIMED trial results, complete food list, weight loss reality, and why it's the best dietary pairing for GLP-1 medications.
Read moreNordic Diet Guide: Rules, Benefits, and Foods to Eat
The Nordic Diet is a whole-food pattern with genuine RCT evidence for weight loss and heart health. Learn the rules, foods, and how it compares to Mediterranean.
Read moreNoom Diet Guide: Rules, Benefits, and Foods to Eat
Noom uses a color-coded calorie-density system and CBT-based lessons for weight loss. We review the evidence, costs, food list, and how it works with GLP-1 medications.
Read moreThe Military Diet: A 3-Day Fad Diet Review
The Military Diet is a 3-day fad diet with no military ties. We review the rules, menu, and risks, including a dangerous grapefruit-drug interaction.
Read moreOn a GLP-1, or thinking about one?
Eating well is the foundation. For some people, a GLP-1 medication is the tool that finally makes appetite manageable.Curex connects you with licensed clinicians for compounded GLP-1 medications, if it's right for you.
- Compounded semaglutide from $49/mo, tirzepatide from $149/mo
- Prescribed by licensed clinicians after an online visit
- Delivered to your door — no in-person clinic required
This content is for general informational purposes only and is not medical or nutritional advice, a diagnosis, or a substitute for professional judgment. It does not account for your health, medications, or goals, and nutrition information changes over time. Always talk with a qualified clinician or dietitian before making significant changes to your diet, supplements, or medications. Curex offers compounded GLP-1 medications through licensed clinicians and does not sell or endorse the food or supplement reviewed on this page.