It seems as though trust in the mainstream media, confidence in what governments are doing & saying, and the quality and consistency of recommendations from global organizations based on sound science is at an all-time low. Back in January, a poll by Edelman demonstrated this, and the trend is only becoming worse.

Almost 3 months to the day, I posted an article on the importance of data integrity, with a focus on reading beyond the salacious headlines to understand whether a research study has been conducted in a sound manner, without bias. Right on cue, we have another “study” that claims to dismantle a growing dietary lifestyle, the ketogenic lifestyle, as being dangerous, even suggesting it is linked to causing cancer, kidney problems and can be especially concerning to pregnant women. Here’s an example of one of the many articles written using the latest study.

With so many people finding success and fulfillment in the ketogenic lifestyle, such daunting words being used to describe the potentially dastardly effects warrant further review. I thought it relevant to analyze the study itself, including who conducted it, how it was prepared, the quality of the data inputs, whether there was sound logic and/or inherent bias, and whether those following the ketogenic diet should legitimately be concerned. Spoiler alert…there’s a lot to cover, and the study bias is strong, so if you’ve got the stomach for it, read on.

Study Intro

The study & review article was published in Frontiers in Nutrition.org. Here’s a link to the actual study. A few things to note right out of the gate:

• The article headline suggests a focus on weighing the benefits vs. the risks of the ketogenic diet. That’s usually a good place to start, and implies a thorough a review of possible pro’s and con’s associated with this lifestyle.
• The funding for this work was provided entirely by one group, the Physicians Committee for Responsible Medicine, which is noted in the study link. Not knowing anything about this committee, I researched them to understand their mission & vision. Taken directly from their own website:
  • Mission: The Physicians Committee is dedicated to saving and improving human and animal lives through plant-based diets and ethical and effective scientific research.
  • Vision: Creating a healthier world through a new emphasis on plant-based nutrition and scientific research conducted ethically, without using animals.

So, the committee entirely funding this study is (1) a self-admitted, fervent proponent of plant-based diets; (2) against animals being considered part of the food chain or research. Right out of the gate, would you expect a committee with these strong beliefs to review a lifestyle that does not adhere to their plant-based diet approach, and embraces animal-based foods as part of the lifestyle? Study bias #1.

In the first few paragraphs of the study, the authors attempt to outline several different descriptions of the ketogenic diet. One of the diets shown in the table is called the 4:1 diet (which, in this table, is not much different than the 3:1), and is in reference to an extremely high-fat variation that is used to treat patients with Epilepsy. Both the 4:1 and 3:1 diet strategies recommend 3 to 4 grams of fat for each 1 gram of carbohydrate or protein. Hence the 4:1 or 3:1 designation, and this diet application will approach 90% of caloric intake from fat. It needs to be clearly understood that this variation of the ketogenic diet is specifically used to successfully treat people with a debilitating disease, and has been shown to reduce the frequency and severity of seizures in children. It is, however, not the broadly-based ketogenic diet that most people who do not have Epilepsy follow, and should not be the reference point when comparing the pro’s and con’s of a ketogenic diet. Most of the ketogenic lifestyle practitioners follow something closer to the modified Atkins diet or a general ketogenic diet, which suggests a macronutrient ratio in the ranges of 65 – 70% Fat; 20 – 25% Protein; 5 – 10% Carbohydrates, based on personal needs. With this guide of carbohydrate consumption, the body will be in a position to utilize gluconeogenesis and create ketones to fuel the body. Study bias #2.

Defining “Low Carb”

The study also mentions “low carb” diets, being defined as <130 grams of carbohydrates consumed daily. Let’s be clear: this is NOT a ketogenic diet. The only reason this is defined as “low carb” is because the American Dietary Guidelines recommend U.S. citizens consume 55% – 65% of their daily calories from carbohydrates, which based on a 2,000 calorie per day diet, would mean consuming 275 – 325 grams of carbohydrates per day. So, while “low carb” is lower than the guidelines recommendation, it is far from a ketogenic diet, and will not be using ketones as fuel. This becomes an important reference point as we continue reading the study, as the authors are very loose at lumping low-carb and ketogenic under one dietary pathway, which is clearly inaccurate. Study bias #3.

Assuming Micronutrient Deficiency

From the study: “Very-low-carbohydrate diets may lack vitamins, minerals, fiber, and phytochemicals found in fruits, vegetables, and whole grains (6–8). Low-carbohydrate diets are often low in thiamin, folate, vitamin A, vitamin E, vitamin B6, calcium, magnesium, iron, and potassium (9).”

The study is making a broad-based, and incorrect assumption, that by eating “very low carbohydrate”, vegetables may be reduced AND micronutrients may also be low. In actuality, those following a general ketogenic diet, which will include 5 – 10% of calories from carbohydrates, will consume most of those carbohydrates from vegetables.

• Ketogenic practitioners are very clear that some vegetables can be nutrient-dense AND low in carbohydrates, which makes them a perfect inclusion into the ketogenic lifestyle. Foods like cauliflower, asparagus, broccoli, lettuce, spinach, kale, zucchini, peppers, mushrooms, cucumbers and celery are all examples of vegetables that are widely consumed by ketogenic practitioners and, ironically, nicely fit into a plant-based diet.
• With regard to fruit, it is absolutely true that fruits are minimized on the ketogenic diet. Avocados are famously part of a staple ketogenic diet due to their healthy fat, lower carbohydrate, high fiber profile (as well as offering excellent micronutrients). Fruits such as blackberries, raspberries, strawberries or blueberries can also be a part of a ketogenic diet, in moderation, as they are usually less carbohydrate dense. However, in general, fruits are carbohydrate-dense and will usually exceed a persons’ daily carbohydrate limit with 1 – 2 servings. Fruits also contain very high levels of fructose, which places a heavy stress on the liver as all fructose has to be digested through the liver first. Fructose is a major contributor to NAFLD. Interestingly, that is never mentioned in the study.

As far as micronutrient deficiency, let’s take a look at some staple ketogenic foods and their ability to deliver more than enough of what the authors claim are often low.

• Thiamine: Lean pork, salmon, asparagus, flax seeds
• Folate: Asparagus, cauliflower, peppers, avocado, broccoli
• Vitamin A: Fish, meat, eggs, cheese, avocados, peppers & carrots
• Vitamin E: Salmon, shrimp, olive oil, spinach, sunflower seeds, almonds
• Vitamin B6: Peanuts, pistachios, spinach, chicken
• Calcium: Dairy, kale, broccoli, sesame seeds, sardines, Greek yogurt
• Magnesium: Pumpkin seeds, halibut, salmon, peanut butter, spinach, beef, dark chocolate
• Iron: Beef, organ meat, flax seeds, hazelnuts, hemp seeds, swiss chard
• Potassium: Salmon, avocados, tomato, flax seeds, pumpkin seeds, brazil nuts, peanut butter

Notice how many plant-based foods are part of a ketogenic diet. I find it ironic how the authors try to disparage the ketogenic diet, yet ignore how much overlap there is between plant-based (ie, vegetables, seeds, berries) & ketogenic lifestyles. Study bias #4.

Fiber

One of the biggest reasons thrown around to NOT follow the ketogenic diet is it is impossible to get enough fiber in your diet without consuming oats, grains or wheat, which are very carbohydrate dense. Not surprisingly this study trumpets the same long-standing narrative, claiming “Ketogenic diets are typically low in fiber needed not only for healthful intestinal function but also for microbial production of beneficial colonic short-chain fatty acids.” Nothing could be further from the truth, and no, you don’t need to eat wheat and grain to consumer fiber. In fact, we see many foods on the ketogenic diet are robust in dietary fiber. Here’s a short list:

Additionally, one of my favorite additions for fiber is psyllium husk powder. Mix it in a protein shake, blend it within Greek yogurt, add it into baking or apply a light dusting to a mixed salad, and you’ll get a tasteless, insoluble fiber source. Study bias #5.

Obesity & Weight Management

Next, the study proceeds to acknowledge that ketogenic diets can induce weight loss, but then proceeds to combine ketogenic and low carb diets into one group, which we’ve already established is absolutely false.

Example: “Low-carbohydrate diets, defined here as <40% of energy from carbohydrate…”

We’ve already highlighted this is NOT ketogenic, however in the next paragraph, they say, “more than half of the diets studied met the criteria for a general ketogenic diet.” Well, that means that almost half were far away from the ketogenic diet, actually closer to the American Dietary Guidelines of 55%+ from carbohydrates. So, could lumping dietary studies that have very high carbohydrate consumption skew data from true ketogenic studies? Of course.

The next two paragraphs then acknowledge that both very low carbohydrate diets AND low carbohydrate diets (their definition: < 130g/day of carbohydrates) provide, at a minimum, directionally improved weight loss, especially among those who are Type II diabetic.

Finally, this section references studies comparing various dietary pathways, but only evaluated them over a very short period of time (usually 2 – 4 weeks). If you’ve been eating a standard American diet for years, if not decades, and you switch to a ketogenic diet, is there a transition period that will require your body to recalibrate? Absolutely! Two common occurrences upon changing to the ketogenic diet are the keto “flu” and the keto “flush.”

• The keto flu references possible nausea, bloating, fatigue, headaches, mood changes and GI imbalance upon converting to the ketogenic diet. This is driven by the body converting away from burning sugar to creating ketones, and the temporary transition can cause an imbalance in hormones, GI track performance, electrolytes & blood sugar levels (hypoglycemia). It usually lasts for 3 – 10 days.
• The keto flush references a common phenomenon that occurs within the first 2 – 3 weeks where the body flushes out a significant amount of water, triggering a large weight loss in a very short period of time. Rumors that this is due to fat cells accumulating water are false, but the actual mechanism of what happens and why this happens is still not clear. It could be related to a dietary change and the body’s adaptation process; a reduction in swelling due to reduced inflammatory foods consumption; or changes in hormones, including cortisol.

Finally, the section concludes with the statement “These results also add to evidence suggesting that the rapid initial weight loss observed on ketogenic diets is due predominantly to loss of fat-free mass (e.g., body water, glycogen, protein, and contents of the gastrointestinal tract).” Losing fat-free mass is not the same as lean body mass (muscular, skeletal), and if the near-term loss on a ketogenic diet is due to a loss in body water weight, sugar (glycogen), old proteins and GI tract contents, is that a bad thing? Study bias #6.

Type II Diabetes (T2D)

The study spends very little time on this topic because of the inconvenient truth that the ketogenic diet can be very effective at reversing insulin resistance (the precursor to T2D), and even in patients with full-on T2D, the ketogenic diet is being used by medical professionals to greatly reduce, if not eliminate, external insulin. Dr. Sarah Hallberg of Virta Health, a leading practitioner and thought leader in applying lifestyle changes to reverse T2D in her patients, has talked at length about her success. I would encourage you to read more about how she approaches her patients and the success she and her team have been able to achieve through use of the ketogenic diet that have yielded consistent results for 3+ years.

In a classic example of stretching the data to fit a narrative, the study references 4 previous studies that support their assertion that foods associated with the ketogenic diet have been associated with increased T2D risk. Here is, verbatim, what one of those previous studies said:

We captured 86 eligible papers (142 associations) covering a wide range of biomarkers, medical conditions, and dietary, lifestyle, environmental and psychosocial factors. Adiposity, low hip circumference, serum biomarkers (increased level of alanine aminotransferase, gamma-glutamyl transferase, uric acid and C-reactive protein, and decreased level of adiponectin and vitamin D), an unhealthy dietary pattern (increased consumption of processed meat and sugar-sweetened beverages, decreased intake of whole grains, coffee and heme iron, and low adherence to a healthy dietary pattern), low level of education and conscientiousness, decreased physical activity, high sedentary time and duration of television watching, low alcohol drinking, smoking, air pollution, and some medical conditions (high systolic blood pressure, late menarche age, gestational diabetes, metabolic syndrome, preterm birth) presented robust evidence for increased risk of T2DM.

Well, there are about a dozen risks covered in this statement. The only area of any relevance that remotely could be linked back to the ketogenic diet is the consumption of processed meat, and as I’ve covered previously in this blog, processed meat (both animal based AND plant based) should be avoided due to the additives, fillers, nitrates and sugars loaded into these foods. Following a clean ketogenic diet means applying the same rigor to choosing quality foods with clean ingredients, regardless of the type of food. Implying the ketogenic diet must be directly linked with processed meat consumption is simply false and misleading. Study bias #7.

Non-Alcoholic Fatty Liver Disease (NAFLD)

As we’ve already shared, non-alcoholic fatty liver disease (NAFLD) can be traced back to the enormous usage of fructose in our diets, including soft drinks, foods processed with high fructose corn syrup, and fruits. In fact, one of the studies referenced in this study says the following:

Foods rich in fructose are the prototype of unhealthy diet. Fructose, contained in fruits and some vegetables, is structurally similar to glucose; it is mostly metabolized in the liver and is fuel for de novo lipogenesis. 

It’s interesting this study, which specifically highlights NAFLD as a danger, never mentions the role fructose plays in stressing the liver and being a major contributor to NAFLD. I suspect, again, it would be an inconvenient truth for the plant-based foods being championed by the funders of this article to talk about how those foods can contribute to such an unhealthy situation.

The assertion that following a true ketogenic diet can lead to NAFLD is simply untrue. In fact, here is a recent research paper that was done to prove the exact opposite…that converting to keto had a material improvement in NAFLD. Study bias #8.

Cancer & Alzheimer’s

As the study describes, there is anecdotal & cause/effect evidence that the ketogenic diet could aid in treating cancer, but definitely not enough quality research in this area to make any recommendations one way or another. This would be an important area of increased research, especially since the three main causes of cancer are tobacco usage, obesity and radiation exposure, and obesity is largely tied to dietary & lifestyle choices.

As for Alzheimers, it’s interesting how this study tries to spin the positive directional data on the ketogenic diet. The study acknowledges that some previous short term studies have shown positive cognitive improvement through the ketogenic diet, but there are few (if any) long-term studies in this area. But, in an effort to further support their narrative, they quickly draw a loose correlation between saturated fat consumption, the ketogenic diet, and the risk to Alzheimer’s, alluding to how near term benefits of the ketogenic diet may be offset by longer-term risk, without any proof or scientific foundation. It’s absolutely fair to say there is inconclusive evidence that the ketogenic diet can benefit Alzheimer’s patients, but it’s false to say it will cause more harm than any other diet. We just don’t know, yet, but as you can see in the attached, there is some research being done (early stages, on rats) that is seeking to evaluate whether the ketogenic diet can have long-term cognitive improvement. Study bias #9.

Cardiovascular Disease (CVD)

Just like we’ve been suffering from 50 years of misinformation based on Ancel Keys’ flawed “research” and narrative that fat is the enemy and we should seek to reduce it in favor of increasing carbohydrates, the research around the drivers of CVD is suggesting a rethink. The narrative is simple: saturated fat, cholesterol and LDL are the key driver’s of CVD, and anything high in saturated fat should be eliminated.

Of course, this study seeks to regurgitate long-held beliefs that saturated fat and cholesterol through animal food consumption are the enemy, and are the primary contributors to CVD. However, I encourage you to read this 10-year epidemical study across 18 countries on 5 continents, which clearly found a correlation between carbohydrate consumption and mortality, an INVERSE relationship between saturated fat & stroke, and no correlation between saturated fat consumption and CVD mortality.

https://www.thelancet.com/journals/lancet/article/PIIS0140-6736(17)32252-3/fulltext

One of the studies this study quotes is a 1-year study funded by Virta Health to show the impact of dietary changes on T2D. While the actual study results show a reduction in insulin resistance, usage of external insulin, triglycerides and HOMA-IR (-55%) and an improvement in HDL (+18%), which are excellent results, the +10% increase in LDL is the one thing the authors focused upon to support their narrative. What they left out is, if you calculate the study’s participants 10-year CVD risk, it went down by 10%. Again, unfortunate bias in a cherry-picking attempt to tell a narrative.

As the study’s authors switch to HDL, they bizarrely quote studies that are tied to “treatment-induced” HDL, which means likely statins or other drugs. How is that relevant to a ketogenic diet and lifestyle, which will increase your HDL naturally and can help people eliminate the need for medical treatments? There seems to be consistency that for CVD prevention, we do not want low-HDL. How do you ensure your HDL levels are healthy? Ketogenic diet. Study bias #10.

Kidney Function

Mentioning kidney stones as a risk for the ketogenic diet is ironic, given some of the foods that are most likely to contribute to such stones are foods high in oxalates. Which foods are high in oxalates? Most of the foods on the list to avoid are staples of a plant-based diet, while only a few are found on a ketogenic diet.

Several studies have been conducted to show ketogenic diets, and even higher protein diets, pose no threat to people who have healthy kidney function. And that’s the key: regardless of what diet you chose to follow, if you have compromised kidney function, you need to seek medical guidance on which foods to eat, potential supplements to take and lifestyle choices to embrace. Study bias #11.

Conclusion

If you’ve made it this far, well done! We covered a lot of information, but the point of this lengthy blog was to point out how a meta-analysis can cherry-pick data to support a narrative, and be loaded with biases that simply are not true. We have to be more cautious than ever with the ever-increasing amount of misinformation being shared as “news”, and really dig into the details. The ketogenic diet may not be for everyone, can be difficult to follow due to the removal of certain “comfort foods”, and not be appropriate for some people depending on their life stage. But, the diet is not a fad…it’s been around for well over 100 years, and has been successfully followed by millions to improve health and wellness. Unfortunately, it does not fit the narrative of the last 50 years, which has told us to eat lots of carbs and sugars in lieu of healthy fats, and as a result, we are more obese, more chronically sick, more diabetic, more diseased and more addicted than ever.