The Best Diet for Keeping Off the Weight

The Best Diet for Keeping Off the WeightBy Daniel Gwartney, MD

Within the United States of America’s Declaration of Independence’s hallowed words is a statement that makes the remainder of the decree necessary and indisputable: “All men are created equal.” Equality is conceptual, as we are all physically unique. Not every student has a high IQ, the majority of adults are less physically gifted than athletes and few are as photogenic as models.

Though calories have never gathered together in adverse conditions to challenge tyranny, they too are considered equal for no reason other than it seems self-evident. Dietitians have ingrained the caloric value of carbohydrates, protein, fats and alcohol into textbook dogma [4, 4, 9 and 7 calories per gram respectively]. Weight-loss diets have juggled these calories around. Once, it was held firm that all calories are the same and a person just needs to eat fewer. Later, the focus was placed on the macronutrients from which the calories came; for instance, dietary fat was held as the source of stored fat, to be avoided. Cardiologists endorsed the low-fat diet, believing that it would reduce cholesterol and triglycerides (blood fats), which are factors involved in cardiovascular disease.

Thankfully, weight-loss experts emerged who began by thinking rather than reacting. Two diet programs gained substantial following due to real-world success and sound scientific rationale— the Zone Diet by Barry Sears, Ph.D. and the Atkins Diet by Robert Atkins, M.D. Dr. Sears presented the concept of controlling insulin to combat obesity and fat storage; Dr. Atkins went further and promoted carbohydrate restriction, shifting the metabolism to preferentially burning fat for energy rather than sugar.

Further progress came with an appreciation of the glycemic index/load and its effect on insulin. Insulin plays a very clear role in fat storage. It potently inhibits the breakdown of stored fat, indirectly transfers fat from the bloodstream to the fat cell and stimulates the creation of more stored fat.1,2 Carbohydrates are the primary driver of insulin release, and the glycemic index is a measure of how quickly dietary carbs enter the bloodstream. The glycemic load is similar but more appropriate in that it factors in how much of the carbohydrate is consumed. Obviously, a slowly released, low-glycemic carbohydrate can be responsible for a delayed insulin surge if one gorges down a few plates of beans or a bowl of Peanut M&Ms.

Low-fat Diets Could Impede Weight Loss

When insulin is released, it is not a bolt of lightning— here and gone in a flash. Instead, it is like Black Friday shoppers storming a department store as soon as the doors are open, then trickling off over the next few hours as the sales items are cleared off the shelves. If a moderate amount of insulin is stimulated, fat release is not blocked as strongly or for as long— also, the blood concentration decreases to fasting level before long. If a large amount of insulin is secreted, fat release is turned off and insulin remains elevated for quite some time.

Insulin also switches the body away from using fat for energy production, instead burning sugar to generate ATP. As glucose (sugar) is metabolized for ATP, chemical intermediates are formed that interfere with fatty acids entering the mitochondria— the energy furnace of the cell.3 When the insulin concentration remains elevated for a prolonged period— like after gorging on Halloween candy— glucose is shuttled into cells that respond to the hormone at the cost of being available to the brain. This may continue for hours, even after the glucose concentration has fallen below fasting level. The brain becomes sugar-starved, interpreting “post-prandial hypoglycemia”— low blood sugar caused by insulin getting too high and staying elevated too long, as being in a low-energy environment.4,5 When the brain thinks there is not enough energy (calories) available, it slows down the metabolism to preserve energy. This reduces the number of calories the body burns at rest and increases hunger signals, resisting weight loss.

These “triple whammy” actions of insulin on fat loss suggest that the traditional low-fat diet might actually impede weight loss over time. Prior studies have compared low-fat and low-carb diets directly, usually showing quicker weight loss with Atkins-like diets but equivalent long-term results.6-9 A new study published in the Journal of the American Medical Association (JAMA) looks deeper into the issue, comparing three different diets used after weight loss.10 The primary goal of the study was to see which type of diet(s) are most advantageous to support continued weight loss or fight weight regain, based upon metabolic and hormonal measures.

In the JAMA study, obese subjects went on a controlled, 12-week diet providing 40 percent fewer calories than maintenance (45% carbs/30% fat/25% protein), resulting in an average weight loss of 12.5 percent— that is, a 200-pound person losing 25 pounds in three months. They then spent four weeks maintaining a stable weight (100% maintenance diet by calories) before being assigned to test diets in random order. Every subject spent four weeks on each diet, with hormonal and metabolic measures taking place during the last few days of each test diet. The test diets included a low-fat (LF) diet (60%carbs/20%fat/20%protein), a low-glycemic index (LGI) diet (40%carbs/40%fat/20%protein) and very low-carbohydrate (VLC) diet (10%carbs/60%fat/30%protin). All the test diets were equal in calorie content and provided 100 percent of maintenance calories.

Glycemic Index and Load

The Best Diet for Keeping Off the WeightThe primary differences between test diets were the carbohydrate content and glycemic index/load of the meals. The LF diet had six times the carbs compared to the VLC diet (310 grams versus 50); the LGI diet contained 205 grams of carbs, similar to the initial 12-week, hypocaloric weight-loss diet (230 grams). The glycemic index/load logically followed a similar pattern. The LF diet had the highest index and load at 67 and 185, which eclipsed the values of the VLC diet’s index of 28 and glycemic load of only 4, due to the low carbohydrate content of the diet. The LGI diet had a similar index of 33, but higher load than the VLC diet at 51. The 12-week weight-loss diet used at the start of the study had a higher index than the LGI diet, but a similar load due to the calorie restriction.

Since the calories and conditions were identical, and the subjects all underwent each of the three diets in random order, the results of this study offer a fair comparison of the three different diets. At the end of each four-week test diet period, the subjects completed food and activity logs to ensure that they adhered to the diets and had not changed their activity levels. They then spent time undergoing metabolic studies and laboratory measures of hormones related to the metabolic rate. Recall, the purpose of the study was to see if any diet improved or hindered the metabolism after weight loss. A common experience for people who have lost weight is to suffer a regain of the weight, or an inability to continue weight loss due to a slowing of the metabolism. This is a major factor in the long-term failure of most weight-loss efforts.

OK, it is getting a bit “blah, blah, blah,” so what is the best diet to avoid regaining weight after a successful diet? Comparing the three test diets, it was clear that the WORST diet was the traditional, cardiologist recommended, low-fat diet. The same people, consuming the same number of calories, had the greatest decline in resting energy expenditure (REE)— the number of calories burned while resting but awake (-205 kcal/day); 20 percent greater the LGI diet (-166 kcal/day); 50 percent greater than the VLC diet (-138 kcal/day). [The baseline REE was determined prior to the initial 12-week diet.]

A similar trend, but a greater degree of effect, was noted when looking at total energy expenditure (TEE), the total number of calories burned throughout the day. Remember, activity logs and pedometers were monitored to ensure that there was no variation in physical activity. Compared to the TEE before weight loss, the LF diet again caused the greatest decrease in calories used, (-423 kcal/day); 30 percent greater than LGI (-297 kcal/day); 325 percent greater than VLC (-97 kcal/day). The trend for changes in REE and TEE were significantly associated with the glycemic load of the diets. The difference between the LF and VLC diets in TEE was the equivalent of an hour of exercise.

Leptin and Metabolic Rate

Clearly, in terms of maintaining the metabolic rate, VLC appears to offer distinct advantages. However, the metabolic rate is not the only relevant factor. The hormones and blood markers assayed during this experiment revealed distinct differences between the diets, some of which suggest a health advantage for LGI diets over VLC diets.

Leptin is a hormone produced in fat cells, which increases the metabolic rate in lean individuals. Contrary to the changes in energy expenditure described above, leptin concentration was greatest in the LF diet, followed by the LGI and then the VLC diet. As this did not support a higher metabolic rate in the LF diet compared to the LGI or VLC diets (actually, the opposite), it may represent a greater retention of fat mass during the LF diet. Leptin is reduced as fat mass decreases during weight loss; even during the LF diet, leptin was decreased by 50 percent from pre-weight loss levels. The changes in thyroid hormone were similar with all diets being decreased from the start of the study, but with LF maintaining the highest TSH and T3 despite having the lowest REE and TEE.

The VLC diet resulted in maintaining a metabolic rate near the pre-weight loss normal; yet, two hormones (T3 and leptin) closely associated with boosting metabolism were higher in the LF diet, which had the most sluggish metabolism. The researchers discussed this, having no answer for the T3 relationship— it remains a curiosity at this point. Relative to the leptin— when compared to the energy expenditure (a way they determine leptin sensitivity), the subjects had much better leptin sensitivity during the VLC diet, possibly responding better (faster metabolism) despite less leptin being released.

Cortisol, the Stress Hormone

The VLC diet demonstrated an effect that might be considered deleterious by some. When blood glucose concentration is low, the brain signals for the release of hormones that promote the release of stored fat, and the creation of glucose from amino acid or fatty acid metabolites. This latter process is called gluconeogenesis, or creating new sugar. One potent hormone involved in this response is cortisol, also known as “the stress hormone” and the catabolic steroid (as opposed to anabolic steroids). Increased cortisol can break down muscle tissue, and the VLC diet did result in a significantly higher cortisol as would be expected. Also, systemic inflammation was higher during the VLC diet as measured by CRP, a significant risk factor in cardiovascular disease. By comparison, the LGI diet did not increase either cortisol or CRP.

The cholesterol and lipid panel was less consistent in supporting one diet over the other. The LF diet resulted in a decrease in HDL (good) cholesterol, but total cholesterol also dropped. The VLC did not affect HDL, actually it increased non-significantly, but total cholesterol did not change. However, the LGI diet maintained the HDL while lowering total cholesterol, giving it the win in this round. Triglycerides are fats that circulate in the blood and are a negative factor in cardiovascular health. The LF diet was not much different from the pre-weight loss value, but the LGI and VLC diets both dropped triglycerides considerably; most notably in the VLC diet.

Insulin sensitivity refers to how well cells respond to insulin. It is like talking to your great-grandparents. You cannot have a conversation at normal volume; in order to include Pappy and Nana, you have to shout in their ears. When insulin sensitivity falls, more insulin has to be released to get cells to take up the sugar from meals. Compared to the pre-weight loss measures, all diets were associated with better insulin sensitivity, as would be expected. However, the least response was noted during the LF diet.

A Lot to Digest

Put it all together, and there is a lot to digest (pun intended). The loser is clear— it’s the low-fat diet. When comparing the low-glycemic index diet to the very low-carbohydrate diet, the VLC diet is the most potent, but has some unhealthy effects noted in this and other studies— primarily, increased cortisol and CRP (inflammation). The LGI diet is nearly as effective as the VLC diet in maintaining the metabolism, demonstrating healthier effects on cholesterol values while avoiding the increased cortisol and CRP noted during VLC.

Based upon this study, those who have been successful in losing weight would be best served to avoid following a low-fat diet after reaching his/her target weight. The body seems to take a metabolic vacation, lowering the metabolic rate and doing less to avoid the between-meal munchies. On the other hand, both low-glycemic index diets and very low-carbohydrate diets maintain the metabolism better, with the very low-carbohydrate diet being the best option. Long term, the picture is not quite so clear. The very low-carbohydrate diet appears to be more stressful, with elevations in cortisol and the inflammatory marker CRP noted. The ideal diet, balancing metabolic support, cardiovascular benefits and avoiding increased stress/inflammation is the low-glycemic index diet.


1. Choi SM, Tucker DF, et al. Insulin regulates adipocyte lipolysis via an Akt-independent signaling pathway. Mol Cell Biol 2010;30:5009-20.

2. Kersten S. Mechanisms of nutritional and hormonal regulation of lipogenesis. EMBO Rep 2001;2:282-6.

3. Cahová M, Vavrínková H, et al. Glucose-fatty acid interaction in skeletal muscle and adipose tissue in insulin resistance. Physiol Res 2007;56:1-15.

4. Rizkalla SW, Bellisle F, et al. Health benefits of low glycaemic index foods, such as pulses, in diabetic patients and healthy individuals. Br J Nutr 2002;88:S255-62.

5. Pawlak DB, Ebbeling CB, et al. Should obese patients be counselled to follow a low-glycemic index diet? Yes. Obes Rev 2002;3:235-43.

6. Stern L, Iqbal N, et al. The effects of low-carbohydrate versus conventional weight loss diets in severely obese adults: one-year follow-up of a randomized trial. Ann Intern Med 2004;140:778-85.

7. Gardner CD, Kiazand A, et al. Comparison of the Atkins, Zone, Ornish, and LEARN diets for change in weight and related risk factors among overweight premenopausal women: the A TO Z Weight Loss Study: a randomized trial. JAMA 2007;297:969-77.

8. Dansinger ML, Gleason JA, et al. Comparison of the Atkins, Ornish, Weight Watchers, and Zone diets for weight loss and heart disease risk reduction: a randomized trial. JAMA 2005;293:43-53.

9. McAuley KA, Hopkins CM, et al. Comparison of high-fat and high-protein diets with a high-carbohydrate diet in insulin-resistant obese women. Diabetologia 2005;48:8-16.

10. Ebbeling CB, Swain JF, et al. Effects of dietary composition on energy expenditure during weight-loss maintenance. JAMA 2012;307:2627-34.