Dr. Saltzman is Assistant Professor of Medicine, Director, Obesity Consultation Center, and Chief, Division of Clinical Nutrition, Tufts-New England Medical Center, and Scientist II, Jean Mayer USDA Human Nutrition Research Center on Aging, Tufts University School of Medicine, Boston.

Within the past three years, Dr. Saltzman has been a consultant for Ortho-McNeil and has been on the Speakers' Bureau for Roche and Abbott Pharmaceuticals.

There is no denying it — Americans are fat. A shocking two-thirds of Americans are now classified as overweight or obese. Even worse, they are getting fatter. According to a 1999 survey, 44% of women and 29% of men currently were trying to lose weight. While it is clear that losing weight is difficult, keeping lost weight off is even more so. Approximately 85% of those who do manage to lose weight will return to their original weight within five years.

It comes as little surprise, then, that weight loss diets are all over the media and the bestseller list. Many of these diets call for significant changes in the proportion of dietary macronutrients — that is, carbohydrates, protein, and fat — we eat, as compared to the "average" U.S. diet. Among the most popular are programs that call for drastically reduced consumption of carbohydrates — the so-called "low carb" diets.

Let's look at what we know about the theory behind these diets, as well as what research studies say about their benefits and safety.

The Goal: a Balance between Energy Intake and Energy Output
In order to change body weight, we must create an imbalance between energy intake and energy expenditure; that is, we must burn more calories than we consume. There are three main types of energy expenditure. The first is called resting energy expenditure (REE); this is the number of calories we burn when our bodies are at rest. Believe it or not, REE accounts for approximately two-thirds of our daily expenditure of calories. Exactly how many calories we burn through REE is determined by an individual's proportion of lean body mass to fat. Leaner, more muscular bodies burn more calories while at rest than do bodies with less muscle and more stored fat. The second is called the thermic effect of feeding (TEF), or the amount of calories expended during food intake, digestion and absorption; this accounts for approximately 10-15% of daily energy expenditure. Finally, there is energy burned up through physical activity (Figure 1), which accounts for the remaining 15-20%.

Figure 1.
The Components of Energy Expenditure.
Figure 1

The amount of calories we eat is influenced by many things, including hunger and emotional factors, as well as environmental factors such as food availability and social norms. Many dietary factors also influence food intake, (Table 1).

Table 1.
Dietary Factors that Influence Energy Intake
  • Macronutrient composition (see below)
  • Fiber content
  • Taste and palatability
  • Energy density (calorie content per unit weight of food)
  • Glycemic index
  • Food form (liquid vs. solid)
  • Location (home, restaurant, etc.)
  • Food variety

We can look at dietary macronutrient composition in two ways: either as an absolute amount of each macronutrient — for example, x number of grams of fat — or as the percentage of our total energy intake (i.e., calories) supplied by each macronutrient — for example, fat consumption as x % of total calories. When we speak of a percentage of total dietary consumption, it is important to remember that any reduction in carbohydrate — without a reduction in overall calories — will necessarily mean an increase the percentage of fat and/or protein in the diet. In order to assess the effects of a low percentage carbohydrate diet, therefore, we also have to take into account the effects of increasing the percentages of other macronutrients.

Do Low Carbohydrate Diets Influence Energy Expenditure and Nutrient Partitioning?
The energy that we take in when we eat must be either stored as body energy, or used during the conversion of food to chemical energy to power our cells, or released as heat, or excreted (as with uncontrolled diabetes). This is called nutrient partitioning. It would be wonderful if a particular diet could somehow influence the body to burn more of the calories it consumes and store fewer. In the ideal weight loss program, we would want the composition of the diet itself to cause us to lose body fat while not affecting muscle and other tissues.

Unfortunately, studies using weight maintenance diets have found that changes in macronutrient composition have little effect on nutrient partitioning; the most significant factor by far was the total amount of calories consumed, regardless of the form those calories took. In studies where subjects were deliberately overfed, the most important factor in determining how much energy was stored as body fat was the amount of excess calories; again, macronutrient composition had little effect on how much of the excess energy was stored as fat.

When individual macronutrients are compared, there are some differences in how efficiently excess amounts are stored as body fat. Efficiency may sound like a good thing, but in this case being efficient means that they are more easily stored as body weight. Protein is stored less efficiently than carbohydrate, which in turn is stored less efficiently than fat. But the range is rather narrow — approximately 90-95% of ingested fat is stored compared to 80-85% of carbohydrate.4 Since no one in the real world eats one macronutrient to the exclusion of another, these differences in efficiency are not very significant. Making moderate changes in the relative amounts of each macronutrient in our diets is unlikely to result in important changes in expended energy. For example, on a 2000 calorie/day diet, shifting protein from 15% of dietary energy to 30%, with a proportional decrease in carbohydrate from 55% to 40%, would increase energy expenditure by only 23 calories a day, or about as many as you might burn by walking a few hundred yards.

One explanation for the differences in energy storage that do exist between nutrients is that different macronutrients have different TEFs. While only a small percentage of fat energy is "burned off" after fat is ingested, the TEF of protein can be more than 25% of ingested calories. Carbohydrate "burn off" is intermediate between fat and protein. These differences in TEF could result in marked loss or gain in body weight over the long term, but only if an extremely high or low amount of a macronutrient were consumed.

In a review of trials prior to 2002, Eisenstein et al. evaluated the effect of low-calorie diets with increased dietary protein (mostly associated with decreased amounts of carbohydrate) on loss of body weight and loss of lean mass. In trials where identical calories were provided, no consistent benefit of increased protein on weight loss was found and these higher protein diets did not spare lean mass during weight loss. Further, diets that induced ketosis had no consistent effect on preservation of lean mass.

Ketosis is the presence in the blood of high levels of acidic substances called ketones, which are produced when there is not enough glucose in the bloodstream, as in starvation, and the body turns to fat as a source of fuel. High levels of ketones make the blood abnormally acid and can be dangerous. Mild ketosis is thought to be a cause of excessive morning sickness in pregnancy; it can also be provoked by a crash diets.

Bravata et al. conducted an exhaustive review of the efficacy and safety of low carbohydrate diets. They found no advantage of lower carbohydrate diets with respect to weight loss or preferential fat loss. According to the researchers, the most important factors in weight loss were the degree of energy restriction and the duration of the study. In other words, regardless of whether they ate relatively more or less fat, protein, or carbohydrate, those who ate less for a longer time lost more weight. In addition, those with greater body mass index lost more weight, which is not surprising given that greater energy expenditure is associated with increased body mass.

Some advocates of low carbohydrate diets have claimed that their diets can cause fat loss while sparing lean mass. Layman and colleagues, in a recent study, provided two reducing diets that differed in protein (mean 125 vs. 68 g/day) and carbohydrate (mean 171 vs. 246 g/day) to overweight women for 10 weeks. They found no significant difference in the total mean weight lost. Body composition assessed by dual x-ray absorptiometry revealed that there were also no statistically significant differences in the amount of fat or lean mass lost between the diets, although there was a trend toward a lower loss of lean mass for those on the higher protein diet.

The amount of weight lost by subjects in Layman is very typical of that experienced by the average dieter, though it is possible that subjects may have needed to lose considerably more weight to allow differences in nutrient partitioning to become statistically significant. The relatively small differences in loss of fat and lean mass observed between the two diets suggest that we need to remain cautious before extrapolating from statistical significance to meaningful differences in real people's diets.

Macronutrient Effects on Energy Intake
It appears that macronutrient composition makes no clear difference on energy expenditure or nutrient partitioning. Within the studies cited above, diets are often isocaloric, (i.e., containing the same amount of calories), so that differences in macronutrient content can be isolated as a factor. However, it is possible that a particular macronutrient composition might result in lower spontaneous energy intake (i.e., snacking), reduced hunger, or greater satiety. Since hunger is a frequent complaint of those trying to decrease intake, it is possible that macronutrient manipulation might result in increased ability to tolerate a low calorie diet.

We do know that the macronutrients do differ in their power to satisfy appetite. In a single meal, for example, the satiating power of protein is superior to carbohydrate, which is in turn superior to fat. If this effect were to persist over weeks or months, it is possible that increased protein content might make a greater overall decrease in intake more attainable.

Evidence for this was found in a study by Skov et al., who utilized an ingenious method to control macronutrient content of the diet while allowing subjects to select their own food. Over a six-month period, subjects were able to choose food from a small shop created by the investigators. The food that was removed from the shop, as well as the amount of food that was consumed, were carefully monitored. After six months, subjects eating a diet with 25% protein/45% carbohydrate/30% fat consumed fewer calories and lost more weight (8.9 kg) than those consuming a diet with 12% protein and a commensurate increase in carbohydrate (loss of 5.1 kg). In this study the increased protein content and decreased carbohydrate content, while still in the range consumed by many Americans, resulted in decreased food intake and superior weight loss.

More recently, several studies have looked at the effects of a very dramatic reduction in carbohydrate intake. Foster et al. educated moderately overweight subjects to follow either a loosely controlled low carbohydrate diet (subjects tried to reduce carbohydrate to an initial goal of 20 g/day of carbohydrate and were given a copy of Dr. Atkins New Diet Revolution) or a low fat diet with a calorie goal of 1200-1500 calories/day). The low carbohydrate diet initially resulted in significantly greater weight loss and at six months the weight loss in the groups was 7.0 vs. 3.2 kg. By the 12-month end of the study, however, the difference in weight between lower and higher carbohydrate diet groups was no longer significant (4.4 vs.2.5 kg).

In another study, Samaha et al.10 instructed extremely obese subjects (mean BMI 43 kg/m2) either to reduce carbohydrate intake to 30 g/day, or to follow a diet of 500 calories/day below expected maintenance needs and to consume less than 30% of total calories from fat. After six months, the low carbohydrate diet resulted in a significantly greater loss of 5.8 kg to 1.9 kg lost on the more conventional diet.

If low carbohydrate diets really do reduce energy intake and body weight, exactly how they do this remains somewhat mysterious. One possible mechanism relies on the concept of glycemic index, or the blood glucose response to food. In short-term studies, lower glycemic index meals suppressed appetite and reduced intake later in the same day in comparison to higher glycemic meals. Proponents of this theory suggest that the insulin response following glucose absorption mediates subsequent food intake, and the blunted insulin response to a lower glycemic index meal translates to decreased food intake. While short trials of this mechanism appear promising, longer-term data have not yet confirmed any effect on weight loss. In the Samaha and Foster studies, as well as others, low carbohydrate diets do not cause improvements in fasting glucose, insulin, or insulin sensitivity.

Ketosis (see above), which is thought to suppress appetite, remains a possible mechanism that might decrease intake. Because the range of foods that can be consumed while following a low carbohydrate diet is limited, it is possible that simple lack of food choices results in better diet compliance. This phenomenon, termed stimulus narrowing, is essentially the elimination of the option to choose foods that might typically be over-consumed. This principle is best illustrated by very restrictive, low calorie diets.

It appears that reducing carbohydrate to very low levels, as with the Foster and Samaha studies, may lead to more weight loss over several months. If hunger is suppressed and intake is reduced, this macronutrient mix might be an effective strategy for some. Critics of this strategy, however, cite both possible effects on bone health and known cardiovascular risk factors from a high fat, high protein diet. High protein intake, especially of sulfur-containing amino acids, can cause bone loss through increased bone resorption and calciuria, or the excretion of calcium in the urine. However, fruit and vegetables can have a buffering effect on these processes and and adequate calcium intake may minimize deleterious effects on bone. At present, experts remain divided on the effects of high protein diets on bone health.

With regard to HDL (the good cholesterol), greater increases have been observed, although the results are unclear in some studies.16 The increase in another lipid, triglycerides, at times observed with diets high in carbohydrates, is also not seen with reduced carbohydrate diets. Does this mean that low carbohydrate diets will not have a negative influence on blood lipids? As noted in an editorial accompanying the Foster and Samaha studies, the idea that we can consume high levels of saturated fat without bad effects on cardiovascular disease is inconsistent with a large body of epidemiologic and intervention data.

What Should We Do?
How can this apparent contradiction be reconciled? First, the low carbohydrate diet studies have been conducted in relatively small numbers of subjects for relatively short periods of time. It is possible that the effects of low carbohydrate diets on lipids may change in the long-term.

Second, it is likely that non-dietary factors help determine the susceptibility of individuals to adverse lipid responses. Genetics may play a significant role, but at present there are no methods to identify susceptible individuals. Without such methods, it would seem prudent for those individuals with very high levels of total LDL cholesterol or triglycerides to avoid diets with extremely high fat or triglyceride intake.

Third, the effect of diet composition may be minimized by negative energy balance. This principle is illustrated by the effects of very low calorie diets on blood glucose control in diabetics. If change in blood sugar is measured over the duration of the very low calorie diet, it becomes clear that reducing intake, even without significant weight loss, has a powerful effect on blood sugar. Within a week of consuming a very low calorie diet, blood sugar decreases quickly and dramatically.


The macronutrient content of a diet affects both sides of the body's energy equation, intake and expenditure. It appears that the greatest effects are on food intake, while the effects on energy expenditure and nutrient partitioning, as well as on body weight, are small. Very low carbohydrate diets may be effective for some, but their long-term benefit remains questionable. In addition, while no overall ill effects on blood lipids or other cardiovascular risk factors have thus far been observed, further research is needed to determine if certain individuals may be more susceptible to the nergative effects of high-fat and other diets that are traditionally associated with increased cardiovascular risk.