From Wikipedia, the free encyclopedia

Low-carbohydrate diets became a major weight loss and health maintenance trend during the late 1990s and early 2000s.^[1][2][3] While their popularity has waned recently from its peak, they still remain popular.^[4][5] This diet trend has stirred major controversies in the medical and nutritional sciences communities and, as yet, there is not a general consensus on their efficacy or safety.^[6][7] As of 2008 the majority of the medical community remains generally opposed to these diets for long term health^[8][9][10] although there has been a recent softening of this opposition by some organizations.^[11][12]

This article summarizes a sampling of the studies and other research that exist related to this diet trend including not only the efficacy of these diets on weight loss, but also their effects on other aspects of health and related topics such as ketosis. This is not a complete listing of all relevant research.

For general information about low-carbohydrate diets see the main article.

Synopsis

Because of the substantial controversy regarding low-carbohydrate diets, and even disagreements in interpreting the results of specific studies, it is currently difficult to objectively summarize the research in a way that reflects scientific consensus.^[13][14][15]

Although there has been some research done throughout the twentieth century, most directly relevant scientific studies have occurred in the 1990s and early 2000s and, as such, are relatively new and the results are still debated in the medical community.^[16] Supporters and opponents of low-carbohydrate diets frequently cite many articles (some times the same articles) as supporting their positions.^[17][18][19] One of the fundamental criticisms of those who advocate the low-carbohydrate diets has been the lack of long-term studies evaluating their health risks.^[20][21] This has begun to change as longer term studies are emerging.^[22]

Specific Research

The following is not a complete list of all relevant research but a selected list of articles demonstrating some of the breadth of scientific knowledge available on this subject.

Journal of the American Medical Association: 1926

Lieb et al., 1926^[23] conducted a case study of Dr. Vilhjalmur Stefansson, an anthropologist and explorer who lived with the Inuit eating a diet consisting almost entirely of meat, fish, and fat. A research team studied Stefansson's health looking for signs that his "unusual" diet had adversely affected his health. The team was unable to find any health problems in Stefansson and noted that the Inuit themselves also were quite healthy. Note that Dr. Stefansson's diet, like that of the Inuit, included substantial amounts of raw liver, an essential source of vitamin C for this diet.

Journal of Biological Chemistry: 1929

Tolstoi, 1929^[24] performed a one-year study of two men who consumed a diet of exclusively meat. One of the subjects demonstrated a dramatic increase in cholesterol levels while the other actually lowered his cholesterol levels by the end of the study. All other health indicators were normal for both subjects.

Note that because of the magnitude of the reported increase (and decrease following the test period) in the cholesterol levels, the measurement methodology has to be questioned by modern standards (i.e. the numbers may not directly correlate to modern cholesterol measurement techniques).

The Lancet: 1956

Kekwick and Pawan, 1956^[25] conducted a study of subjects consuming 1000-calorie diets, some 90% protein, some 90% fat, and some 90% carbohydrates. Those on the high fat diet lost the most, the high protein dieters lost somewhat less, and the high carbohydrate dieters actually gained weight on average.

Kekwick and Pawan noted irregularities in their study (patients not fully complying with the parameters of the study). As such the validity of the conclusions has to be questioned.

Annals of Internal Medicine: 1965

A study conducted in 1965 at the Naval Hospital Oakland (Oakland, California) used a diet of 1000 calories per day, high in fat and limiting carbohydrates to 10 grams (40 calories) daily. Over a ten-day period, subjects on this diet lost more body fat than did a group who fasted completely (Benoit et. al. 1965). Some advocates, such as Atkins, of low-carbohydrate diets have termed this the metabolic advantage of such diets.

Journal of the American Dietetic Association: 1980

Larosa, 1980^[26] followed 24 patients for 12 weeks, 8 weeks on a high-protein, low-carbohydrate diet, studying their weight loss and cholesterol changes. The authors concluded the following.

The high-protein, low-carbohydrate dieting resulted in substantial weight loss ... Significant increases occurred in LDL-cholesterol, uric acid, and free fatty acid levels. HDL-cholesterol levels failed to rise despite significant weight loss ...

Despite weight loss the study found serious negative metabolic changes.

American Journal of Clinical Nutrition: 1997

Holt et al., 1997^[27] performed a study of glucose and insulin responses for test subjects to a variety of foods, both high- and low-carbohydrate. The conclusions state the following.

Our study was undertaken to test the hypothesis that the postprandial insulin response was not necessarily proportional to the blood glucose response and that nutrients other than carbohydrate influence the overall level of insulinemia ... The results of this study confirm and also challenge some of our basic assumptions about the relation between food intake and insulinemia. Within each food group, there was a wide range of insulin responses, despite similarities in nutrient composition ... As observed in previous studies, consumption of protein or fat with carbohydrate increases insulin secretion compared with the insulinogenic effect of these nutrients alone (22, 30-32) ... However, some protein and fat-rich foods (eggs, beef, fish, lentils, cheese, cake, and doughnuts) induced as much insulin secretion as did some carbohydrate-rich foods (eg, beef was equal to brown rice and fish was equal to grain bread).

This study challenges the general assertion that only carbohydrates significantly impact insulin production.

The authors describe their work as "preliminary" and so the results should be judged with caution.

Journal of the American College of Nutrition: 2000

Anderson et al., 2000^[28] performed computer simulations analyzing eight popular diet programs including Atkins and Protein Power. Their conclusions state the following.

The Atkins and Protein Power diets are very high in total and saturated fat compared to current dietary guidelines. Long-term use of these diets for weight maintenance are likely to significantly increase serum cholesterol concentrations and risk for CHD.

...

While high fat diets may promote short-term weight loss, the potential hazards for worsening risk for progression of atherosclerosis or atherosclerotic events override the short-term benefits.

Note that this was not a study of human trials, only a theoretical computer analysis.

American Journal of Kidney Diseases: 2002

Reddy et al., 2002^[29] studied ten subjects consuming a low-carbohydrate high-protein (LCHP) diet over six weeks. The patients were found to have substantially increased calcium loss compared to their conventional diet among other negative health indicators. The authors conclude the following.

Consumption of an LCHP diet for 6 weeks delivers a marked acid load to the kidney, increases the risk for stone formation, decreases estimated calcium balance, and may increase the risk for bone loss.

Journal of the American College of Nutrition: 2002

Bowman et al., 2002^[30] completed a survey study of 10,014 adults correlating carbohydrate intake to body mass index in addition to other measurements in the diet. The subjects surveyed had not necessarily been dieting per se. The authors concluded the following.

Adults eating a high-carbohydrate diet are more likely to have Body Mass Index values below 25.

This suggests that over the long run those eating more carbohydrates have better body proportion.

New England Journal of Medicine: 2003

Two important NEJM studies from this year are mentioned here. Samaha et al., 2003^[31] completed a study of 132 obese subjects comparing the efficacy of low-carbohydrate and low-fat diets. The conclusions of the article state the following.

Severely obese subjects with a high prevalence of diabetes or the metabolic syndrome lost more weight during six months on a carbohydrate-restricted diet than on a calorie- and fat-restricted diet, with a relative improvement in insulin sensitivity and triglyceride levels, even after adjustment for the amount of weight lost. This finding should be interpreted with caution, given the small magnitude of overall and between-group differences in weight loss in these markedly obese subjects and the short duration of the study. Future studies evaluating long-term cardiovascular outcomes are needed before a carbohydrate-restricted diet can be endorsed.

Foster et al., 2003^[32] performed a study with 63 obese subjects randomly assigned either to low-carbohydrate or conventional low-fat diets for one year. Their conclusion was the following.

The low-carbohydrate diet produced a greater weight loss (absolute difference, approximately 4 percent) than did the conventional diet for the first six months, but the differences were not significant at one year. The low-carbohydrate diet was associated with a greater improvement in some risk factors for coronary heart disease. Adherence was poor and attrition was high in both groups. Longer and larger studies are required to determine the long-term safety and efficacy of low-carbohydrate, high-protein, high-fat diets.

Foster's study claims to be the first truly randomized, controlled study of the efficacy and safety of low-carbohydrate diets.

In essence these studies showed that, setting aside their short-term nature and some safety questions, the low-carbohydrate diet was at least somewhat more effective in weight loss and in improvement of other health issues in an important demographic.

Journal of the American Medical Association: 2003

Bravata et al., 2003^[33] conducted a literature search study of low-carbohydrate diet studies conducted between 1966 and 2003. The paper stated the following conclusion.

There is insufficient evidence to make recommendations for or against the use of low-carbohydrate diets, particularly among participants older than age 50 years, for use longer than 90 days, or for diets of 20 g/d or less of carbohydrates. Among the published studies, participant weight loss while using low-carbohydrate diets was principally associated with decreased caloric intake and increased diet duration but not with reduced carbohydrate content.

The study determined that carbohydrate reduction did not significantly contribute more to weight loss than simply reducing calories. The article does state that

Low-carbohydrate diets had no significant adverse effect on serum lipid, fasting serum glucose, and fasting serum insulin levels, or blood pressure.

Journal of Child Neurology: 2003

Evangeliou et al., 2003^[34] completed a 6-month study of 30 autistic children following a low-carbohydrate, ketogenic diet. The paper stated the following conclusions.

Of the remaining group who adhered to the diet, 18 of 30 children (60%) [the rest did not complete the study], improvement was recorded in several parameters and in accordance with the Childhood Autism Rating Scale. Significant improvement (> 12 units of the Childhood Autism Rating Scale) was recorded in two patients (pre-Scale: 35.00 ± 1.41[mean ± SD]), average improvement (> 8–12 units) in eight patients (pre-Scale: 41.88 ± 3.14[mean ± SD]), and minor improvement (2–8 units) in eight patients (pre-Scale: 45.25 ± 2.76 [mean ± SD]).

The authors state clearly that the study was limited and the results are preliminary.

Journal of the American Academy of Neurology: 2003

Kossoff et al., 2003^[35] conducted a small study of six epileptic patients studying the effects of the Atkins diet. The abstract states the following.

The ketogenic diet is effective for treating seizures in children with epilepsy. The Atkins diet can also induce a ketotic state, but has fewer protein and caloric restrictions, and has been used safely by millions of people worldwide for weight reduction. Six patients, aged 7 to 52 years, were started on the Atkins diet for the treatment of intractable focal and multifocal epilepsy. Five patients maintained moderate to large ketosis for periods of 6 weeks to 24 months; three patients had seizure reduction and were able to reduce antiepileptic medications. This provides preliminary evidence that the Atkins diet may have a role as therapy for patients with medically resistant epilepsy.

In a 2004 Lancet article,^[36] Dr. Kossoff also stated that

The ketogenic diet is a high-fat, adequate protein, low carbohydrate diet that has been used for the treatment of intractable childhood epilepsy since the 1920s ... Although less commonly used in later decades because of the increased availability of anticonvulsants, the ketogenic diet has re-emerged as a therapeutic option.

Harvard University: 2003

Greene et al. studied participants consuming one of three diet regimens over 12 weeks: a low-fat diet, a low-carbohydrate diet with the same number of calories, and a low-carbohydrate diet with 300 extra calories per day.^[37] The researchers found that the low fat fat group lost 17 pounds on average, the low carbohydrate group eating the same number of calories lost 23 pounds, and the low-carbohydrate group eating more calories lost 20 pounds. In commenting on their results Greene stated

There does indeed seem to be something about a low-carb diet that says you can eat more calories and lose a similar amount of weight ...^[38]

Annals of Internal Medicine: 2004

Two significant studies can be found in the Annals of Internal Medicine in 2004. Yancy et al., 2004^[39] completed a study of 120 overweight, high-lipid-count subjects comparing the efficacy of low-carbohydrate and low-fat diets. The conclusions of the article state the following.

Compared with a low-fat diet, a low-carbohydrate diet program had better participant retention and greater weight loss. During active weight loss, serum triglyceride levels decreased more and high-density lipoprotein cholesterol level increased more with the low-carbohydrate diet than with the low-fat diet.

Stern et al., 2004^[40] conducted a one-year study of 132 obese adults. The conclusions state the following.

Participants on a low-carbohydrate diet had more favorable overall outcomes at 1 year than did those on a conventional diet. Weight loss was similar between groups, but effects on atherogenic dyslipidemia and glycemic control were still more favorable with a low-carbohydrate diet after adjustment for differences in weight loss.

Nutrition Journal: 2004

Feinman and Fine, 2004 present an argument refuting the "calorie is a calorie" principle cited by some as an argument against the weight-loss benefits of low-carbohydrate diets.^[41] The "calorie is a calorie" argument, loosely speaking, states that the laws of thermodynamics imply that calories ingested from any source are burned at the same rate in the body (meaning that, for the purposes of weight loss, all sources of calories are the same).

The paper refutes this (the argument is omitted here) stating the following in the conclusion.

Thus, ironically the dictum that a "calorie is a calorie" violates the second law of thermodynamics, as a matter of principle.

The authors' point is that while some have argued that there is no point in comparing the effectiveness of diets based on the sources of calories (proteins, fats, or carbohydrates), the arguments in favor of this viewpoint are not supported by science. This paper is not directly based on any clinical studies but rather is a discussion of basic scientific theory related to this subject.

Cancer Epidemiology, Biomarkers & Prevention: 2004

Romieu et al. 2004^[42] completed a survey-based study of a selected group of 475 women against a control group of 1391 correlating diet and breast cancer rates. The study concluded the following.

In this population, a high percentage of calories from carbohydrate, but not from fat, was associated with increased breast cancer risk.

Lancet: 2004

Astrup et al., 2004^[43] completed a Rapid Review of published research regarding low-carbohydrate diets. The authors concluded the following.

A systematic review of low-carbohydrate diets found that the weight loss achieved is associated with the duration of the diet and restriction of energy intake, but not with restriction of carbohydrates.

...

There is an urgent need for longer and larger studies in obese and moderately overweight individuals ... The studies should be sufficiently long (up to 2 years) to enable careful monitoring of cardiovascular risk factors during the weight-stability phase, and should also include obese individuals with impaired glucose-tolerance to examine the potential of low-carbohydrate diets to prevent type 2 diabetes.

...

Patients who want to try these diets should be told that, although safety cannot be guaranteed, they seem to be safe for short-term use (up to 6 months) as long as weight loss occurs.

Although the authors question the merits of low-carbohydrate diets they nevertheless recommend longer-term studies to judge them more effectively.

American Journal of Clinical Nutrition: 2004

Mozaffarian et al.^[44] studied 235 postmenopausal women with coronary heart disease for 3 years. They monitored intake of fat and carbohydrate as well as progression of the narrowing of the arteries (atherosclerosis). The authors conclude the following.

In postmenopausal women with relatively low total fat intake, a greater saturated fat intake is associated with less progression of coronary atherosclerosis, whereas carbohydrate intake is associated with a greater progression.

An editorial in the same journal observed that these and other results seem to entirely contradict the prevailing theory that saturated fat causes heart disease.^[45]

American Journal of Epidemiology: 2005

Ma et al., 2005^[46] completed a one-year study of 572 healthy adults monitoring their diet and physical activity. The study concluded the following.

In conclusion, results from our study suggest that daily dietary glycemic index is independently and positively associated with BMI [Body Mass Index]. This finding is consistent with the hypothesis that with increased glycemic index, more insulin is produced and more fat is stored, suggesting that type of carbohydrate may be related to body weight. Our data did not support the current public trend of lowering total carbohydrate intake for weight loss or of lowering glycemic load for weight loss, as suggested by other researchers.

This study refutes the suggestion that total carbohydrate consumption directly correlates with weight loss but does support the notion that the glycemic index of foods consumed correlates with weight loss. The study does not specifically distinguish between nutritive and non-nutritive carbohydrate consumption nor is it clear that any of the diets was ketogenic (a key factor for most low-carbohydrate diets).

Journal of Nutrition and Metabolism: 2005

Yancy et al., 2005^[47] completed a study of 28 overweight subjects with type 2 diabetes. The conclusion of the study was the following.

The LCKD [low carbohydrate, ketogenic diet] improved glycemic control in patients with type 2 diabetes such that diabetes medications were discontinued or reduced in most participants. Because the LCKD can be very effective at lowering blood glucose, patients on diabetes medication who use this diet should be under close medical supervision or capable of adjusting their medication.

The article lends support to the argument that low carbohydrate diets can be at least a partial remedy for some forms of diabetes (and may lend support to the argument that some forms of diabetes may in fact be caused by high carbohydrate diets).

New England Journal of Medicine: 2006

Halton et al., 2006^[48] completed a study analyzing the long-term (20 years) health effects of low-carbohydrate diets. The study was limited to women and followed 82,802 subjects. Based on questionnaires, the study determined the correlation between the carbohydrate intake and coronary heart disease risk.

The conclusion in the article states the following.

Our findings suggest that diets lower in carbohydrate and higher in protein and fat are not associated with increased risk of coronary heart disease in women. When vegetable sources of fat and protein are chosen, these diets may moderately reduce the risk of coronary heart disease.

American Journal of Clinical Nutrition, 2006

Johnston et al., 2006^[49] completed a study of 20 subjects over a 6-week period comparing ketogenic low-carbohydrate diets (i.e. very low carbohydrate) and non-ketogenic low-carbohydrate diets (i.e. moderate carbohydrate). The authors of the paper concluded the following.

KLC and NLC diets were equally effective in reducing body weight and insulin resistance, but the KLC diet was associated with several adverse metabolic and emotional effects. The use of ketogenic diets for weight loss is not warranted.

This study suggests that ketosis has no real benefit and is potentially harmful in a diet regimen.

International Journal of Cancer, 2006

Bravi et al., 2006^[50] completed a study of 2301 subjects, 767 with renal cell carcinoma (cancer of the kidneys), analyzing the effects of various types of foods on the risk of developing the cancer. The authors of the paper concluded the following.

A significant direct trend in risk was found for bread (OR = 1.94 for the highest versus the lowest intake quintile), and a modest excess of risk was observed for pasta and rice (OR = 1.29), and milk and yoghurt (OR = 1.27). Poultry (OR = 0.74), processed meat (OR = 0.64) and vegetables (OR = 0.65) were inversely associated with RCC [renal cell carcinoma] risk.

This, in effect, says that bread consumption was strongly correlated with increased risk of this carcinoma whereas the consumption of meats and vegetables decreased the risk.

Nutrition and Metabolism, 2006

Nielsen et al., 2006^[51] completed a study of type 2 diabetics randomly assigned to low-carbohydrate diets and conventional high-carbohydrate diets monitored over 22 months. All test subjects consumed the same amount of calories. Over the first 6 months the low-carbohydrate group was found to have significantly greater weight loss and glycemic control, after which many of the control (high-carbohydrate) group changed diets. The low-carbohydrate group was found to mostly maintain their weight loss and glycemic control through the 22 months of the study. The authors conclude

Weight reduction is primarily caused by decreased caloric intake although decreased energy efficiency has also been found. A high-starch, high-carbohydrate diet excessively stimulates appetite and disturbs energy balance in patients with the metabolic syndrome and type 2 diabetes. A reduction of carbohydrates normalises the balance, reduces insulin concentrations and favours utilization of stored fat as fuel as well as significantly reducing insulin resistance. Weight loss in overweight persons is improved by a higher proportion of protein, presumably due to protein's effect on satiety and/or metabolic efficiency. A reduction in carbohydrates for patients with type 2 diabetes effectively reduces both fasting and postprandial glucose as well as HbA1c. These effects can be independent of weight loss.

Journal of the American Medical Association, 2007

Gardner et al., 2007^[52] studied 311 overweight women each following one of four diet plans (Atkins, Zone, LEARN, and Ornish) in 12-month trials. The authors concluded the following.

In this study, premenopausal overweight and obese women assigned to follow the Atkins diet, which had the lowest carbohydrate intake, lost more weight and experienced more favorable overall metabolic effects at 12 months than women assigned to follow the Zone, Ornish, or LEARN diets. While questions remain about long-term effects and mechanisms, a low-carbohydrate, high-protein, high-fat diet may be considered a feasible alternative recommendation for weight loss.

Epilepsia, 2008

Kossoff et al., 2008^[53] studied adult epileptic patients (as opposed to children used in other studies) following a modified Atkins diet for up to 6 months. The authors concluded the following.

After 3 months, 47% of patients had a >50% seizure reduction, and after 6 months, 33% were similarly improved.

...

When the modified Atkins diet led to seizure reduction, it was relatively quick, usually within 2 weeks.

New England Journal of Medicine, 2008

Shai et al., 2008^[22] studied 322 moderately obese adults over a 2-year period assigning each to one of three diets: a low-fat, restricted-calorie diet (based on the 2000 AHA recommendations); a "Mediterranean", restricted-calorie diet; or a low-carbohydrate, non-restricted-calorie diet (based on the Atkins Diet). Adherence among the participants was high (84.6%). The low-carbohydrate group showed both the greatest weight loss and the most improvement in lipids (cholesterol). The Mediterranean group showed the greatest improvement in glycemic control (related to diabetes). They conclude

The more favorable effects on lipids (with the low-carbohydrate diet) and on glycemic control (with the Mediterranean diet) suggest that personal preferences and metabolic considerations might inform individualized tailoring of dietary interventions.

Interestingly, this study was significant enough that the American Heart Association issued an immediate response to clarify its position (essentially saying that the low-fat diet used in the study is no longer recommended by the AHA and that the AHA's 2006 guidelines emphasize more fiber, vegetables, and "lean" meats).^[54]

PNAS, 2009

Research upon apolipoprotein E(-/-) mice suggest that diets low in carbohydrates and high in protein and fats might reduce vascular regenerative capacity and so increase atherogenesis and through this cardiovascular risk even though this diet does not necessarily effect serum levels of lipids.^[55] The researchers noted "Although caution is warranted in extrapolating from such animal studies, these data at least raise concern that low carbohydrate high-protein diets could have adverse vascular effects not adequately reflected in serum risk markers."^[55]

Meta-analytic summaries

Meta-analysis is a method to succinctly summarize and combine the results from multiple individual studies. The following meta-analyses of low carbohydrate diets are limited to randomized controlled trials that directly compare low carbohydrate diets to other diets. Some of the studies listed above are randomized controlled trials and are included in these meta-analyses.

A meta-analysis of randomized controlled trials by the Cochrane Collaboration in 2002 concluded^[56] that fat-restricted diets are no better than calorie restricted diets in achieving long term weight loss in overweight or obese people.

A more recent meta-analysis that included randomized controlled trials published after the Cochrane review^[57][58][59] found that "low-carbohydrate, non-energy-restricted diets appear to be at least as effective as low-fat, energy-restricted diets in inducing weight loss for up to 1 year. However, potential favorable changes in triglyceride and high-density lipoprotein cholesterol values should be weighed against potential unfavorable changes in low-density lipoprotein cholesterol values when low-carbohydrate diets to induce weight loss are considered."^[60]

An even more recent meta-study of randomized controlled studies that compared low-carbohydrate diets to low-fat/low-calorie diets found that measurements of weight, HDL cholesterol, triglyceride levels and systolic blood pressure were significantly better in groups that followed low-carbohydrate diets. The authors also found a higher rate of attrition in groups with low-fat diets. They conclude that "Evidence from this systematic review demonstrates that low-carbohydrate/high-protein diets are more effective at 6 months and are as effective, if not more, as low-fat diets in reducing weight and cardiovascular disease risk up to 1 year." They also call for more long-term studies.^[61]

References