Obesity is a major contributor to morbidity and mortality in the United States. In recent years, low-carbohydrate diets have attracted increasing interest as a means of losing weight. However, questions about the long-term efficacy and safety of low-carbohydrate diets remain. Specifically, concerns have been raised about the accumulation of ketones and impaired liver and kidney function, salt and water depletion, and excessive consumption of animal proteins and fats. Bravata and colleagues examined the medical literature on low-carbohydrate diets.

A MEDLINE search of English-language studies published between January 1, 1966, and February 15, 2003, identified 2,609 titles and abstracts. Bivariate analyses estimated the differences in weight loss, serum lipid levels, fasting serum glucose levels, fasting serum insulin levels, and blood pressure between two groups participating in lower and higher carbohydrate diets. The authors characterized diets as lowest carbohydrate diets (up to 20 g per day), lower carbohydrate diets (between 21 and 60 g per day), and higher carbohydrate diets (more than 60 g per day).

A total of 107 studies met the authors' inclusion criteria. Study designs were highly heterogeneous. The lower carbohydrate diets were generally of shorter duration than the higher carbohydrate diets; only five studies of low-carbohydrate diets lasted more than 90 days, and none of these was randomized or controlled. Lower carbohydrate diets had fewer calories than higher carbohydrate diets.

The authors found that the greatest weight loss was achieved by participants who took in the fewest calories, regardless of carbohydrate content. In 34 of 38 lower carbohydrate diets, weight loss appeared to be greater than in higher carbohydrate diets, but when only the randomized controlled trials were included in the analysis, the difference between these diets was not statistically significant. In 13 studies of lowest carbohydrate diets, the heterogeneity of study design led the authors to conclude that the 1.2 kg (2.6 lb) weight change associated with these diets was not statistically significant. In the studies demonstrating the greatest weight loss, participants were significantly overweight at baseline, and the diets were longer in duration and in restricted caloric intake. In general, no change was noted in serum lipid levels, fasting glucose levels, or blood pressure.

Bravata and colleagues conclude that there is insufficient evidence to make recommendations on low-carbohydrate diets because studies were short-term, have included only certain types of patients, and have not adequately measured biologic indexes. Caloric restriction and diet duration correlated with weight loss, but the authors found insufficient evidence to determine that low-carbohydrate diets independently produce more weight loss than higher carbohydrate diets. There is a need for studies of isocaloric diets that compare different carbohydrate contents. Lower carbohydrate diets may be effective in achieving weight loss over the short-term without compromising health, as measured by effects on glucose and lipid levels, and blood pressure. This conclusion does not apply to persons with diabetes, nor can any statements or recommendations be made for longer term use of these diets.