FYI: This is the full text of Walter Willett's editorial on this study. It's kinda hard to track down on the Internet so I thought I'd include it here.
For decades, nutritionists and dietitians have disparaged the very-low-carbohydrate Atkins diet because it is high in saturated fat and because its purported benefits had not been tested in formal studies. According to the conventional wisdom, a diet low in fat and high in starch reduces the risks for heart disease and cancer and promotes weight loss. However, these claims do not have solid evidence to support them. In large prospective studies, total fat intake does not predict cancer risk (1). In addition, dietary fat per se does not predict the risk for coronary heart disease in ecologic and prospective studies or clinical trials (2, 3); however, intake of specific fatty acids is important. Moreover, high-carbohydrate diets, which reduce high-density lipoprotein (HDL) cholesterol and raise triglyceride levels, exacerbate the metabolic manifestations of the insulin resistance syndrome (4). The primary remaining justification for high-starch diets has been weight control, but even this rationale is on shaky ground (5, 6). Meta-analyses of studies that mostly lasted 6 months or less suggest a small benefit for low-fat diets (typically 20% to 25% of energy) compared with moderate-fat diets (usually 35% to 40% of energy) (5). However, patients on low-fat diets typically regain lost weight. Two meta-analyses of studies lasting 1 year or more found no sustained reduction in weight; the weighted mean difference in weight change (low-fat diet group minus control group) was -0.25 kg in one analysis (7) and +3.7 kg in another (8), even though the low-fat groups had more intensive interventions, which would create bias in their favor.
Recently, 4 randomized trials in adults have compared very-low-carbohydrate diets with low-fat diets (9-12). This issue contains 2 of these studies: the 6-month study by Yancy and colleagues (11) and the 12-month report by Stern and colleagues (12); the latter is an update of a report of weight loss after 6 months (13). Although the trials differed in the target population, study design, and intensity of intervention, all 4 studies found that weight loss at 6 months was 4 to 6 kg greater in the low-carbohydrate group than in the low-fat group. In the 2 studies that lasted for 1 year, the differences in weight loss had narrowed to 2 kg by 1 year and were no longer statistically significant. In the report by Foster and colleagues (10), this convergence occurred because the low-carbohydrate group regained weight. In the study by Stern and colleagues (12), the low-carbohydrate group maintained the weight loss seen at 6 months, whereas the low-fat group continued to lose weight.
The low-carbohydrate diet had similar effects on blood lipid levels in the 4 studies. The low-carbohydrate, high-fat diets minimally changed low-density lipoprotein (LDL) cholesterol levels, slightly increased HDL cholesterol levels, and markedly decreased fasting triglyceride levels. The changes in HDL cholesterol and triglyceride levels are consistent with the known effects of reducing carbohydrate intake and body fat. The small effects on total and LDL cholesterol probably result in part from greater weight loss, which reduces total and LDL cholesterol levels, thereby offsetting the effects of a higher intake of saturated fat. Although the low-carbohydrate diet had little effect on weight loss by the end of the two 12-month studies, its favorable effects on triglyceride and HDL cholesterol levels persisted. Thus, despite earlier concerns, the low-carbohydrate diets did not, on average, harm blood lipid levels. The dropout rates have tended to be higher for persons on the low-fat diet; this finding is important because the value of any diet depends on the degree to which patients adhere to it over time.
We can no longer dismiss very-low-carbohydrate diets. The findings raise important questions, foremost being the long-term effect on weight. Almost any diet can lead to weight loss over 6 months, but most individuals regain weight. The earlier experience with low-fat diets and now the diminished effects of low-carbohydrate diets at 12 months emphasize the need for additional randomized trials that last for longer than 1 year and that monitor weight and lipid profile. Such studies should compare various levels of fat, different forms of carbohydrate, and combinations with physical activity programs. More intensive re-enforcement of the dietary interventions is another strategy; whether it will lead to sustained differences in weight requires further research.
If low-carbohydrate diets are effective, what pathophysiologic mechanisms would be responsible? One underlying hypothesis is that a high dietary glycemic load (the contribution to blood glucose of all foods consumed in a meal) increases the difficulty of weight control because the high intake of refined starches and sugars causes rapid swings in blood insulin and glucose levels; these, in turn, stimulate hunger between meals and lead to more snacking. This mechanism has support (14). Some have dismissed low-carbohydrate diets because participants have reported lower caloric intake, even though the diet does not specifically restrict calories. However, this criticism is hardly valid because a diet that facilitates lower caloric intake is the key to sustained weight loss. Whether the mild ketosis that occurs with severe carbohydrate restriction contributes to weight loss is unclear; notably, only half of the participants in the study by Yancy and colleagues remained ketotic during the follow-up, and even fewer did so in the less intensive study by Samaha and colleagues (13). Ketosis may simply be part of a beneficial continuum of carbohydrate restriction. Alternatively, perhaps monitoring for ketosis may be an effective feedback mechanism for maintaining dietary adherence. Some have suggested that food energy is used less efficiently during ketosis, but as Yancy and colleagues point out, testing this hypothesis will require monitoring and much more precise measures of intake and energy expenditure.
How can we maximize the health-enhancing effects of the Atkins diet? Dr. Atkins deserves credit for his observations that many persons can control their weight by greatly reducing carbohydrate intake and for his funding of trials by independent investigators. Nevertheless, advocating unlimited servings of beef, sausage, and butter would not serve our overweight patients well. Other dietary strategies, such as replacing saturated fat with a combination of monounsaturated or polyunsaturated fats, will reduce LDL cholesterol levels, platelet aggregation, endothelial dysfunction, and insulin resistance (15). In addition, considerable evidence suggests that replacing red and processed meats with a combination of fish, nuts, legumes, and poultry would reduce the risks for colon cancer, prostate cancer, diabetes, and heart disease, even if total fat remains high. Also, eating several servings of whole grains high in fiber per day, which is possible while maintaining a relatively low total carbohydrate intake, has consistently been associated with lower risks for type 2 diabetes and coronary heart disease. Replacing refined carbohydrates with whole grains, vegetables, and some fruits such as apples will also reduce the spikes of glucose and insulin that provoke hunger and will provide additional micronutrients and fiber. In his last book, Atkins (16), too, had shifted considerably toward this healthier version of a low-carbohydrate diet.
In the context of these studies of the low-carbohydrate diet, what advice can we offer our patients who want to lose weight? Responses to weight loss interventions typically vary widely; for example, some persons in both groups of the study by Yancy and colleagues lost more than 20 kg, some had modest weight changes, and others lost nothing. These differences are probably due to a combination of genetic, environmental, and psychosocial factors. As an example of a genetic influence, the relation between specific types of dietary fat and body weight appears to depend on a specific polymorphism in the peroxisome proliferator-activated receptor-gamma gene (17). One implication of the large variation in response to diet is that the mean values from randomized trials apply to few individuals and that 'n of 1' experiments with specific diets are valid for any patient. Thus, we can encourage overweight patients to experiment with various methods for weight control, including reduced-carbohydrate diets, as long as they emphasize healthy sources of fat and protein and incorporate regular physical activity. Patients should focus on finding ways to eat that they can maintain indefinitely rather than seeking diets that promote rapid weight loss. For many patients, the roll will have little role.
- Walter C. Willett, MD, DrPH Harvard School of Public Health; Boston, MA 02115
REFERENCES:
1. Hunter DJ, Spiegelman D, Adami HO, Beeson L, van den Brandt PA, Folsom AR, et al. Cohort studies of fat intake and the risk of breast cancer -a pooled analysis. N Engl J Med. 1996;334:356-61. [PMID: 8538706] PubMed
2. U.S. National Research Council, Committee on Diet and Health. Diet and Health: Implications for Reducing Chronic Disease Risk. Washington, DC: National Academy Pr; 1989. PubMed
3. Hu FB, Stampfer MJ, Manson JE, Rimm E, Colditz GA, Rosner BA, et al. Dietary fat intake and the risk of coronary heart disease in women. N Engl J Med. 1997;337:1491-9. [PMID: 9366580] PubMed
4. Institute of Medicine. PubMed
5. Bray GA, Popkin BM. Dietary fat intake does affect obesity! Am J Clin Nutr. 1998;68:1157-73. [PMID: 9846842] PubMed
6. Willett WC. Dietary fat and obesity: an unconvincing relation [Editorial]. Am J Clin Nutr. 1998;68:1149-50. [PMID: 9846838] PubMed
7. Willett WC, Leibel RL. Dietary fat is not a major determinant of body fat. Am J Med. 2002;113 Suppl 9B:47S-59S. [PMID: 12566139] PubMed
8. Pirozzo S, Summerbell C, Cameron C, Glasziou P. Advice on low-fat diets for obesity. Cochrane Database Syst Rev. 2002:CD003640. [PMID: 12076496] PubMed
9. Brehm BJ, Seeley RJ, Daniels SR, D'Alessio DA. A randomized trial comparing a very low carbohydrate diet and a calorie-restricted low fat diet on body weight and cardiovascular risk factors in healthy women. J Clin Endocrinol Metab. 2003;88:1617-23. [PMID: 12679447] PubMed
10. Foster GD, Wyatt HR, Hill JO, McGuckin BG, Brill C, Mohammed BS, et al. A randomized trial of a low-carbohydrate diet for obesity. N Engl J Med. 2003;348:2082-90. [PMID: 12761365] PubMed
11. Yancy WS Jr, Olsen MK, Guyton JR, Bakst RP, Westman EC. A low-carbohydrate, ketogenic diet versus a low-fat diet to treat obesity and hyperlipidemia. A randomized, controlled trial. Ann Intern Med. 2004;140:69-77. PubMed
12. Stern L, Iqbal N, Seshadri P, Chicano KL, Daily DA, McGrory J, 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. PubMed
13. Samaha FF, Iqbal N, Seshadri P, Chicano KL, Daily DA, McGrory J, et al. A low-carbohydrate as compared with a low-fat diet in severe obesity. N Engl J Med. 2003;348:2074-81. [PMID: 12761364] PubMed
14. Ebbeling CB, Leidig MM, Sinclair KB, Hangen JP, Ludwig DS. A reduced-glycemic load diet in the treatment of adolescent obesity. Arch Pediatr Adolesc Med. 2003;157:773-9. [PMID: 12912783] PubMed
15. Hu FB, Willett WC. Optimal diets for prevention of coronary heart disease. JAMA. 2002;288:2569-78. [PMID: 12444864] PubMed
16. Atkins R. Atkins for Life. New York: St. Martin's Pr; 2003. PubMed
17. Memisoglu A, Hu FB, Hankinson SE, Manson JE, De Vivo I, Willett WC, et al. Interaction between a peroxisome proliferator-activated receptor gamma gene polymorphism and dietary fat intake in relation to body mass. Hum Mol Genet. 2003;12:2923-9. [PMID: 14506127] PubMed
|