DIRECT Study result: Low-carb, Mediterranean diets win weight-loss battle 19. July 2008 William Davis (14) Drs. Iris Shai and colleagues released results of a new Israeli study, the Dietary Intervention Randomized Controlled Trial (DIRECT) Trial, that compared three different diet strategies. Of those tested, a low-carbohydrate diet was most successful at achieving weight loss. You can find the full-text of the study on the New England Journal of Medicine website.322 participants followed one of three diets over two year period. Compared head-to-head, the (mean) weight loss in each group was:• 2.9 kg (6.4 lbs) for the low-fat group• 4.4 kg (9.7 lbs) for the Mediterranean-diet group• 4.7 kg (10.3 lbs) for the low-carbohydrate group(Average age 52 years at start; average body-mass index, or BMI, 31.)The conclusion was that the low-carb diet performed the best, with 60% greater weight loss, with the Mediterranean diet a close second. The dietsThe low-fat diet was based on the American Heart Association diet, with 30% of calories from fat (10% from saturated fat) and food choices weighted towards low-fat grains, vegetables, fruits, and legumes and limited additional fats, sweets, and high-fat snacks; calorie intake of 1500 kcal per day for women and 1800 kcal per day for men was encouraged. The Mediterranean diet was a moderate-fat diet rich in vegetables, with reduced red meat, and poultry and fish replacing beef and lamb. Total calories from fat of 35% per day or less was the goal, with most fat calories from olive oil and a handful of nuts. Like the low-fat program, calories were limited to 1500 kcal per day for women, 1800 kcal per day for men. The low-carbohydrate diet was patterned after the popular Atkins’ program, with 8% participants achieving the ketosis that Dr. Atkins’ advocated as evidence that a fat-burning metabolism was activated, rather than sugar-burning as fuel. For the 2-month “induction phase,” 20 grams of carbohydrates per day was set as the goal, followed by 120 grams per day once the weight goal was achieved. Unlike the other two diets, calories, protein and fat were unlimited. Weight loss, lipids, inflammationYou can see from the weight loss graph that the low-carb approach exerted the most dramatic initial weight loss. Interestingly, much of the weight-loss benefit was lost as the carbohydrate intake increased, by study design, back to 120 mg per day. However, the other two diet approaches showed similar phenomena of “giving back” some of the initial weight loss. The low-carbohydrate diet exerted the greatest change in cholesterol, or lipid, panels: increased HDL 8.4 mg/dl vs. 6.3 mg/dl on low-fat; the triglyceride response was the most dramatic, with a reduction of 23.7 mg/dl vs. 3.7 mg/dl on low-fat. Interestingly, the LDL cholesterol-reducing effect of all three diets was modest, with the most reduction achieved by the Mediteranean diet. The inflammatory measure, C-reactive protein (CRP), was reduced most effectively by the low-carb and Mediterranean diets, least by the low-fat diet. HbA1c, a measure of long-term blood sugar, dropped significantly more on the low-carb diet. When the final dietary composition was examined, interestingly, there really were only modest differences among the three diets, with 8% less calories from carbs, 8% greater calories from fat, comparing low-carb to low-fat, with Mediterranean intermediate. Taken at face value, this useful exercise quite clearly shows that, from the perspective of weight loss and correction of metabolic parameters like triglycerides, HDL,CRP, and blood sugar, low-carbohydrate wins hands down, with Mediterranean diet a close second. It also suggests that a return to a carbohydrate intake of 120 mg/day allows a partial return of initial weight lost, as well as deterioration of metabolic parameters after the initial positive changes.Although the study has already received some criticism for such potential flaws as the modest number of Atkins’ followers achieving ketosis (8%), suggesting lax adherence, and the reintroduction of the 120 mg/day carbohydrate advice, I can suspect that these may have been compromises drawn to satisfy some Institutional Review Board. (Whenever a study is going to be conducted involving human subjects, a study needs to pass through the review of an Institutional Review Board, or IRB. IRB’s, while charged to protect human subjects from experimental abuses, also tend to be painfully conservative and will block a study or demand changes even if they are not dangerous, but just veer too far off the mainstream.) However, several unanswered questions remain:1) How would the diets have compared if the carbohydrate restriction were continued for a longer period, or even indefinitely? (The divergences would likely have been dramatic.) 2) Will low-carb exert the same cardiovascular event reduction that the Mediterranean approach has shown in the Lyon study and others? 3) Are there effects on health outside of the measures followed that differ among the three diets, such as cancer? (I doubt it, especially given the modest real differences over time. But this will be the objection raised by various "official" organizations.)I would further propose that:Low-fat diets are deadThe AHA will cling to their version of low-fat diet, based on difficulty in changing course for any large, consensus-driven organization, not to mention the substantial ($100’s of millions) revenues derived from endorsing low-fat manufactured products. The AHA will also point to the lack of difference in LDL cholesterol among the three, since they cannot get beyond the fact that there’s more to coronary risk—a lot more—than LDL. Off-the-shelf diets achieve off-the-shelf resultsIf you just need a T-shirt, a medium might fit fine. But if you’d like a nicely fitting suit or dress, then tailoring to your individual proportions is needed. When aiming towards maximizing benefits on lipoproteins and coronary risk, none of these diets achieve the kinds of changes we often need for coronary plaque reversal, as in the Track Your Plaque program. That requires making dietary changes that exert maximal effects on lipoprotein patterns.