By: Dr. Geoffrey Modest
A recent short-term study of overweight individuals did not show a benefit for a low glycemic index (low GI) diet (see JAMA. 2014;312(23):2531-2541). In this randomized cross-over trial, overweight adults were given 4 different complete diets including all of their meals, snacks, and beverages, each for 5 weeks, with data below for those completing a minimum of 2 such diet rotations. Details:
–163 adults enrolled, mean age 53, 52% female, 51% black, 40% non-hispanic white, BMI 32 and waist circumference 104cm, 26% hypertensive, 16% smokers
–Diets (all based on DASH diet — high in whole grains, fruits, veges, and lower in low-fat dairy, lean meats/fish/poultry and nuts/seeds/legumes). all with similar amounts of potassium (4gm) , calcium (1 gm), sodium (2.3 gm), and all were 2000 kcal/day:
–High GI (65% on the glucose scale, high carbohydrate (58% of the energy)
–Low GI (<45% on glucose scale), high carbohydrate
–High GI, low carbohydrate (40% of the energy)
–Low GI, low carbohydrate
–Results (primary outcomes of insulin sensitivity, systolic BP, HDL, LDL, triglycerides):
–High carbohydrate diets: those on low GI vs high GI haddecreased insulin sensitivity (-20%, p=0.002), increased LDL from 139 to 147 (6%, p<0.001), no change in HDL, triglycerides, or blood pressure
–Low carbohydrate diets: those on low GI vs high GI had decreased triglycerides from 91 to 86 (-5%, p=0.02) and no change in other measures
–Low GI low carb vs high GI high carb: decreased triglycerides from 111 to 86 (-23%, p<0.001). No change in other measures
–All of these diets, based on the DASH diet, led to decreases in LDL from baseline (from when people were eating their own diet) of 153 to 140 and systolic blood pressure from 132 to 123 mmHg.
Their conclusion is that the glycemic index of the foods “does not significantly improve cardiovascular risk factors or insulin resistance.” I am commenting on this very short-term study mostly because the last study published seems to have disproportionate weight in the minds of most of us… a few issues:
–This is a really short-term, though highly controlled study (done in research units of academic medical centers)
–Several, but not all studies of low GI diet, have found an increase in HDL and decrease in triglycerides (eg, see JAMA. 2007;297:2092-2102; or N Engl J Med 2008;359:229-41). this last study compared a calorie restricted low fat diet (1500 kcal/d for women, 1800 kcal/d for men) with calorie-restricted Mediterranean diet (same restrictions) vs. unrestricted low carb diet, finding the most weight loss over 2 years was in the calorie-unrestricted low carb diet. So, one interesting point is that those on low carb diets naturally eat less, perhaps related to the fact that, as opposed to carbs, fats do decrease appetite by their effect on the hypothalamic satiety center.
–A low carb vs low fat diet in 148 men and women at 12 months recently found a significant weight loss (3.5 kg), ratio of HDL/triglycerides, and increases in HDL (see Ann Intern Med. 2014;161:309-318)
–An interesting study found that low fat diets, as compared to very low carbohydrate and low GI diets (all isocaloric), led to decreases in resting and total energy expenditure – I.e., the low fat diet led to a compensatory decrease in energy expenditure, which could contribute to weight gain (see JAMA. 2012;307(24):2627-2634).
So, as a perspective on this study, there is reaffirmation that a DASH diet is healthier than the “usual diet”, with decreases in blood pressure and LDL. Although this very short-term study did not find additional value to the low GI diet, most other studies I’ve seen show lipid benefits even in nondiabetic patients. perhaps the issue with this study is having the underlying DASH diet (the first such study I have seen), though before making broad conclusions, it would be useful to have a confirmatory trial with longer diet durations. I also think that for many patients, eating a calorie-unrestricted low GI diet is easier than the comparable calorie-restricted low fat or Mediterranean diets, with at least as good results. This study, of note, does not deal with an important aspect of the low GI diet: in the prevention or treatment of diabetes.