Carbs don't tell the whole story

According to a 2005 study, the OMNIheart study found a way to improve on the DASH diet -- cut the carbs. But you wouldn't know it from any of the DASH-related websites, which still recommend the same number of servings of grain per day. You might think that family doctors would start handing out prescriptions for OMNIheart instead of DASH to their patients with high blood pressure... But no. Maybe they're all on the South Beach diet now. Maybe the OMNIheart study was a waste of US taxpayer dollars.

According to the OmniHeart paper and the original DASH paper, the macronutrient profiles of the diets stacked up like this:

	Control	DASH	OMNI Carb	OMNI Protein	OMNI Fat	Barnard
Carbohydrates	48%	55%	58%	48%	48%	75%
Fat	37	27	27	27	37	10
Protein	15	18	15	25	15	15

DASH and OMNIheart diets were all markedly better for the heart than control, but the lower-carb Protein and Fat diets did significantly better. Protein was best. But wait! Fat looks like Control here. And it did better? Distribution of macronutrients must not tell the whole story. And then there's Neil Barnard's 2006 trial of a high-carb vegetarian diet that improved triglycerides much more than any of the OMNIheart diets. Of course, Barnard's trial started with diabetics with worse triglycerides, and ran three times as long as the OMNIheart trial. So comparing it to DASH and OMNIheart is a bit like comparing apples to oranges. Still -- the very-high-carbohydrate diet is very effective at improving metabolic syndrome risk factors. It clearly does not cause the blood sugar, inflammation, and cholesterol problems that low-carbers have warned us about.

How can this be? If cutting carbs is a good idea, how can increasing carbs be a great idea? Which side is wrong? Maybe neither side -- there's more to the story. Barnard's diet has a lot of whole grains. And by whole, I mean entire. Not degermed, defatted, polished, ground, cut, rolled, folded, spindled, or mutilated in any way. His diet is very high in low-glycemic-index foods. It is very low in fats, and exceptionally low in saturated fats. He had to cut out animal foods entirely to get there.

It has been well-known for some time that the glycemic index (GI) and glycemic load (GL) of foods can predict their effect on blood sugar, cholesterol, inflammation, and heart health. But I didn't see any of this discussion in any of the studies. The two things the Barnard and OMNI Fat diets have in common are better carbs and less saturated fat than the Control diet.

Wouldn't you like to see these two diets in a cage match? Or else, can someone who has access to all of the data analyze the GIs of the meals and GLs of the diets to look for the correlation?

1. Glycemic index in chronic disease: a reviewL S Augustin et al. European Journal of Clinical Nutrition (2002) 56, 1049-1071. doi:10.1038/sj.ejcn.1601454
2. Effects of protein, monounsaturated fat, and carbohydrate intake on blood pressure and serum lipids: results of the OmniHeart randomized trial. Appel LJ et al. JAMA. 2005 Nov 16;294(19):2455-64
3. A clinical trial of the effects of dietary patterns on blood pressure. DASH Collaborative Research Group.Appel LJ et al. N Engl J Med. 1997 Apr 17;336(16):1117-24.