It has been known for many years that foods that have the same carbohydrate (CHO) content can still have a very different impact on blood sugar (BS) levels after consumption. The relative increase in the BS level relative to consuming the same amount of CHO as a glucose solution is known as the glycemic index (GI). Multiplying the GI by the grams of CHO in a meal determines its glycemic load (GL). The scientific evidence linking GI and/or GL to changes in insulin sensitivity and/or cardiovascular (CVD) risk factors has long been complicated by a lack of adequate control of other dietary factors that can also impact insulin sensitivity and CVD risk factors. If the high-GI and low-GI diets differ significantly in salt, potassium, magnesium, fiber, saturated fat, and cholesterol content, then it may be those variables, rather than GI per se, that are the true causal factor.
Back in 2002, Dr. Pi-Sunyer published an excellent review of the scientific evidence linking GI to disease. He provided convincing arguments as to why it was still premature to recommend that people limit their intake of high-GI foods to prevent or treat any disease. Nevertheless, GI became a major focus of many fad diet books like the South Beach Diet, Zone Diet, and Sugar Busters. Even some nutrition researchers have been advocating the consumption of low-GI diets. Why all this enthusiasm for the unproven benefits being claimed for the metabolic impact of GI or GL on insulin sensitivity and CVD risk?
If you look at studies showing favorable effects a reduced GI diet in clinical trials or at data from population studies, it is clear that GI tends to increase as plant foods become more refined and processed. However, the refining and processing of plant foods often results in increased added salt and a marked reduction in fiber, along with reduction in nutrients such as potassium, magnesium, vitamins, and phytochemicals, which may themselves be largely responsible for whatever health benefits that are simply correlated with GI or GL. When health benefits due to other confounding variables is mistakenly attributed to GI or GL, and then clinicians use this data to make dietary recommendations, this can distract patients from what dietary factors they really need to be focusing on. Many studies that purported to show benefits to lowering GI or GL were complicated by changes in BMI. Absent the control of energy intake and other confounding variables, any correlations between improved CVD risk or insulin sensitivity being attributed to GI would be premature.
New GI Study Controls Confounding Variables: A randomized, crossover design study conducted by Dr. Frank Sacks and associates at Harvard Medical School examined the relative effects of diets composed of mostly high-GI or low-GI foods. Unlike many prior studies, the high-GI and low-GI diets in this new study did an excellent job of controlling for other dietary variables known or suspected to impact insulin sensitivity, blood pressure, and/or blood lipids. Importantly, calorie intake on all 4 experimental diets was similar. The high-GI (65) and low-GI (40) diets were each fed at a relatively high-carbohydrate (58% energy) fairly low-CHO (40% energy) level. Each of the 4 experimental diets were fed to 163 overweight and obese adult subjects for 5 weeks each. Most subjects (135) consumed all 4 experimental diets, and the rest followed 2-3 of the 4 experimental diets for 5 weeks. Each of the 4 experimental diets was based on a healthful Dietary Approaches to Stop Hypertension (DASH) diet. All 4 experimental diets had a similar low saturated fat, cholesterol, and salt content and had a similar but higher fiber, potassium, and magnesium content than a typical American diet. This study appeared in the December 17, 2014 issue of JAMA.
Dr. Sacks et. al. found that at the high dietary carbohydrate content, the low-GI (40) diet compared with high-GI (65) diet significantly decreased insulin sensitivity and raised low-density lipoprotein (LDL) cholesterol by 6%, but they found no significant changes in the levels of HDL-C, triglycerides, or blood pressure. On the low carbohydrate content (40% energy) diets, the high-GI level compared to low-GI level did not affect the outcomes, except for increasing serum triglycerides. Comparing the low-CHO (40% energy) diet (with more protein and unsaturated fat in place of mostly refined CHO) to the high-CHO (58% energy) diet, they did observe that the low-GI, low-CHO diet did not affect insulin sensitivity, systolic BP, LDL-C, or HDL-C, but the lower CHO (whether high or low GI) did significantly lower fasting serum triglycerides by about 20% and slightly lower diastolic BP. The significant drop in fasting TG levels and the modest drop in diastolic BP when unsaturated fat and protein displace some of refined CHO in a DASH-style diet had previously been observed in the Omni Heart Trial. This study showed that replacing refined CHOs with high-fiber plant foods such as nuts, seeds, avocado, and very lean animal protein such as egg whites, nonfat dairy, and very lean poultry or seafood and even modest amounts of unsaturated vegetable oils may help lower serum TGs and improve blood sugar control, particularly in people with type 2 diabetes mellitus (DM) with poor beta-cell function.
Bottom Line: Dr. Sacks noted, “In the context of an overall DASH-type diet, using glycemic index to select specific foods may not improve cardiovascular risk factors or insulin resistance.” Indeed, it is becoming increasingly clear that focusing primarily on GI or GL is unlikely to have any meaningful impact on the risk of developing CVD, insulin resistance, and/or type 2 DM.
For most people, a focus on GI seems misguided. It is far better to focus on reducing refined carbohydrates and consuming more minimally-processed plant foods, reducing added dietary salt, and limiting the intake of animal products, particularly those high in saturated fat and/or cholesterol than it is to limit healthy foods like potatoes or bananas simply because they have a high GI.
By James J. Kenney, PhD, FACN
Stephanie Ronco has been editing in a professional capacity for the past 10 years. In addition to her work as an editor, Ronco has also served as a ghostwriter and writing tutor. A voracious reader, Ronco loves watching language evolve and change. When she’s not delving into her latest project, Ronco can be found teaching acting classes, performing in community theater, or sailing with her husband.