With greater availability of energy-rich foods and the rising prevalence of obesity, type 2 diabetes and associated complications are increasing alarmingly. Currently afflicting 8% of our population and associated with annual expenditures of $159 billion/y in 2007, a 30% increase to 22.8 million Americans is expected by 2025.25
The global prevalence of type 2 diabetes is expected to rise to 329 million affected people by 2030. Type 2 diabetes is characterized by insulin resistance and reduced responsiveness of the pancreatic islet cells to glucose, ultimately leading to hyperglycemia and the development of clinical diabetes. In animal models, hyperglycemia contributes to insulin resistance and defects in insulin secretion.26–28
Thus, dietary factors that decrease plasma glucose and insulin demand could plausibly decrease the risk of type 2 diabetes.
Positive associations were obtained in seven of the eleven prospective epidemiologic studies 8–13, 29–32
that examined the relation between GI, and risk of type 2 diabetes (). The GL was also positively associated with diabetes24
() and this finding was confirmed based on 20 years of follow-up.14
Methodological difficulties might explain the four studies with null findings.29–32
A possible reason for the lack of association in the Iowa Women’s Study is that the diagnosis of diabetes was made only on self report without confirmation.29
Stevens et al. used an abbreviated food questionnaire that deliberately focused on dietary fat rather than carbohydrate.30
Sahyoun et al. included only 99 cases of diabetes.31
The Whitehall II study used dietary information collected from only one baseline questionnaire to relate to diabetes incidence in 13 years.32
Studies relating glycemic index to type 2 diabetes.
Studies relating glycemic load to type 2 diabetes.
In a recent meta-analyses summarizing studies of GI and GL in relation to risk of type 2 diabetes, Barclay calculated 40% and 27% higher summary RRs when comparing the highest with lowest quartiles of GI (95% CI: 1.23, 1.59; P < 0.0001) and for GL (95% CI: 1.12, 1.45; P < 0.0001), respectively.33
All the studies included in this meta- analysis were adjusted for fiber. Findings from the Black Women’s Health Study12
and the Shanghai Women’s Health Study13
provide valuable evidence that the adverse effects of GI and GL also apply to non-Caucasian ethnic groups. In summary, although not every study found positive associations between GI and GL and risk of type 2 diabetes the overall epidemiologic evidence strongly supports a positive relationship.
Although there have been no clinical trials to determine whether low-GI or low-GL diets can prevent diabetes, the effect of dietary CHO on comorbidities of diabetes has been investigated. Importantly, those who develop diabetes are unable to compensate for increased age-related insulin resistance by secreting more insulin.34
In normal subjects and subjects with impaired glucose tolerance,35, 36
as well as in patients with diabetes or CHD, low-GI diets limit reductions in insulin sensitivity 37–41
and reduce insulin secretion.42–45
Further, a meta-analysis of 14 randomized trials of people with diabetes indicated that glycated proteins (HbA1c
or fructosamine) were reduced 7.4% (95% CI, 8.8-6.0) more on a low-GI diet compared to a conventional diet with a higher GI after adjusting for baseline differences.46
In another meta-analysis of 11 relevant randomized controlled trials, the benefit of a low-GI diet on improving glycemic control in diabetes is further confirmed.47
The analysis also concluded that episodes of hypoglycemia were significantly fewer with low- compared to high-GI diets. In a recent trial among patients with type 2 diabetes, a low-GI diet, compared with a high-fiber control diet, improved HbA1c.48
However, neither a lower GI diet nor a lower CHO (higher mono-unsaturated fat) diet improved HbA1c in patients with near-normal HbA1c and only the lower-GI diet elicited sustained reductions in postprandial glucose and C-reactive protein.49
These data suggest that diabetics gain more salutary advantage than non-diabetics from low-GI diets and low-GI diets confer additional advantage compared to high-fiber diets.50
Acarbose is a drug which can delay CHO digestion by inhibiting intestinal glucosidase and thus mimic the effect of a low-GI diet. Use of acarbose, reduced diabetes risk in a randomized trial.51
This data also implies that an aspect of low-GI diets other than high fiber is related to their salutary advantage. Overall, available experimental evidence corroborates the findings from epidemiologic studies that low GI/GL are associated with reduced risk of type 2 diabetes.
Rats in which diabetes was modeled by partially pancreatectomy, feeding of a high- versus low-GI diet for 18 weeks showed decreased glucose tolerance and plasma adiponectin, doubled body fat, increased plasma triglyceride concentrations and exacerbated fibrosis of the pancreatic islets.52
Epidemiologic evidence regarding GI/GL and cardiovascular disease
Over 16 million people are affected by CHD, the cause of ~20% of deaths and associated costs of ~$165 billion in the US in 2006.53
Insulin resistance increases multiple risk factors for cardiovascular disease (CVD), a broad category of circulatory diseases which affect the heart and blood vessels. These risk factors include dyslipidemia, hypertension and hyperglycemia, and many of the same dietary factors that are related to enhanced risk for diabetes also appear to be related to higher risk of CHD.54
The relation between GI and GL and incidence of CVD has been examined in five prospective studies ( and ). GL has been related to a higher risk than GI. In a large female American cohort, a high dietary GL was associated with markedly increased risk of CHD, independent of conventional CHD risk factors over 10 years of follow-up 1
and a 90% increased risk after 20 years of follow-up ().2
In a Dutch female cohort, there was a 47% increased risk of CVD in women followed up for 9 years ().15
However, results from a much smaller study following 646 elderly Dutch men for 10 years did not corroborate these observations ().55
In a cohort of Italian men (n=15 171) and women (n= 32 578) originally recruited to the European Prospective Investigation into Cancer and Nutrition study, after a median of 7.9 years of follow-up, high dietary GL (RR=2.24; 95% CI: 1.26, 3.98) and high carbohydrate intake from high-GI foods (RR=1.68; 95% CI: 1.02, 2.75) increase the overall risk of CHD in women but not men.56
In a cohort of 36 246 Swedish men aged 45–79 y without diabetes or prior cardiovascular disease, after 8 years of follow-up dietary GI and dietary GL were not associated with ischemic cardiovascular disease or mortality, but dietary GL was associated with a greater risk of hemorrhagic stroke (RR = 1.44 comparing extreme quartiles; 95% CI: 0.91, 2.27; P for trend = 0.047). The authors concluded that the discrepancies between these findings and those of previous studies may be due to variations in the associations by sex or to differences in dietary contributions to GI and GL.23
Compared with other US cohorts, such as the NHS cohort, these Swedish men had much higher cereal fiber intake (6.2 g per 1000 kcal compared with ~2–3 g per 1000 kcal in the NHS) and physical activity (approximately twice as much time as the NHS). In a meta-analysis of CVD,33
there were 25% higher summary RRs for GI. Corroborating the epidemiologic data above, high-GI or GL diets were found to be strongly and inversely associated with risk factors for CHD, including reduced HDL levels 57–60
and increased insulin resistance, metabolic syndrome 61
and C-reactive protein.62
Another meta-analysis also lists foods with a high GI or GL as harmful factors.63
Studies relating glycemic index to cardiovascular disease.
Studies relating glycemic load to cardiovascular disease.
Intervention studies regarding GI/GL and metabolic risk factors for coronary heart disease
There have not been randomized trials in humans using CHD as the end point, however, evidence indicates that diets high in CHO can increase plasma levels of triglycerides and reduce HDL cholesterol, both of which are risk factors for CHD.64
Further, partial replacement of CHO with either protein or unsaturated fat improved CVD risk factors.65, 66
Three controlled intervention studies show that low-GI diets reduced levels of plasminogen-activator inhibitor-1 (PAI-1), a marker of thrombogenicity, in overweight 67
as well as diabetic adults.68, 69