Background: Epidemiologic studies suggest that diets with a low glycemic index (GI) or glycemic load (GL) are associated with a decreased risk of developing metabolic syndrome. Intervention studies are inconsistent, however, possibly due to differences in macronutrient and fiber compositions of the experimental diets.

Objective: We tested side-by-side test foods with similar macronutrient and fiber compositions but with different sucrose-flour ratios or carbohydrate source to determine the effects of different GIs and GLs on metabolic risk markers in overweight subjects.

Design: Overweight men (n = 9) and women (n = 6) received in random order for 11 wk 4 test foods with an increased GI or a decreased GI (69 compared with 40, 86 compared with 48, 63 compared with 37, and 51 compared with 20, respectively). There was a GL difference of 32 units between the 2 interventions.

Results: At the end of the 11-wk intervention periods, the decreased GL test foods did not change fasting plasma glucose (mean +/- SD: 5.83 +/- 0.6 compared with 5.94 +/- 0.6 mmol/L) or insulin (8.3 +/- 2.8 compared with 9.8 +/- 5.1 mU/L) concentrations compared with increased GL test foods. Serum total cholesterol (5.56 +/- 0.90 compared with 5.76 +/- 1.04 mmol/L), LDL-cholesterol (3.57 +/- 0.72 compared with 3.68 +/- 0.80 mmol/L), HDL-cholesterol (1.21 +/- 0.38 compared with 1.24 +/- 0.37 mmol/L), and triacylglycerol (1.61 +/- 0.77 compared with 1.78 +/- 1.04 mmol/L) concentrations were also not significantly different for decreased and increased GL test foods, respectively. Finally, proinflammatory (high-sensitivity C-reactive protein, interleukin-6, tumor necrosis factor-alpha, monocyte chemoattractant protein 1) and prothrombotic (plasminogen activator inhibitor 1) markers were not affected. Glucose and lipids were also analyzed after 1 and 5 wk of intervention and were not affected by the intervention.

Conclusion: When incorporated into a habitual diet, consumption of test foods with a decreased GL does not ameliorate metabolic risk markers in overweight subjects.