clinicaltrials.gov Identifier: NCT00438425.

The apparently smaller LDL cholesterol (LDL-C)-lowering effect of soy in recent studies has prompted the U.S. FDA to reexamine the heart health claim previously allowed for soy products. We therefore attempted to estimate the intrinsic and extrinsic (displacement) potential of soy in reducing LDL-C to determine whether the heart health claim for soy continues to be justified. The intrinsic effect of soy was derived from a meta-analysis using soy studies (20-133 g/d soy protein) included in the recent AHA Soy Advisory. The extrinsic effect of soy in displacing foods higher in saturated fat and cholesterol was estimated using predictive equations for LDL-C and NHANES III population survey data with the substitution of 13-58 g/d soy protein for animal protein foods. The meta-analysis of the AHA Soy Advisory data gave a mean LDL-C reduction of 0.17 mmol/L (n = 22; P < 0.0001) or 4.3% for soy, which was confirmed in 11 studies reporting balanced macronutrient profiles. The estimated displacement value of soy (13-58 g/d) using NHANES III population survey data was a 3.6-6.0% reduction in LDL-C due to displacement of saturated fats and cholesterol from animal foods. The LDL-C reduction attributable to the combined intrinsic and extrinsic effects of soy protein foods ranged from 7.9 to 10.3%. Thus, soy remains one of a few food components that reduces serum cholesterol (>4%) when added to the diet.

OBJECTIVEFat intake, especially monounsaturated fatty acid (MUFA), has been liberalized in diabetic diets to preserve HDL cholesterol and improve glycemic control, yet the exact sources have not been clearly defined. Therefore, we assessed the effect of mixed nut consumption as a source of vegetable fat on serum lipids and HbA1c in type 2 diabetes.RESEARCH DESIGN AND METHODSA total of 117 type 2 diabetic subjects were randomized to one of three treatments for 3 months. Supplements were provided at 475 kcal per 2,000-kcal diet as mixed nuts (75 g/day), muffins, or half portions of both. The primary outcome was change in HbA1c.RESULTSThe relative increase in MUFAs was 8.7% energy on the full-nut dose compared with muffins. Using an intention-to-treat analysis (n = 117), full-nut dose (mean intake 73 g/day) reduced HbA1c (−0.21% absolute HbA1c units, 95% CI −0.30 to −0.11, P < 0.001) with no change after half-nut dose or muffin. Full-nut dose was significantly different from half-nut dose (P = 0.004) and muffin (P = 0.001), but no difference was seen between half-nut dose and muffins. LDL cholesterol also decreased significantly after full-nut dose compared with muffin. The LDL cholesterol reduction after half-nut dose was intermediate and not significantly different from the other treatments. Apolipoprotein (apo) B and the apoB:apoA1 ratio behaved similarly. Nut intake related negatively to changes in HbA1c (r = −0.20, P = 0.033) and LDL cholesterol (r = −0.24, P = 0.011).CONCLUSIONSTwo ounces of nuts daily as a replacement for carbohydrate foods improved both glycemic control and serum lipids in type 2 diabetes.

Aims/hypothesisSugar has been suggested to promote obesity, diabetes and coronary heart disease (CHD), yet fruit, despite containing sugars, may also have a low glycaemic index (GI) and all fruits are generally recommended for good health. We therefore assessed the effect of fruit with special emphasis on low GI fruit intake in type 2 diabetes.MethodsThis secondary analysis involved 152 type 2 diabetic participants treated with glucose-lowering agents who completed either 6 months of high fibre or low GI dietary advice, including fruit advice, in a parallel design.ResultsChange in low GI fruit intake ranged from −3.1 to 2.7 servings/day. The increase in low GI fruit intake significantly predicted reductions in HbA1c (r = −0.206, p = 0.011), systolic blood pressure (r = −0.183, p = 0.024) and CHD risk (r = −0.213, p = 0.008). Change in total fruit intake ranged from −3.7 to 3.2 servings/day and was not related to study outcomes. In a regression analysis including the eight major carbohydrate foods or classes of foods emphasised in the low GI diet, only low GI fruit and bread contributed independently and significantly to predicting change in HbA1c. Furthermore, comparing the highest with the lowest quartile of low GI fruit intake, the percentage change in HbA1c was reduced by −0.5% HbA1c units (95% CI 0.2–0.8 HbA1c units, p < 0.001).Conclusions/interpretationLow GI fruit consumption as part of a low GI diet was associated with lower HbA1c, blood pressure and CHD risk and supports a role for low GI fruit consumption in the management of type 2 diabetes.Trial registrationClinicalTrials.gov NCT00438698FundingThe Canadian Institutes of Health Research; Canada Research Chair Endowment of the Federal Government of Canada, Loblaw Companies, and Barilla (Italy).Electronic supplementary materialThe online version of this article (doi:10.1007/s00125-010-1927-1) contains supplementary material, which is available to authorised users.