*Antioxidants* **2019**, *8*, 302

The supplementation of the diet with pistachios for 9 months in prediabetic participants had a significant lowering effect in fasting plasma glucose, fasting plasma insulin, and HOMA-IR (*p* < 0.001 for all), but only borderline significance in decreasing Wc, BW, and BMI [57]. Also, a short 4 week trial with a pecan-rich diet showed significant reductions in serum insulin and HOMA-IR compared to the control diet (*p* < 0.05) [41].

Similar to our findings, two meta-analysis of RCTs pointed out that nut consumption was related to a significant decrease in BW, BMI, and Wc [62,63].

One reason for these outcomes can be the fact that the metabolizable energy content, or energy available to the body, of several nuts is less than predicted by the Atwater factors: 5% less for pistachios [64], 16% less for cashews [65], 21% less for walnuts [66], and as much as 32% less in the case of almonds [67]. Also, despite 10% higher energy intake from peanuts in another trial, a less than expected increase in BW compared to the control was found [68].

The meta-analysis of Viguiliouk et al. [69] established that tree nut intake significantly lowered fasting blood glucose (*p* = 0.03) and HbA1c (*p* = 0.0003) compared with control diets, but no significant effects were detected for fasting insulin and HOMA-IR. The very recent meta-analysis from Tindall et al. [70] observed that consumption of tree nuts or peanuts significantly decreased fasting insulin and HOMA-IR with no significant effect on fasting blood glucose or HbA1c.

Positive associations between nut consumption and cardiometabolic biomarkers were seen in two cross-sectional studies reported in the literature in the last ten years, although the strength of the association is weaker in these types of studies, which look at a single point in time, compared to longitudinal and intervention studies. A cross-sectional study conducted on an elderly population at high CV risk, showed that nut intake decreased both Wc and BMI (*p*-trend < 0.005 in both) [71]. In the study of Jaceldo-Siegl et al. [72] obesity was 11% less likely in the high peanut/low tree nut group (odds ratio, OR = 0.89; 95% CI: 0.53–1.48), 37% less likely in the high tree nut/high peanut group (OR = 0.63; 95% CI: 0.40–0.99), and 46% less likely in the high tree nut/low peanut group (OR = 0.54; 95% CI: 0.34–0.88) (*p*-trend = 0.006) compared to low tree nut/low peanut group. Moreover, in a large US population cross-sectional survey, walnut consumers had a significantly lower prevalence of diabetes with lower levels of fasting blood glucose and HbA1c [73].
