Effects of Mixed Nut Consumption on Blood Glucose, Insulin, Satiety, and the Microbiome in a Healthy Population: A Pilot Study

Abstract

Nuts contain many health-promoting nutrients, fiber, and phytochemicals. Nut consumption has been reported to improve several chronic disease risk factors. Most studies to date have investigated single variety nut consumption. A nut mixture may offer a more diverse array of nutrients over single variety nuts. The primary outcome of this study was to examine the effects of mixed nut consumption on postprandial glucose, insulin, and satiety in healthy young adults. Exploratory outcomes include the effects of daily nut consumption on stool microbiome and bowel movement patterns. Twenty participants were randomized to consume either 42 g of mixed nuts or 46 g of potato chips daily for 3 weeks. Mixed nut consumption did not alter postprandial blood glucose and insulin, while potato chip consumption increased glucose and insulin (P < .05). There were no significant differences in fasting blood glucose or insulin for either snack after 3 weeks of daily consumption. Both snacks increased satiety while there were no significant differences in body weight, body fat, blood pressure, waist-to-hip ratio, or anxiety. After 3 weeks of snack consumption, both groups significantly reduced straining during bowel movements while the mixed nut group slightly increased stool amount. There were no significant changes in microbiome composition for either group; however, there was a nonsignificant trend toward increased Firmicutes to Bacteroidetes ratio in the potato chip group and an opposite trend in the mixed nut group. The results of this study suggest that mixed nuts are a healthy alternative for blood sugar control.

The study was registered at ClinicalTrials.gov, Number: NCT03375866.

INTRODUCTION

Nuts are an important component of many healthy dietary patterns, including the Mediterranean diet because they are rich in vitamins, minerals, unsaturated fatty acids, protein, fiber, and phytochemicals.^1,2 Nut consumption has been shown to lower the risk for chronic diseases such as heart disease and is inversely associated with total and cause-specific mortality.³ The 2012 Global Burden of Disease study similarly found that diets low in nuts and seeds are among the top dietary risk factors for disease burden and disability.⁴ In addition to physical health, there is increasing evidence that nuts may improve cognitive states by reducing stress and alleviating symptoms of depression.^5,6

Cohort studies examining nut consumption and type 2 diabetes (T2DM) risk have been mixed with some studies finding reduced risk and another study finding no association.^7–9 A meta-analysis of observational studies found no association between nut consumption and T2DM, although peanut butter was inversely related to T2DM incidence.¹⁰ Regarding glycemic control, a meta-analysis of randomized controlled trials assessing the effects of different tree nuts (including almonds, cashews, hazelnuts, pistachios, and walnuts) in those with diabetes found benefits for fasting glucose and glycosylated hemoglobin (HbA1c) with no advantages for fasting insulin and homeostatic model assessment for insulin resistance.¹¹ Mixed nuts contain a greater variety of nutrients than single variety nuts and warrant further investigation for their potential effects on glycemic control.

There is growing evidence of the importance of the microbiome and species diversity in promoting health and modulating chronic disease risk.¹² The intestinal microbiome refers to the bacteria, viruses, fungi, and protozoa that reside in the human intestinal tract.¹³ Nuts contain prebiotic substrates including fibers and polymerized polyphenols that can feed beneficial short-chain fatty acid producing gut bacteria.^14,15 Pistachio consumption decreases lactic acid bacteria and increases the number of beneficial butyrate-producing bacteria in the gut.¹⁶ Butyrate has been found to increase intestinal barrier function, improve insulin sensitivity, and has anti-inflammatory properties.¹⁷ Almond consumption increases the relative abundances of Lachnospira, Roseburia, and Dialister, while walnut consumption increases the relative abundances of Firmicutes species in butyrate-producing Clostridium clusters XIVa and IV including Faecalibacterium and Roseburia.^18,19

Most studies to date have examined the effects of individual types of nuts only. The phytochemical composition of nuts varies by type. Almonds, pecans, and pistachios are rich in proanthocyanidins, while walnuts and Brazil nuts have high phenolic acid content.² Therefore, a nut mixture may offer a greater variety of bioactive compounds with potential for greater synergistic effects than any single variety. The objective of this study was to assess the effects of mixed nut consumption on postprandial satiety, glucose, and insulin as well as the effects of 3 weeks of daily mixed nut consumption on fasting glucose and insulin, anxiety, bowel movement patterns, and gut microbiome changes.

MATERIALS AND METHODS

Study population

Inclusion criteria included healthy males and females between the ages of 20 and 35 years. Exclusion criteria included individuals with metabolic disorders or chronic inflammation; required medications, supplement, or antibiotic use; allergies to nuts; and pregnant women, which was verified with a brief medical history form. Based on a previous human trial of nut consumption on postprandial glucose, power analysis (G*Power, Germany) indicated that significant differences would be detected with a sample of 10 subjects per trial at 70% power and an alpha level of P < .05.²⁰ The experimental protocols and the process for obtaining informed consent were approved by San Diego State University’s Institutional Review Board (IRB) Committee (No. HS-2019-0094). The study was registered at ClinicalTrials.gov

Study design

A two-factor mixed design study was conducted with participants consuming either 42 g (250 kcal, 112 mg Na) of mixed nuts (Kirkland, Costco Wholesale Corp., Issaquah, WA, USA) or 46 g of isocaloric lightly salted potato chips (46 g, 250 kcal, 137 mg Na; Frito-Lay Inc., Plano, TX, USA). Potato chips were chosen for comparison because they are a common snack category in the United States based on market research.²¹ The nut mixture contained cashews, almonds, Brazil nuts, pecans, macadamia nuts, peanuts, walnuts, and pistachios. Participants were given prepackaged servings of the snacks to consume daily for 3 weeks. Participants were randomly assigned to one of the interventions in a block size of five until all participants were assigned.

All participants were instructed to maintain their usual diets and exercise habits. Both groups were instructed to not consume nuts or potato chips for the 7 days before the first visit and those in the potato chip group were asked not to consume nuts during the intervention period. Participants were instructed to do an overnight fast before each laboratory visit (baseline and 3 weeks).

Anthropometrics and blood collection

Anthropometric measurements were taken for height, weight, percent body fat (bioelectrical impedance; Omron Healthcare Inc., Bannockburn, IL, USA), and waist-to-hip ratio at baseline and week 3. Personnel were trained to take waist measurements 1 inch above the navel, and hip measurements were taken around the widest part of the buttocks region. If the value at follow-up differed greatly from baseline, measurements were taken again. Baseline and 3-week measurements for blood pressure were taken using a ReliON automatic BP monitor (Omron Healthcare Inc.). Blood pressure readings were taken after participants were in a rested and quiet state. Capillary blood was collected in the morning via finger pricks at baseline, 45 min postprandial, and at 3 weeks. Blood samples were centrifuged at 1200 g for 10 min at 4°C and stored at −80°C until analysis.

Dietary assessment and physical activity

Two 24-hour food logs were collected for the 2 days before the baseline and week 3 visit. The nutrition analysis program Cronometer (Cronometer, Revelstoke, British Columbia) was used to analyze energy and nutrient intakes of the diets. Participants were trained to enter data with generic foods and not by brand name ensuring that the data would be the laboratory-analyzed data from the United States Department of Agriculture database. Physical activity was measured at baseline and week 3 using a validated Physical Activity Recall Questionnaire.²²

Satiety and anxiety questionnaires

During the baseline visit, participants filled out a five-question 10 cm visual analog scale measuring satiety before eating their snack and at every 15-min increment through 75 min following snack consumption, which included 16 oz of water.²³ The satiety scale included the following questions: ‘‘How hungry are you?’’ ‘‘How full are you?’’ ‘‘How strong is your desire to eat?’’ How much do you think you could eat right now?’’ and ‘‘How thirsty are you?’’ Participants recorded their responses on a line with the left side indicating low and the right side indicating high feelings of hunger, fullness, desire to eat, and thirst. Responses were quantified in centimeters from the left side of the line to the participants’ mark.

Participants filled out a six-question State–Trait Anxiety Inventory (STAI) Scale at baseline, 45 min postprandial, and at 3 weeks to assess acute and long-term effects of nut consumption.²⁴ The STAI asked participants to rank on a 1–4 scale how much they agree with the following statements: I feel calm, tense, upset, relaxed, content, and worried. Each item has the responses of ‘‘not at all,’’ ‘‘somewhat,’’ ‘‘moderately,’’ and ‘‘very much,’’ which are assigned numerical values of 1–4 with higher scores indicating greater anxiety tendencies. These are then scored according to Spielberger’s manuals.

Biochemical analysis

Finger-stick blood glucose was measured using a glucose monitor (BD Biosciences, San Jose, CA, USA). Insulin was measured using an ultrasensitive insulin enzyme-linked immunosorbent assay according to the ALPCO protocol (ALPCO, Salem, NH, USA). Total antioxidant capacity (TAC) was determined from blood samples using the Cayman Anti-oxidant Assay Kit (Sigma–Aldrich, St. Louis, MO, USA).

Bowel movement patterns and microbiome composition

Participants filled out a 7-day bowel movement log for the week before the baseline and week 3 visit, which quantified participants’ frequency, amount, consistency, strain, pain, and overall feeling of constipation during bowel movements.²⁵ Stool samples were collected at baseline and week 3 in a DNA/RNA shield solution, and isolated DNA was sequenced using 16S rRNA gene sequencing at Laragen (Culver City, CA, USA).

Statistical analyses

Data were analyzed by mixed repeated analysis of variance (ANOVA) using SPSS software (IBM, Armonk, NY, USA). Bonferroni post hoc tests were used for glucose, insulin, satiety, and anxiety. For anthropometrics, blood pressure, diet analysis, and bowel movement questionnaire, paired t-tests were used to compare between pre- and post-snack consumption and independent t-tests to compare values between trials. All data are presented as mean – standard deviation and considered statistically significant when P < .05. The normality of the data was examined using the explore procedures in SPSS. When data showed a normal distribution, a mixed repeated ANOVA or t-test was performed. When the normality assumption was suspect, the Wilcoxon nonparametric analysis was also performed.

Microbiome data were analyzed using R (version 4.0.0) with the following packages: edgeR, phyloseq, and vegan. Alpha diversity index measures (Chao1 richness, Abundance-based Coverage Estimator richness, Shannon–Wiener diversity, and Simpson diversity) were analyzed using the nonparametric Kruskal–Wallis test. The Bray–Curtis dissimilarity index and the Jaccard index were used to assess beta diversity and were analyzed using permutational multivariate analysis of variance (PERMANOVA). Differential abundance of taxa as a result of treatment was analyzed using a negative binomial model. Data were filtered to remove taxa with <100 counts per million and were normalized using the trimmed mean of M-values method before analysis.

RESULTS

Demographics and baseline characteristics

Twenty-four participants were recruited to join the study. Two participants did not meet the inclusion criteria due to gastrointestinal and dental issues. Of the 22 participants who were enrolled in the study, 2 participants did not complete all the study requirements leaving 10 participants in each group who completed the study. There were eight females and two males in the potato chip group and nine females and one male in the mixed nuts group. Participants’ mean age was 24.4 years. The mean body mass index (BMI) was 23.3 kg/m² in the potato chip group and 24.8 kg/m² in the mixed nuts group. There were no significant differences between groups for age, height, body weight, or BMI.

Anthropometric measurements, blood pressure, diet, and physical activity levels

There were no significant differences for body weight, BMI, systolic blood pressure, diastolic blood pressure, waist circumference, hip circumference, waist-to-hip ratio, and body fat percentage after 3 weeks between or within groups (Table 1). Results revealed no significant differences in food intake between or within groups for all nutrients analyzed (Table 2) and for physical activity levels.

Table 1.

Anthropometric and Blood Pressure Measurements at Baseline and Week 3 in the Potato Chips and Mixed Nuts Groups

	Potato chips (n = 10)		Mixed nuts (n = 10)
Measurements	Baseline	Week 3	Baseline	Week 3
BMI (kg/m2)	23.3 ± 3.2	23.5 ± 3.6	24.8 ± 4.8	24.8 ± 4.8
Body fat (%)	22.4 ± 2.2	22.6 ± 2.3	23.7 ± 2.2	24.0 ± 2.3
SBP (mmHg)	105.6 ± 7.2	113.8 ± 21.8	109.1 ± 10.6	111.1 ± 12.9
DBP (mmHg)	71.1 ± 3.7	74.5 ± 7.8	73.6 ± 8.0	77.1 ± 15.2
W/H ratio	0.8 ± 0.1	0.8 ± 0.1	0.8 ± 0.1	0.8 ± 0.1

Data are presented as mean ± SD. There were no significant differences between or within groups. Paired t-tests were used to compare between pre- (baseline) and post-snack consumption (week 3) and independent t-tests to compare values between trials.

BMI, body mass index; DBP, diastolic blood pressure; SBP, systolic blood pressure; SD, standard deviation; W/H ratio, waist-to-hip ratio.

Table 2.

Energy and Nutrient Intakes at Baseline and Week 3 in the Potato Chips and Mixed Nuts Groups

	Potato chips (n = 10)		Mixed nuts (n = 10)
Nutrients	Baseline	Week 3	Baseline	Week 3
Energy (kcal/day)	1647 ± 563	1930 ± 622	1635 ± 361	1717 ± 413
Protein (g/day)	75.3 ± 7.4	74.3 ± 9.2	72.9 ± 7.4	72.5 ± 8.0
CHO (g/day)	212.0 ± 96.9	250.7 ± 93.5	201.5 ± 60.0	188.3 ± 43.6
Fat (g/day)	69.3 ± 47.8	81.9 ± 31.2	57.4 ± 14.5	72.2 ± 22.3
SFA (g/day)	16.0 ± 3.4	21.6 ± 3.4	19.6 ± 2.7	20.4 ± 3.1
MUFA (g/day)	15.5 ± 10.1	20.5 ± 10.6	15.9 ± 8.4	16.5 ± 12.9
PUFA (g/day)	12.5 ± 9.0	16.3 ± 10.4	8.4 ± 3.5	8.9 ± 6.0
n-3 FA (g/day)	1.9 ± 1.4	2.3 ± 2.0	1.2 ± 1.0	1.5 ± 1.5
n-6 FA (g/day)	10.0 ± 7.4	13.5 ± 8.7	6.3 ± 3.4	7.1 ± 5.4
n-6:n-3 ratio	5.9 ± 1.9	7.0 ± 2.8	7.3 ± 2.7	6.8 ± 3.3
Fiber (g/day)	22.9 ± 20.7	28.8 ± 22.8	22.5 ± 9.9	25.5 ± 10.9

Data are presented as mean ± SD. Data within rows with different superscript letters are statistically different at P < .05. Paired t-tests were used to compare between pre- (baseline) and post-snack consumption (week 3) and independent t-tests to compare values between trials.

CHO, carbohydrate; kcal, kilocalories; MUFA, monounsaturated fatty acids; n-3 FA, omega-3 fatty acids; n-6 FA, omega-6 fatty acids; PUFA, polyunsaturated fatty acids; SFA, saturated fatty acids.

Satiety and anxiety

Results indicated no significant between-group differences for satiety measures (Fig. 1). There was a significant effect for time with a decrease between 0 and 15 min for Q1 (hunger feeling), Q3 (desire to eat), Q4 (prospective food consumption), and Q5 (thirst feeling) in both groups (P < .01) but no significant difference between interventions at any time point. Both groups significantly increased feelings of fullness (Q2) from 0 to 15 min (P < .001).

FIG. 1.

Effects of potato chips and mixed nuts on satiety values as measured by a 10 cm visual analog satiety scale. (A) How hungry are you? (B) How full are you? (C) How strong is your desire to eat? (D) How much do you think you could eat right now? (E) How thirsty are you? Data are presented as mean - SD. *Significantly different from baseline in mixed nuts and potato chips intervention at P < .05. Mixed repeated ANOVA was used for statistical analysis, followed by Bonferroni post hoc tests. ANOVA, analysis of variance; SD, standard deviation.

STAI values for anxiety ratings were not significantly different between groups at baseline (13.10 ± 4.07 vs. 13.90 ± 4.31), 45 min postprandial (14.20 ± 4.92 vs. 12.70 ± 3.86), or at week 3 (14.10 ± 3.63 vs. 12.70 ± 3.77).

Glucose and insulin responses, and TAC

There was a significant increase in postprandial glucose (111.9 ± 9.31 vs. 131.3 ± 16.02 mg/dL; P = .009) and insulin (3.79 ± 2.15 vs. 10.69 ± 5.59 mIU/L; P = .005) from baseline to 45 min in the potato chips group, with no significant change in the mixed nuts group for postprandial glucose (102.2 ± 12.46 vs. 104.6 ± 8.55 mg/dL) and insulin (5.33 ± 2.25 vs. 5.02 ± 2.66 mIU/L; Fig. 2). There were no changes in fasting glucose and insulin following 3 weeks of daily snack consumption for either intervention.

FIG. 2.

Effects of potato chips and mixed nuts on (A) glucose and (B) insulin levels at baseline, 45 min postprandial, and week 3. Data are presented as mean ± SD. Bars with * are significantly different at P < .05 compared with baseline value. Mixed repeated ANOVA was used for statistical analysis, followed by Bonferroni post hoc tests.

There were no significant differences in TAC between groups at baseline (0.21 ± 0.15 vs. 0.27 ± 0.16 mM/L), 45 min postprandial (0.28 ± 0.16 vs. 0.21 ± 0.18 mM/L), and week 3 (0.25 ± 0.19 vs. 0.28 ± 0.19 mM/L).

Bowel movement patterns

There were no significant differences between groups for frequency, stool hardness, bowel pain, or feeling of constipation (Table 3). Stool amount per day slightly increased from baseline to week 3 (P = .013) in the mixed nuts group. There was a significant time effect for bowel movement straining as both groups decreased straining from baseline to week 3 (P < .05).

Table 3.

Bowel Movement Characteristics at Baseline and Week 3 in the Potato Chips and Mixed Nuts Groups

	Potato chips (n = 10)		Mixed nuts (n = 10)
Measurements	Baseline	Week 3	Baseline	Week 3
Stool frequency (stools per day)	2.4 ± 1.2	2.1 ± 0.9	1.6 ± 0.5	1.6 ± 0.5
Stool volume (cups per day)	1.2 ± 0.9a	1.3 ± 0.9a,b	1.1 ± 0.5a	1.5 ± 0.7b
Stool hardness*	3.0 ± 0.7	2.8 ± 0.8	3.3 ± 0.6	3.0 ± 0.4
Straining during BM*	2.7 ± 1.1a	2.2 ± 0.8b	2.1 ± 0.8a	1.8 ± 0.5b
Pain during BM*	1.8 ± 1.0	1.5 ± 0.6	1.4 ± 0.5	1.3 ± 0.4
Feeling of constipation during BM*	2.6 ± 2.0	1.8 ± 1.0	1.6 ± 0.7	1.4 ± 0.5

Data are presented as mean - SD. For stool frequency, ANOVA was performed using baseline as a covariate. Paired t-tests were used to compare between pre- (baseline) and post-snack consumption (week 3) and independent t-tests to compare values between trials. Data in rows with different superscript letters are statistically different at P < .05.

^*Stool hardness, straining, pain, and feeling of constipation are based on a 1–7 Likert scale with 1 representing very soft, no straining, no pain, and no feeling of constipation through 7 representing very hard, extreme straining, extreme pain, and feeling very constipated.

ANOVA, analysis of variance; BM, bowel movement.

Microbiome composition

There were no significant changes in alpha and beta diversity measures. However, a trend toward an increased Firmicutes to Bacteroidetes (F/B) ratio was observed after potato chip consumption and an opposite trend following mixed nut consumption (Fig. 3A). Although not statistically significant, the mixed nuts group had a 7.1 log fold decrease in Prevotella bivia (P = .058), a 3.8 log fold decrease in Prevotella disiens (P = .130), a 4.5 log fold decrease in Eubacterium dolichum (P = .185), and a 3.6 log fold increase in Bifidobacterium longum (P = .150) (Fig. 3B). The potato chip intervention led to an 11.9 log fold increase in Dialister invisus (P = .018), a 2.2 log fold increase in Blautia coccoides (P = .174), a 1.7 log fold increase in Faecalibacterium prausnitzii (P = .316), and increases in several Clostridium species (Fig. 3C).

FIG. 3.

(A) Phylum-level microbiome composition. The numbers above the bars are the Firmicutes to Bacteroidetes ratios expressed as mean ± SD. (B) Top taxa changes as a result of mixed nuts intervention for 3 weeks. (C) Top taxa changes as a result of potato chips intervention for 3 weeks. Nonparametric Kruskal–Wallis test was used to analyze the alpha diversity. PERMANOVA was used to analyze the beta diversity. PERMANOVA, permutational multivariate analysis of variance.

DISCUSSION

This study showed maintenance of postprandial glucose and insulin responses following mixed nut consumption, while potato chip consumption raised glucose and insulin. Previous studies similarly found almond and peanut consumption lowered postprandial glucose.^26–28 The lower carbohydrate content of nuts (9 g vs. 25 g) contributes to the lower glycemic index (34 vs. 73) and lower postprandial glucose responses of nuts compared with potato chips.^29,30 Nuts are a good source of healthy unsaturated fatty acids. Substituting carbohydrates with monounsaturated fatty acids (MUFAs) and especially polyunsaturated fatty acids (PUFAs) has been shown to improve glycemic control.³¹ Nuts contain a variety of polyphenols such as ellagitannins and proanthocyanidins in addition to soluble fiber, which all exert antidiabetic effects.³² Soluble fiber delays gastric emptying and reduces postprandial glucose and insulin responses.³³

There were no changes in fasting glucose and insulin levels after 3 weeks of daily nut consumption. Previous studies have produced mixed results on this topic. A study comparing a diet supplemented with 30 g/day of mixed nuts daily (walnuts, pine nuts, and roasted peanuts) versus no supplementation over 6 weeks found no changes in fasting glucose and insulin concentrations in women with metabolic syndrome.³⁴ An 8-week intervention of 42.5 g of mixed nuts compared with pretzels found reduced fasting insulin levels but no change in fasting glucose.³⁵ Research in patients with T2DM found 12 weeks of pistachio consumption (50 g/day) significantly reduced fasting glucose and HbA1c, while 12 weeks of almond consumption (60 g/day) reduced fasting glucose and insulin values.^36,37 The longer duration and larger dose of these studies may have contributed to the beneficial effects on glycemic control. Additionally, responses may be more pronounced in those already with T2DM compared with participants with normal glycemic functioning.

Nuts are a good source of unsaturated fatty acids, protein, and fiber, which can all contribute to slower gastric emptying and longer satiety. There are higher levels of fiber (2.9 g vs. 1.9 g), MUFA (12.0 g vs. 6.9 g), and PUFA (6.1 g vs. 5.7 g) in nuts compared with potato chips per provided snack. Despite higher levels of these nutrients in the provided snacks, the dietary intake data indicated a numerically greater increase in fiber, MUFA, and PUFA in the potato chip group, although not statistically significant. This may be due to the nonsignificant higher energy intake in the potato chip group or other foods eaten during the study period since only the provided snacks were controlled. Two previous studies using various forms of almonds and one study using peanut butter found increases in subjective satiety ratings following nut consumption.^26–28

However, these studies compared nuts with control meals that were not matched for energy density. A study examining various macronutrient compositions, energy densities, weight, volume, and sensory properties on hunger and food intake found energy content to be the primary determinant of hunger.³⁸ This may explain the current findings with no differences in appetite ratings because the snacks were matched for energy content. Another study comparing peanuts with potato crisps similarly found no differences in satiety ratings but did find a decrease in subsequent energy intake for those consuming peanuts when presented with a buffet meal.³⁹ Future studies should include objective measures of hunger and satiety including appetite hormones in addition to measuring subsequent energy intake. Moreover, our results and others have consistently found that nut consumption is not associated with weight gain and can be incorporated into a weight maintaining or weight loss diet.⁴⁰

Although there were no changes in anxiety scores following mixed nut consumption, longer studies may be warranted since nuts contain many neuroprotective compounds, such as vitamin E, folate, zinc, selenium, polyphenols, and omega-3 fatty acids, which may reduce the risk for cognitive decline and mood disorders.^41–43 Walnuts, which are particularly high in antioxidants, polyphenols, and a-linolenic acid, improved total mood disturbance scores in young healthy males and are associated with reduced depression scores compared with non-nut consumers. ^5,44

Our study found no change in 45 min postprandial TAC, although longer time points may be needed as Torabian et al. previously reported plasma TAC peaking at 150 min following almond and walnut consumption.⁴⁵ There were also no changes in TAC after 3 weeks of mixed nut consumption, which is similar to previous studies finding no change in plasma oxygen radical absorbance capacity and ferric reducing antioxidant power levels with walnut and cashew consumption.^46,47 Diets rich in a variety of antioxidants such as those in a nut mixture may offer a more diverse range of antioxidants and greater protection against oxidative stress. However, a larger serving of nuts and control over the antioxidant content of the rest of the diet may be needed to detect changes in antioxidant capacity.

The consumption of mixed nuts did not lead to a significant shift in microbiome composition. Our study showed a trend toward a decreased F/B ratio following consumption of mixed nuts with an opposite trend with potato chip consumption. Despite wide variations in species diversity between individuals, Bacteroidetes and Firmicutes phyla dominate >90% of the microbiome.⁴⁸ They reported that individuals with obesity had fewer Bacteroidetes and more Firmicutes compared with lean individuals and weight loss increased Bacteroidetes and decreased Firmicutes. Other studies have similarly shown that a greater relative abundance of Firmicutes is associated with higher risks of obesity, cardiovascular disease, and inflammation.^49,50

However, as Magne et al. pointed out, there are not enough convincing data to support the general associations of obesity with alterations in the F/B ratio as many factors including diet, physical activity, antibiotics, food additives, and contaminants can influence the microbiota.⁵¹ One study found Clostridium bolteae and Flavonifractor plautii to be common Firmicutes bacteria in lean individuals.⁵² Others have associated a decreased F/B ratio with inflammatory bowel disease.⁵³ Continuing to identify the specific species over just the F/B ratio will further our understanding of the role of the microbiome and human health. This is important as the microbiome can influence obesity and chronic disease through a variety of mechanisms including modulating energy extraction from food, metabolism, inflammation, and insulin resistance among many others.⁵⁴

There was a nonsignificant trend toward decreased P. bivia and P. disiens following the mixed nuts intervention. Increased abundance of Prevotella bacteria has been associated with inflammatory disorders. Specifically, Prevotella-rich dysbiosis in gut microbiome has been linked to insulin resistance, obesity, hypertension, and nonalcoholic fatty liver disease.⁵⁵ Mixed nut consumption also resulted in a decrease in E. dolichum, a species that has been associated with diet-induced obesity.⁵⁶ In contrast, an increase in B. longum and Bifidobacterium choerinum was observed following the mixed nuts intervention. B. longum has been used as a probiotic for its benefits in modulating luminal metabolism, stabilizing gut microbiome, and improving homeostatic balance within the host–microbiome interaction.⁵⁷ In addition to the increased bifidobacteria, consumption of mixed nuts resulted in an increase in Bacteroides nordii, a bacterium associated with healthy lipid profile and insulin sensitivity.⁵⁸

Participants were consuming less than the dietary recommendations for fruits and vegetables averaging one fruit and two cups of vegetables per day based on the 24-h recalls. Therefore, a small change such as adding nuts may be an effective dietary strategy to improve metabolic markers and microbiome composition.

A limitation to the current study is the use of several self-reported measures for dietary intake and bowel movement habits and thus should be interpreted with caution. This study also had a relatively small sample size with power calculations assessed for postprandial glucose responses, which may not give sufficient power for measures, such as stool microbiome. Future studies should also include a more diverse sample as the participants in the current study were mostly female and with no known medical conditions. Forty-two grams of nuts was chosen because it is a standard serving size for nuts and feasible for real-world implementation. However, larger servings of nuts may have a larger impact on variables such as TAC and microbiome composition. A full nutrient analysis of the nuts will better identify the specific nutrients and phytochemicals that a nut mixture provides. Standardization over the participants’ background diet with controlled feeding trials and diet lead-in periods would better isolate the effects of the provided snacks.

Nuts are a nutrient-rich food providing healthy fats, fiber, and phytochemicals. This study suggests that the consumption of a nut mixture may be a good low glycemic snack choice for maintaining steady postprandial blood sugars compared with other refined snack options. Trends toward favorable microbiome changes suggest that mixed nuts may also benefit gut health, although more research is needed.