Walnut Consumption and Cardiac Phenotypes: the Coronary Artery Risk Development in Young Adults (CARDIA) Study

Abstract

Background and Aims.

Observational studies and clinical trials have shown cardiovascular benefits of nut consumption, including walnuts. However, the relations of walnut consumption with systolic and diastolic function, risk factors for heart failure, are unknown. We examined the associations of walnut consumption with cardiac structure and function parameters in black and white adults enrolled in the Coronary Artery Risk Development in Young Adults (CARDIA) study.

Methods and Results.

After exclusions, the study population included 3,341 participants. Dietary intake was assessed using the CARDIA Diet History questionnaire at baseline, year 7 and year 20 exams. Cardiac structure and function were measured by echocardiography at year 25. Multivariable linear regression evaluated the associations of walnut consumption with blood pressure (BP), heart rate, and cardiac phenotypes, adjusting for age, sex, race, lifestyle habits, and clinical characteristics. We found the majority of walnut consumers compared to non-consumers were females, whites, and more highly educated, and had lower waist circumference, diastolic BP, and heart rate, and higher diet quality score. Even though cardiac structure and function measures were generally within normal ranges among participants, walnut consumers had significantly better values for diastolic function parameters A wave, E/A ratio, septal and lateral e’ than non-consumers. Further adjustment for body mass index and diabetes status did not materially change the significance between walnut consumer groups. Systolic function parameters did not differ by walnut group.

Conclusion.

Compared to non-consumers, walnut consumption is associated with better diastolic dysfunction in young to middle-aged adults.

INTRODUCTION

Risk factors for heart failure (HF) include high blood pressure, coronary heart disease (CHD), and systolic and diastolic dysfunction.¹ Dietary intake is an important modifiable risk factor in the development of cardiovascular disease (CVD) and CVD risk factors, such as high blood pressure.^2–4 Investigators of long-term prospective studies report lower risk of CVD, CHD mortality, and/or HF among those consuming daily intakes of total nuts or tree nuts, including walnuts, almonds, and Brazil nuts.^5–7 Results from numerous small clinical intervention studies and short-term randomized clinical trials (RCTs) demonstrate improved lipids and oxidative stress with walnut consumption compared to usual dietary intake without walnuts.^8–10 Similarly, long-term prospective studies of plant-based diet patterns that include nuts, such as walnuts, predict lower blood pressure,⁴ lower serum total cholesterol and low-density lipoprotein (LDL) cholesterol, greater endothelial function (i.e., flow mediated dilation),¹¹ and lower risk of developing type 2 diabetes.¹²

Because walnuts are rich in alpha-linolenic acid (ALA), melatonin, magnesium, and antioxidants, it is biologically plausible that walnut consumption would promote better heart structure and function. In 336 older white adults enrolled in the Hoorn study, higher serum ALA was related to higher left ventricular ejection fraction and lower left ventricular mass; notably, serum eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA) were not related to either cardiac measure.¹³ In the NHLBI Family Heart Study participants, dietary ALA intake was inversely related to lower odds of CAD, a known risk factor for HF.¹⁴ Similar beneficial findings were reported in a meta-analysis of 13 cohort studies examining dietary intake of ALA with risk of developing CVD, but not for serum or plasma ALA.¹⁵ Melatonin, a hormone and biomarker possessing anti-inflammatory and antioxidant properties, may protect against CVD.¹⁶ In a mouse model, improved diastolic function was demonstrated in mice fed a high-fat diet supplemented with melatonin.¹⁷ Further, melatonin supplementation improved ejection fraction in patients with HF,¹⁸ blood pressure in hypertensive individuals¹⁹ and other cardiac diseases induced by oxidative stress.¹⁶

These reports support the hypothesis that walnuts, rich in ALA, melatonin, and other bioactive compounds may potentially extend their cardioprotective effects to systolic and diastolic function.^13–19 However, few studies, if any, have examined the associations of walnut intake with heart structure and function. Therefore, investigation of these associations is warranted, especially in a biracial population of young adults followed prospectively over 25 years. Using data from the Coronary Artery Risk Development in Young Adults (CARDIA) study, we examined associations of walnut consumption over 20 years with cardiac phenotypes measured at Y25. We hypothesized that walnut consumption compared to non-consumption is related to lower left ventricular mass and a favorable left ventricular systolic function and diastolic function.

METHODS

The CARDIA study was designed to investigate the development of CVD risk factors and subclinical and clinical disease in young adults at baseline and currently has followed participants for over 30 years. All study protocols at CARDIA field centers were approved by each institution’s Institutional Review Board. After the study protocol was described to participants at each exam, consent forms were signed.

Study Population

At baseline, 5,115 black and white young women and men ages 18–30 years enrolled in the CARDIA study.²⁰ Study participants were recruited at one of four CARDIA field centers: Birmingham, AL; Chicago, IL; Minneapolis, MN; and Oakland, CA. Retention of participants by Y25 was 72% (n=3,449) of survivors. Exclusion criteria for this analysis include: missing heart structure and function measures (n=112); implausible energy intake: <600 and >6,000 kcal/day for women and <800 and >8,000 kcal/day for men (n=119); and participants with mitral valve prolapse or mitral regurgitation (n=33). The total number of participants included in the analyses is n=3,341.

Data Collection

Diet Assessment. The CARDIA Diet History²¹ was used to assess dietary intake by trained interviewers at baseline (Y0), year 7 (Y7) and year 20 (Y20). Information obtained included specific (and brand name) food and beverage intake, frequency (per day, per week, or per month) and estimated portion size using food models, measuring cups and spoons. The CARDIA nutrient and food group databases for the CARDIA Diet History at each diet data collection year were created using information from the Nutrient Data System for Research (NDSR) software program (Nutrition Coordinating Center, University of Minnesota; Minneapolis, MN). The food groups in the NDSR program are based on the USDA food grouping system, including 13 major food groups (fruit, vegetables, grains, legumes, meat, fish and seafood, dairy products, eggs, condiments, sugar, fats, nuts, and beverages) and 166 sub food groups. Using unique food and ingredient codes in NDSR, the ‘Walnut food group’, was created by extracting ‘walnuts’ (frequency and amount consumed) from the ‘Nuts’ food group, which included walnuts, almonds, pecans, pistachios, cashews, peanuts, other nuts, and peanut butter. One question on the CARDIA Diet History queried intake of NUTS AND SEEDS. Participants were asked if nuts or seeds were consumed; and if so, what kind of nuts or seeds, amount and frequency consumed; and whether they were salted or unsalted. Another question on the CARDIA Diet History queried ‘additions to foods’. For example, after asking participants about ice cream and frozen dessert intake, they were also asked if anything was added to their ice cream. This provided an opportunity to report (and record) additions, i.e., if they added ‘walnuts’ to their ice cream. Another opportunity to quantify ‘Walnuts’ would be if walnuts were an ingredient in another food consumed, such as breakfast cereal. A walnut consumer is defined as anyone who consumed walnuts included in food items or just walnuts alone. To validate walnut consumption, we found that walnut consumers had greater plasma ALA % than non-consumers (0.28% vs. 0.21%; p=0.01). To account for the rest of dietary intake in statistical models, the Healthy Eating Index-2015 (HEI-2015), a diet quality score that aligns with the 2015–20 US Dietary Guidelines for Americans (DGA), was created.²²

Echocardiography. At Y25, experienced sonographers performed Doppler echocardiography and 2-dimensional–guided M-mode echocardiography using an Artida cardiac ultrasound scanner (Toshiba Medical Systems, Otawara, Japan) using standardized protocols across all field centers. The sonographers made measurements from digitized images using a standard off-line image analysis software system (Digisonics, Inc., Houston, Texas). Left ventricular ejection fraction and left atrial volume were measured from the apical 4-chamber view on the basis of the American Society of Echocardiography (ASE) guidelines. Left ventricular mass was derived from the Devereux formula using the ASE guidelines.²³ Left atrial volume and left ventricular mass were indexed to body surface area (left atrial volume index and left ventricular mass index).²⁴ In the diastolic functional parameters, peak velocities of the early phase (E) and late phase (A) of the mitral inflow and their ratio (E/A ratio) were measured from pulsed Doppler echocardiography recordings of transmitral flow; early peak diastolic mitral annular velocity (e’) at the septal mitral annulus was measured using tissue Doppler imaging. E/e’ ratio was calculated as an index of left ventricular filling pressures. Strain was calculated based on speckle tracking as the peak systolic change in segment length relative to its end diastolic value. For this analysis of cardiac phenotypes structure and function, exclusion criteria include adults with mitral valve prolapse or mitral regurgitation (n=33).

Other measures. Interviewer-administered questionnaires were used to obtain information about demographic characteristics, medical history and medication use, smoking and alcohol habits, and physical activity. Height and weight were measured by wall-mounted stadiometer and calibrated balance-beam scale, respectively. Body mass index (BMI) was calculated as weight in kg/height in m². Using an anthropometric tape, waist circumference was measured (cm) midway between the bottom of the ribcage and the iliac crest. Abdominal obesity was defined as waist circumference greater than 88 cm for women and 102 cm for men. After a 5 minute rest in a quiet room, sitting blood pressure was measured three times using an Omron blood pressure device (Y20 and Y25; Omron Healthcare Inc., Lake Forest, Illinois). The average of the last two of three measures for each of systolic blood pressure (SBP) and diastolic blood pressure (DBP) was used. Pulse pressure (PP) was calculated as SBP minus DBP. Diabetes was defined as medication use for diabetes, fasting plasma glucose ≥126 mg/dl, 2-h glucose ≥200 mg/dl, or glycosylated hemoglobin ≥6.5%. Fasting plasma phospholipids (%), including ALA, was measured at Y20.²⁵

Statistical Methods

SAS, version 9.4 (SAS Institute Inc., Cary, NC) was used for all data analysis. Characteristics from year 20 were reported as means (standard errors, SEs) or frequencies (%) by walnut consumption (yes/no) and adjusted for age, sex, race, education, field center, and energy intake. Multivariable linear regression analyses (Proc GLM) evaluated the differences in dietary intake and cardiac phenotypes by walnut consumption (the average of walnut intake at years 0, 7, and 20). Models were adjusted for age, sex, race, field center, education, energy intake, physical activity, current smoking, drinking alcohol status, and HEI-2015 diet score (without walnuts). Systolic and diastolic function models were also adjusted for BMI and diabetes, in study of possible mediating factors. Statistical significance is represented as p-value ≤0.05.

RESULTS

Prevalence of walnut consumption at baseline was 2.5% and increased to 9% by Y20. Study participants were mean age 45 years for both walnut consumer and non-consumer groups at Y20 (Table 1). More whites and women reported eating walnuts than blacks or men; and walnut consumers had significantly lower DBP, waist circumference and prevalence of abdominal obesity as well as higher ALA (%) levels than non-consumers. Walnut consumers generally consumed a healthier diet than non-consumers as evidenced by higher average HEI-2015 diet quality score (Table 2). Compared to non-consumers, walnut consumers ate more nuts (in addition to walnuts), whole grain products, fruit, vegetables, and legumes and lower intakes of refined grain products, red and processed meat, candy/sugar/jams, and diet beverages.

Table 1.

Mean (SE) and proportion (SE) of year 20 characteristics by walnut consumption (Y0, 7, 20) among CARDIA participants, n=3341

Year 20 Characteristics	Walnut consumers n=340	Walnut nonconsumers n=3001	p-value
Age, years	45.222(0.213)	45.205(0.085)	0.939
Men, %	32.219(2.773)	43.844(1.111)	<0.001
White race, %	62.575(2.794)	46.911(1.121)	<0.001
Education, % >HS	80.960(2.353)	75.302(0.840)	0.023
Current smoking, %	15.471(2.146)	18.406(0.766)	0.194
Current drinking, %	23.353(2.292)	20.980(0.818)	0.327
Physical activity score	362.096(14.941)	341.553(5.321)	0.193
Body mass index, kg/m2	29.048(0.405)	29.675(0.145)	0.143
Obesity, %	35.238(2.732)	39.206(0.975)	0.169
Waist circumference, cm	90.886(0.829)	92.751(0.296)	0.033
Abdominal obesity, %	32.075(2.741)	38.847(0.979)	0.0193
SBP, mmHg	116.062(0.816)	117.229(0.291)	0.175
DBP, mmHg	71.904(0.624)	73.610(0.223)	0.009
Pulse pressure, mmHg	32.982(0.297)	33.294(0.106)	0.319
Hypertension, %	20.357(2.292)	22.202(0.818)	0.446
Diabetes, %	8.307(1.510)	7.115(0.539)	0.455
Plasma ALA, %	0.193(0.503)	0.182(0.181)	0.022

SE, standard error; HS, high school; SBP, systolic blood pressure; DBP, diastolic blood pressure; ALA, alpha linolenic acid

Linear regression models adjusted for age, sex, race, and field center, education and energy intake

Table 2.

Average (SE) nutrient and food intake over 20 Years (Y0, Y7, Y20) by walnut consumption in CARDIA participants, n=3341

Average dietary intake	Walnut consumers (n=340)	Walnut non-consumers (n=3001)	p-value
HEI-2015 score (without walnuts)	63.89 (0.53)	61.69 (0.19)	<0.001
Daily nutrient intake
Energy, kcal	2877.35 (54.44)	2787.39 (21.18)	0.105
Protein, % kcal	14.97 (0.12)	15.05 (0.04)	0.501
Total fat, % kcal	36.48 (0.27)	36.48 (0.09)	0.993
Saturated fat, % kcal	12.32 (0.12)	12.66 (0.04)	0.009
Monounsaturated fat, % kcal	13.65 (0.11)	13.64 (0.04)	0.891
Polyunsaturated fat, % kcal	7.74 (0.10)	7.39 (0.03)	<0.001
18:3, gm (GLA+ALA)	2.43 (0.06)	2.05 (0.02)	<0.001
Carbohydrate, % kcal	47.77 (0.34)	47.25 (0.12)	0.156
Added sugar, % kcal	11.46 (0.30)	12.43 (0.10)	0.002
Dietary fiber, gm	23.94 (0.51)	20.43 (0.14)	<0.001
Magnesium, mg	411.00 (6.11)	374.04 (2.18)	<0.001
Potassium, mg	2139.40(20.01)	2035(10.71)	0.001
Vitamin E, alpha tocopherol, mg	15.22 (0.33)	12.99 (0.12)	<0.001
Food group intake, serving/day
Walnuts, 1 oz.	0.69 (1.01)	0
Other nuts, 1 oz.	1.38 (0.10)	1.02 (0.04)	0.001
Whole grain products, 1 oz or slice	1.97 (0.06)	1.71 (0.02)	<0.001
Refined grain products, 1 oz or slice	3.34 (0.09)	3.63 (0.03)	0.004
Fruit, ½ cup or 1 medium piece	2.02 (0.07)	1.56 (0.03)	<0.001
Vegetables, ½ cup	4.84 (0.12)	4.11 (0.05)	<0.001
Dairy products, 1 cup or 1 oz	2.60 (0.09)	2.75 (0.03)	0.133
Legumes, ½ cup	0.31 (0.01)	0.22 (0.01)	<0.001
Red meat, 1 oz	1.93 (0.07)	2.22 (0.03)	<0.001
Processed meat, 1 oz	1.09 (0.05)	1.32 (0.02)	<0.001
Poultry, 1 oz	1.39 (0.05)	1.38 (0.02)	0.940
Fish, 1 oz	1.12 (0.05)	1.04 (0.02)	0.131
Eggs, 1 egg	0.60 (0.02)	0.60 (0.01)	0.959
Candy, sugar, honey, jam, 1 TB or oz	1.71 (0.11)	1.94 (0.040)	0.043
Sugar-sweetened beverages, 8 FO	2.91 (0.11)	2.83 (0.04)	0.444
Artificially sweetened beverages, 8 FO	0.48 (0.06)	0.66 (0.02)	0.010
Coffee/tea (unsweetened), 8 FO	1.79 (0.14)	1.82 (0.05)	0.836

SE, standard error; HEI-2015, Healthy Eating Index-2015; GLA+ALA, gamma linolenic acid plus alpha linolenic acid (these fatty acids are not differentiated in the nutrient database); oz=ounce; FO=fluid ounce; TB=tablespoon

Linear regression models adjusted for age, sex, race, education, field center, and energy intake

Results of the echocardiography parameters by walnut consumption are shown in Table 3 along with normal values of each parameter for age group 41–60 years (if applicable).²⁶ In our generally healthy population, we found that walnut consumers compared to non-consumers had significantly lower DBP and heart rate. There was no significant difference in left ventricular mass index between groups (p=0.478). Systolic function parameters ejection fraction and myocardial strain, including radial, longitudinal, and circumferential strain, were not different between two consumer groups. Mean values for diastolic function parameters were within the normal range for age for all adults in this analysis. Compared to non-consumers, mean A wave was significantly lower, E/A ratio and e’ were significantly higher; and the E/e’ ratio was marginally significantly lower among walnut consumers. Although not directly part of diastolic function, left atrial volume was marginally higher. Adjustment for BMI and diabetes status attenuated the association of walnut consumption (yes vs. no) with E/e’ ratio (p=0.152), but not with the other parameters (data not shown).

Table 3.

Components of mean (SE) systolic and diastolic function measured by echocardiography at year 25 by walnut consumption: CARDIA, n=3341

Y25 Echocardiography measures	Reference range	Walnut consumer (n=340)	Walnut nonconsumer, (n=3001)	p-value
Systolic blood pressure	≤120 mmHg	118.59 (0.88)	120.16 (0.32)	0.094
Diastolic blood pressure	≤80 mmHg	73.60 (0.61)	75.19 (0.22)	0.013
Pulse pressure (SBP – DBP)	45–55 mmHg	44.99 (0.51)	44.97 (0.18)	0.979
Heart rate, beats/min	60–100 beats/m	63.71 (0.62)	65.34 (0.22)	0.014
LVMI, g/m2	43–95 g/m2	49.80 (0.77)	50.38 (0.28)	0.478
Systolic function
LV Ejection fraction, %	>55%	69.72 (0.48)	69.36 (0.17)	0.469
Radial strain		38.43 (0.73)	37.46 (0.26)	0.217
Longitudinal strain		−15.12 (0.14)	−14.98 (0.05)	0.345
Circumferential strain		−15.37 (0.17)	−15.26 (0.06)	0.542
Diastolic function
E wave		78.54 (0.90)	78.26 (0.32)	0.767
A wave		61.47 (0.91)	63.58 (0.32)	0.024
E/A ratio	Age grp: reference 41–60: 0.78–1.78	1.36 (0.02)	1.29 (0.01)	0.002
Deceleration time, ms	41–60: 143–219	181.33 (2.30)	178.21 (0.82)	0.200
Septal e’	41–60: 7.6–16.8	9.53 (0.12)	9.18 (0.04)	0.007
Lateral e’	41–60: 11.5–20.7	12.05 (0.15)	11.66 (0.06)	0.014
e’		10.79 (0.12)	10.42 (0.04)	0.003
E/e’ ratio	40–59: 2.8–12.8	7.61 (0.12)	7.84 (0.05)	0.072
LAVI, mL/m2	16–28	25.02 (0.39)	24.22 (0.14)	0.052

LVMI, left ventricular mass index (left ventricular mass/body surface area); LV, left ventricular; LAVI, left atrial volume index (left atrial volume/body surface area)

Linear regression models adjusted for age, sex, race, education, field center, energy intake, drinking alcohol status, current smoking, physical activity, and HEI-2015 w/o walnuts

DISCUSSION

Walnut consumers compared to non-consumers had higher HEI-diet quality scores, represented by higher intakes of nutrient-dense plant foods and lower intakes of refined grain products, red meat, and processed meat. At Y25, walnut consumers had lower waist circumference and fewer were abdominally obese than non-consumers. Walnut consumers had significantly better diastolic function parameters than non-consumers. Moreover, the walnut consumers have a significantly lower DBP and heart rate as well as marginally lower SBP than non-consumers.

Higher heart rate and blood pressure are risk contributors to the development of diastolic dysfunction.²⁷ Echocardiographic markers such as E wave, A wave, and e’ are highly relevant, because they directly measure blood filling rate and mitral annular wall motion during diastole. These findings are very important in primary prevention of HF. Taken to an extreme, diastolic dysfunction manifests clinically as diastolic HF (HF with preserved ejection fraction). It is known that diastolic dysfunction is a major health-economic burden because 65% of the patients admitted in the hospital for HF are due to diastolic dysfunction.²⁸ Medical therapy to treat diastolic HF is limited and diastolic HF leads to frequent hospital re-admissions. These data show in an apparently healthy young population who consume walnuts a better diastolic function in comparison to those who do not consume walnuts. The significance of the differences in echocardiographic measures between walnut consumers and non-consumers is potentially important for early detection of changes in LV diastolic function.²⁹ Assessment of diastolic function is based on the profile of different parameters derived from echocardiographic measurement. Although the mean levels of the several diastolic function parameters are within the normal range, it is likely that the profile of less favorable values among non-consumers of walnuts than among walnut consumers represents a tendency towards subclinical pathology. Despite the LV diastolic function in both groups (walnut and non-walnut consumers) is normal, several parameters are significantly different.

Recent RCTs demonstrate lower blood pressure and improved lipids with daily walnut intake of 2–3 ounces over 6 weeks³⁰ and 1–2 ounces over 2 years. ³¹ In our prospective study, better diastolic blood pressure and function was associated with an average walnut intake of 0.70 ounce per day. Walnut consumption may improve these factors through individual or synergistic effects of its bioactive constituents including ALA, melatonin, potassium, magnesium, and polyphenols. In addition, the polyphenols in walnuts enhance the antioxidant effects of vitamin C, tocopherols, and B-sitosterol.³²

These bioactives may help in unraveling the beneficial effects of walnut consumption on the prevention of diastolic dysfunction.¹⁶ Walnuts are rich in ALA, which exerts beneficial effects on cardiovascular risk through a better lipid profile, anti-thrombotic and anti-inflammatory action, and vascular protection. ³³ Melatonin, a bioactive compound with free radical scavenger activity and antioxidant and anti-inflammatory functions,¹⁶ improved body composition through weight and body fat reduction (1 or 3 mg/night supplemental melatonin for 1 year in women);^34,35 CVD benefits via lipid lowering (5 mg/night supplemental melatonin for 3–4 weeks);³⁶ and blood pressure reduction (3 mg/night supplemental melatonin for 3 weeks in women; 2–5 mg/night for 3–4 weeks in men).^37–39 Even though both magnesium (Mg) and potassium (K) are widely distributed in plant foods, magnesium (Mg) and potassium (K) intakes are relatively low in American. The recommended intake for Mg is 310–420 mg/day and for potassium 3500–4700 mg/day. Lower intake of both minerals are associated with higher risk of hypertension. ^40,41 Finally, RCTs have demonstrated reduced blood pressure with polyphenol intakes that range from 35–1050 mg/day for up to 18 weeks.⁴²

Although melatonin and total polyphenols are not included in the CARDIA nutrient database, these compounds are contained in a variety of foods, particularly plant foods.^43,44 While fruit and vegetables are rich in potassium, commonly consumed foods rich in melatonin include nuts, legumes, whole grain products, fruit, vegetables, fish, and eggs.^43,44 In our study, walnut consumers compared to nonconsumers ate more daily servings of nuts, including walnuts, legumes, whole grain products, fruit, and vegetables; therefore, we expect that individuals who ate walnuts consumed more melatonin as well as polyphenols. Furthermore, dietary intakes of magnesium and potassium were higher in walnut consumers.

There are strengths and limitations in our study. Limitations of this study include the self-reported dietary intake that is subject to measurement error; nevertheless, the CARDIA Diet History method is among the most comprehensive and in-depth of diet assessment instruments. We expect that walnut reporting was accurate since walnuts are rich in ALA and walnut consumers had higher proportion of plasma phospholipid ALA compared to non-consumers. Given accurate walnut reporting, the differences between walnut consumers and non-consumers for cardiac structure and function parameters are likely unbiased.

The strengths of this study include the prospective study design and 25 years of follow-up, the cohort of biracial men and women, repeated measures of dietary intake assessed by the CARDIA diet history over 20 years, and measurement of cardiac phenotypes measured at Y25. Cardiac phenotypes were measured in a middle-age, generally healthy population, with the majority within the normal ranges of systolic and diastolic function parameters. Because it is likely that participants enrolled in short-term randomized clinical trials revert to their usual dietary habits after completion of intervention,⁴⁵ it may be difficult to assure that the effects of walnut intake remain constant after many years. However, observational studies with regular follow-up exams are able to monitor lifestyle, physical, and clinical factors over time. Thus, CARDIA is an ideal study population to investigate the long-term association of walnut consumption with the risk trajectory of the decline in diastolic function beyond changes in heart rate and blood pressure to study the development of new onset of diastolic HF.

These study findings are in line with the health claim to include walnuts as part of usual dietary intake.⁴⁶ In addition, our findings are consistent with the 2015–20 DGA recommendation to consume a variety of protein sources, including nuts in a healthy eating pattern.⁴⁷

Highlights.

• Walnut consumers had higher diet quality score than non-consumers

• Walnut intake was associated with better CVD risk factors

• Walnut consumers had better diastolic function parameters