Walnut intake, cognitive outcomes and risk factors: a systematic review and meta-analysis

Figures & data

Table 1. Study Eligibility Criteria^a.

Population	Adults $\ge$ 18 years
	Healthy
	At risk for increasing cognitive decline (e.g. those with baseline obesity, hyperlipidaemia, hypertension, diabetes, or metabolic syndrome)
Intervention	Walnut
	Walnut oil
	Walnut extract
Comparator	No walnut
	Lower dose of walnut
	Other nuts/foods
Outcome	Cognitive function
	Dementia
	Alzheimer’s disease and pathology
	Mild cognitive impairment
	Cerebrovascular disease (e.g. stroke)
	Brain imaging
	Mood (e.g. anxiety, depression)
	Glucose metabolism (e.g. HbA1c and HOMA-IR)
	Inflammation (e.g. hsCRP, IL-1β, IL-6 TNFɑ, E-selectin, sICAM-1, sVCAM-1)
Design	All designs except case reports
Duration of exposure (RCTs only)	Any duration: inflammatory markers and mood
	$\ge$1 week intervention: glucose outcomes
	$\ge$3 week intervention: all other outcomes

^aHbA1C: haemoglobin A1C; HOMA-IR: homeostatic model assessment-insulin resistance; hsCRP: high sensitivity c-reactive protein; IL: interleukin; RCT: randomized-controlled trial; sICAM-1: soluble intracellular adhesion molecule-1; sVCAM-1: soluble vascular cell adhesion molecule-1; TNFɑ: tumour necrosis factor-alpha; USA: United States of America.

Figure 1. Study flow diagram showing the number of abstracts identified (n = 522); abstracts not meeting criteria (n = 438); full-text articles retrieved (n = 84); full-text articles excluded after screening (n = 53); full-text articles added from grey literature search (n = 6); full-text articles suggested by key informants (n = 2); full-text articles meeting study eligibility criteria (n = 39); eligible randomized controlled trials (n = 32), and observational studies (n = 7).

Table 2. Study characteristics and key findings of RCTs reporting the effects of walnut on cognition-related outcomes^a.

Study (Year)	Country (funding); design	Walnut intervention	Control intervention^b	Baseline health (%)^c	Mean BMI (SD)	N Randomized (N analyzed)	Mean age (SD), years	% Male	Washout period	Total duration (exposure duration)	Cognition-related outcomes (measurement tool(s))	Key findings	ROB^e
Miller et al. (2018) [38]	England (Gov, Aca); RP	30 mL walnut oil in milkshake	Milkshake with no walnut	Healthy (100), Overweight (17)	22.5 (NR)	62 (59)	19.9 (2.5)	37.2	NA	100 mins (NA)^d	Mood (POMS)	No significant difference between walnut and control on POMS.	High
Pribis et al. (2012) [35]; (2016) [36]^f	USA (Ind); RCO	60 g/d walnuts in banana bread	Banana bread with no walnuts	Healthy (100), Overweight/ Obese (26.6)	22.9 (3.4)	64 (52)	20.7 (2.1)	NR	6 wks	22 wks (8 wks)	Non-verbal Reasoning (APM) Memory (WMS-III) Verbal Reasoning (WGCTA) Mood (POMS [TMD])	No significant differences on APM or WMS-III. WGCTA: walnut improved performance only on interference subtest vs. control (11.2%, p = .009, d = 0.567). Walnut improved TMD score in males only (−27.49%, p = .043, d = 0.708).	SC
Probst et al. (2012) [39]	Austria (NR); RCO	41.6 g/d walnuts in 300 g of 1% fat yogurt	300 g of 1% fat yogurt	Healthy (100), Obese (0)	23.1 (NR)	14 (14)	24 (NR)	100	1 wk	6 wks (15 mins)^d	Mood (MDMQ [fatigue, calmness, mood])	No significant difference between walnut and control on MDMQ.	High
Sala-Vila et al. (2020) [37]	USA, Spain (Gov, Ind, Aca); RP	30–60 g/d walnuts	No walnuts	T2D (9.6), HTN (52.2), HLD (54), Overweight/ Obese (NR)	27.3 (NR)	708 (657)	69.2 (NR)	33	NA	2 yrs (2 yrs)	Composite Scores for: Memory (RAVLT, ROCF) Language (semantic fluency, BNT) Perception (VOSP, WAIS-III) Frontal Function (TMT [A, B], FAS, SCWT, SDMT, Digit span [WAIS-III], CPT-III) Global Cognition (all tests above) Structural Brain MRI (T1-weighted) fMRI (during N-back test)	No significant differences between walnut and control on any composite score (adjusted mean change). In Barcelona subgroup, walnut improved global cognition (p = .016) and perception (p = .005) scores. No significant differences in Loma Linda subgroup. Structural MRI: no significant difference between walnut and control. fMRI: at 2 yrs vs. baseline, increased activation in control group, but not walnut group (suggests ↑ brain efficiency in walnut group).	High

^aAca: Academia; APM: Raven’s Advanced Progressive Matrices; BMI: Body Mass Index (kg/m²); BNT: Boston Naming Test; CPT-II: Conners Continuous Performance Test-II; FAS: Phonemic fluency test; fMRI: Functional MRI; Gov: Government; HLD, hyperlipidaemia (includes hypercholesterolemia); HTN: hypertension; Ind: industry; MDMQ; Multidimensional Mood State Questionnaire; MRI: Magnetic Resonance Imaging; N: number; NA: not applicable; NR: not reported; POMS: Profile of Mood States Questionnaire; RAVLT: Rey Auditory Verbal Learning Test; RCO: randomized cross-over; RCT: randomized-controlled trial; ROB: risk of bias; ROCF: Rey-Osterrieth Complex Figure; RP: randomized parallel; SC: some concerns; SCWT: Stroop Colour Word Test; SD: standard deviation; SDMT: Symbol Digit Modalities Test; T2D: type-II diabetes; TMD: Total Mood Disturbance Score; TMT: Trail Making Test; Part A and B; USA: United States of America; VOSP: Visual Object and Space Perception Battery; WAIS-III: Wechsler Adult Intelligence Scale; WGCTA: The Watson-Glaser Critical Thinking Appraisal; WMS-III: Wechsler Memory Scale III.

^bFor studies with multiple walnut-free intervention arms, the intervention most similar to the walnut intervention was chosen as the control group.

^c‘Healthy’: no chronic illness (e.g. T2D, MetS, cardiovascular disease, etc.); ‘Health condition (NR)’: unspecified proportion of population has condition.

^dIntervention was a single dose.

^eROB: Overall risk of bias assessment.

^fSame trial, but reported different outcomes.

Figure 2. Risk of bias assessment based on the Cochrane risk-of-bias tool for randomized trials (RoB 2) conducted for five publications reporting cognition and mood outcomes. SC: some concerns.

Table 3. Effects of walnut on cognition-related outcomes reported in five RCTs^a.

Study (Year)	Group comparisons	Cognition-related outcome	Estimate (95% CI)^b
Sala-Vila et al. (2020) [37]	Walnut vs. Control	Global Cognition	0.02 (–0.02, 0.06)
		Memory	0.00 (–0.08, 0.09)
		Language	−0.01 (–0.09, 0.07)
		Perception	0.04 (–0.06, 0.14)
		Frontal Function	0.01 (–0.04, 0.06)
Pribis et al. (2012) [35]	Walnut vs. Placebo	General intellectual capacity/non-verbal reasoning	0.20 (–0.80, 1.00)
		Verbal reasoning	1.60 (–2.20, 5.50)
		General memory	−1.30 (–3.90, 1.20)
		Working memory	−0.20 (–1.20, 0.80)
Pribis et al. (2016) [36]	Walnut vs. Placebo	Total Mood Disturbance (Combined Sexes)	−4.40 (–12.3, 3.50)
		Total Mood Disturbance (Males)	−11.0 (–33.13, 11.13)
		Total Mood Disturbance (Females)	0.10 (–15.42, 15.62)
Miller et al. (2018) [38]	Walnut vs. Control	Total Mood Disturbance	−4.99 (–10.75, 0.07)
Probst et al. (2018)[39]	Walnut vs. Control	Mean AUC Values of Mood	0.40 (NR)

^aAUC: area under the curve; CI: confidence interval; NR: not reported; RCT: randomized-controlled trial.

^bMean differences are reported, with the corresponding 95% CIs.

Table 4. Study characteristics and key findings of cross-sectional (n = 6) and prospective cohort studies (n = 1) reporting cognition-related outcomes^a.

Study (Year)	Country (Funding); Design	Study Name (Enrollment Years)	N Enrolled (N Analysed)	Baseline Health (%)^b	Mean BMI (SD)	Mean Age (SD), Years	% Male	Walnut Intake Amount(s)	Adjustments	Cognition-related Outcomes (Measurement Tool(s))	Key Findings	ROB^c
Arab et al. (2015) [30]	USA (Ind); Cross-sectional	NHANES III (1988–1994); NHANES (1999–2002)	20,050 (12,693)	Overweight (NR)	26.8 (NR)	56.4 (NR)	44.7	NRNC: 0 g/dTertiles for: WWHC: >0 to 5; >5 to 11; >11 g/d WWON: > 0 to 4; > 4 to 7.7; > 7.7 g/d	Age, sex, race, PA, SMK, ALC	NHANES III Age 20–59: All subtests of NES2: Visuomotor Speed (SRTT), Information Processing Speed Concentration, Motor Control (SDST), Learning and Recall (SDLT); NHANES III Age 60+: Attention and Verbal Memory (SRT); NHANES 60+: Attention, Visual Spatial Skills, Learning, Memory (DSST)	Higher walnut intake associated with better scores on: SRTT: p-trend <.001 SDST: p-trend = .003 SDLT: p-trend = .004 SRT: p-trend <.001 DSST: p-trend <.001	SC
Arab et al. (2019) [29]	USA (Ind); Cross-sectional	NHANES (1988–1994)	50,965 (26,656)	Overweight (NR)	26.8 (NR)	46 (NR)	49	NN, ON: 0 g/d WWON: 4 g/d WWHC: 24 g/d	Age, sex, race, BMI, SMK, ALC, annual household income, marital status	Depression (PHQ-9)	WWHC and WWON significantly associated with reduced depression scores vs. NN. For WWHC, least squared mean for total scores was 26% lower than NN (p < .0001) and 17% lower than ON (p = .01). Stronger association among women (32%, p < .0001) than men (21%, p = .05).	SC
Bishop et al. (2020) [31]	USA (Ind); Cross-sectional	HCNA and HRS (2012–2016)	3,632 (2,821)	Overweight (37.8), Obese (30.2)	NR	74.3 (NR)	42.9	No intake: 0 oz/d Low intake: 0.04 oz/d, Moderate intake: 0.49 oz/d	Age, sex, race, marital and retirement status, EDU, occupation, household income, BMI, PA, SMK, ALC, chronic conditions	Global Cognitive Function (TICS Total Score)	Walnut intake (low or moderate vs. none) was associated with better test scores in 2012, 2014, and 2016 (p < .001), but no association was seen between walnut consumption and cognitive change overtime.	Low
Guasch-Ferré et al. (2017) [20]	USA (Gov, Ind); Cross-sectional	NHS (1986–2012); NHS II (1991–2013); HPFS (1986–2012)	289,900 (210,863)	All cohorts: HLD (NR), HTN (NR), T2D (NR)	NR	NR	19.6	28 g serving; 0 vs. ≥ 1 servings/wk	Age, race, BMI, PA, SMK, vitamin and aspirin use, family history of T2D, cancer, MI, HTN, HLD, menopausal status, hormone and oral contraceptive use, diet (total energy, ALC, red or processed meat, fruit, vegetable intake)	Fatal and Non-Fatal Stroke (Medical Records)	Walnut consumption was associated with a 17% reduction in risk for stroke (HR: 0.83; 95% CI, 0.71–0.96)	SC
Liu et al. (2020) [33]	USA (Ind); Cross-sectional	NHS (1986–2012); NHS II (1991–2013); HPFS (1986–2012)	289,660 (192,655)	NHS: HTN (26.4), HLD (22.4); NHS II: HTN (7.1), HLD (15.8); HPFS: HTN (17.4), HLD (21.1)	25.4 (NR); 24.7 (NR); 25.4 (NR)	58.4 (NR); 41.3 (NR); 58.2 (NR)	17.7	28 g serving; per 0.5 serving/d	Age, race, sex, family history of CVD, calendar year in 4-year intervals, intake of nuts at each 4-year interval, SMK, ALC, menopausal status, hormone use, oral contraceptive use, PA, BMI, changes in energy intake, HLD, HTN	Fatal and Non-Fatal Stroke (Medical Records)	Per 0.5 servings/d, increased walnut intake during a 4-year interval was associated with a 20% lower risk for stroke (RR: 0.80, 95% CI, 0.67–0.95) in the subsequent 4-year interval.	SC
O’Brien et al. (2014) [34]	USA (Gov, Ind); Prospective cohort	NHS (1995–2001)	19,415 (15,467)	HTN (55.1), HLD (65.3), MI (5.9), T2D (10), Overweight (35), Obese (17.2)	NR	74.2 (NR)	0	28 g serving < 1/mo, 1–3/mo, ≥ 1/wk	Age, EDU, time between tests, BMI, anti-depressants, SMK, PA, total energy, ALC, vitamin intake, HTN, HLD, MI, T2D	Cognitive Function (TICS), Verbal Memory Score (average of EBMT [immediate, delayed recall], fluency test [animals], TICS [delayed recall of word list], DS-backward), Global Score (average of all 6 tests)	Significantly higher global cognitive and verbal memory scores in 1–3 servings/mo group vs. < 1 serving/mo group (global score: 0.03, 95% CI: 0.01–0.06; verbal memory score: 0.03, 95% CI: 0.00–0.06). No overall trend of increasing walnut intake with improved cognitive performance and no significant differences between intake groups in rates of cognitive decline over the six years of follow-up.	SC
Valls-Pedret et al. (2012) [32]	Spain (Gov); Cross-sectional	PREDIMED (2004–2009)	578 (447)	T2D (55.9), HLD (72.0), HTN (75.2)	28.5 (NR)	66.9 (NR)	47.9	Per 30 g/d walnut	Age, sex, EDU, BMI, SMK, ApoE e4 allele, PA, T2D, HTN, HLD	Global Cognitive Function (MMSE), Immediate and Delayed Episodic Verbal Memory (RAVLT), Episodic Memory of Performance (VPAT [WMS]), Semantic Fluency (AFT), Immediate Memory (DS-forward [WAIS]), Working Memory (DS-back [WAIS]), Executive Function (CTT-I, II)	Higher walnut intake was associated with better scores on digit span backward test of working memory (p = 0.039). Higher intake was not associated with better performance on other measures.	SC

^aAca: Academia; AFT: Animal Fluency Test; ALC: alcohol; BMI: Body Mass Index (kg/m²); CI: confidence interval (95%); CTT-I; II: Colour Trail Test (parts I and II); CVD: cardiovascular disease (includes coronary artery disease, peripheral vascular disease, congestive heart failure, or carotid stenosis); WWHC: walnuts with high certainty; DS-back: Digit Span Backward; DS-forward: Digit Span Forward; DSST: Digit-Symbol Substitution Test; EBMT: East Boston Memory Test (immediate and delayed recall); EDU: education; Gov: Government; HCNS: Health Care and Nutrition Study; HLD: hyperlipidaemia (includes hypercholesterolemia); HPFS: Health Professionals Follow-up Study; HR: hazard ratio; HRS: Health and Retirement Study; HTN: hypertension; Ind: Industry; MI: myocardial infarction; MMSE: Mini-Mental State Examination; N: number; NA: not applicable; NES2: Neurobehavioral Evaluation System 2; NHANES: National Health and Nutrition Examination Survey; NHS: Nurses’ Health Study; NN: no nuts; NR: not reported; ON: other nuts; oz: ounce; PA: physical activity; PHQ-9: Patient Health Questionnaire – 9; Major Depressive Disorder Module; PREDIMED: Prevención con Dieta Mediterránea; RAVLT: Rey Auditory Verbal Learning Test; ROB: risk of bias; RR: risk ratio; SC: some concerns; SD: standard deviation; SDLT: Serial Digit Learning Test; SDST: Symbol Digit Substitution; SMK: smoking; SRT: Story Recall Test; SRTT: Simple Reaction Time; T2D: type-II diabetes; TICS: Telephone Interview of Cognitive Status; USA: United States of America; VPAT: Verbal Paired Associates Test (WMS subtest); WAIS: Wechsler Adult Intelligence Scale; WMS,: Wechsler Memory Scale; WWON: walnuts with other nuts.

^bHealth condition (NR): unspecified proportion of population has condition.

^cROB: Overall risk of bias assessment.

Table 6. ROB assessment of included observational studies using the Newcastle-Ottawa score for cohort and cross-sectional studies^a.

Cross-sectional studies
Study (Year)	Selection (maximum: ★★★★★)	Comparability (maximum: ★★)	Outcome (maximum: ★★★)	Score (out of 10)	Overall ROB
Arab et al. (2015) [30]	★★★★	★★	★★	8	SC
Arab et al. (2019) [29]	★★★★	★	★★	7	SC
Bishop et al. (2020) [31]	★★★★	★★	★★★	9	Low
Guasch-Ferré et al. (2017) [20]	★★	★	★★★	6	SC
Liu et al. (2020) [33]	★★	★	★★★	6	SC
Valls-Pedret et al. (2012) [32]	★★★★	★★	★	7	SC
Prospective Cohort Study
Study (Year)	Selection (Maximum: ★★★★★)	Comparability (Maximum: ★★)	Outcome (Maximum: ★★★★)	Score (Out of 11)	Overall ROB
O’Brien et al. (2014) [34]	★★★	★★	★★	7	SC

ROB: risk of bias; SC: some concerns.

^aLower scores indicate increased risk; Cross-Sectional: 0–4 = High; 5–7 = SC; 8–10 = Low; Cohort: 0–4 = High; 5–7 = SC; 8–11 = Low.

Figure 3. Risk of bias assessment based on the Cochrane risk-of-bias tool for randomized trials (RoB 2) conducted for twenty-seven publications reporting glucose homeostasis and inflammatory outcomes [HbA1c, HOMA-IR, hsCRP, TNF-a, IL-6, IL-1$\mathrm{\beta },$ E-selectin, sVCAM, sICAM]. SC; some concerns.

Table 5. Study characteristics of RCTs reporting risk factors for cognitive-decline^a.

Study (Year)	Country (Funding)	Walnut Intervention	Control Intervention^b	RCT Design	Baseline Health (%)^c	Mean BMI (SD)	N Randomized (N Analysed)	Mean Age (SD), Years	% Male	Washout Period	Total Duration (Exposure Duration)	Risk Factor Outcome(s)	ROB^d
Aronis et al. (2012) [50]	USA (Aca)	48 g/d walnuts (liquid meal)	Isocaloric liquid meal	RCO	MetS (100): Obese (100), HTN (NR), HLD (NR)	36.6 (1.7)	15 (15)	58 (2.5)	60	1 mo	8 wks (4d)	hsCRP, sICAM-1, sVCAM-1, E-selectin, TNFɑ, IL-6	Low
Bamberger et al. (2017) [40]	Germany (Ind)	43 g/d shelled walnuts	Nut-free diet	RCO	Healthy (100), Overweight/Obese (NR), CVD (0)	25.1 (4.0)	204 (194)	63 (7)	30.9	4 wks	24 wks (16 wks)	HbA1C	High
Bhardwaj et al. (2018) [51]	USA (NR)	60 g shelled walnuts in single high-fat (62.3% energy) meal	77 g almonds in a single high-fat (62.3% energy) meal	RCO	Overweight (100), T2D (0), CVD (0), HTN (0)	28.6 (2.2)	27 (27)	42.3 (6.8)	40.7	2 wks	2 wks (4 hrs)	sVCAM-1	Low
Burns-Whitmore et al. (2014) [52]	USA (Gov, Ind)	28.4 g walnuts, 6x/wk	Standard egg, 6x/wk	RCO	Healthy (100)	23 (1.0)	26 (20)	38 (NR)	20	4 wks	32 wks (8 wks)	TNFɑ, E-selectin, sICAM-1, hsCRP, IL-6	High
Canales et al. (2011) [53]	Spain (Aca)	21.4 g/d walnuts in steak and sausage (walnut-enriched meat)	Low-fat meat without walnuts	RCO	Increased CVD risk: Overweight (60), Obese (40), HTN (NR), HLD (NR)	NR	25 (22)	54.8 (NR)	60	4–6 wks	14–16 wks (5 wks)	sVCAM-1, sICAM-1	SC
Chiang et al. (2012) [54]	USA (Ind)	42.5 g walnuts, 6x/wk	No walnuts or fish	RCO	HLD (NR)	24.8 (NR)	27 (25)	33 (NR)	56	1 d	12 wks (4 wks)	hsCRP, sICAM-1, sE-selectin, TNFɑ, IL-6, IL-1Β	High
Cortes et al. (2006) [55]	USA (Gov, Aca)	42 g/d walnuts in high-fat (63% energy) meal	25 g olive oil in high-fat (63% energy) meal	RCO	HLD (50), Overweight (NR), HTN (0)	25.5 (3.3)	24 (24)	38.5 (10.5)	83.3	1 wk	4 hrs	sICAM-1, sVCAM-1, E-selectin	Low
Damasceno et al. (2011)[56]	Spain (Ind)	40–65 g/d walnuts on Mediterranean-type diet	Isoenergenic Mediterranean-type diet with 35–50 g/d virgin olive oil	RCO	HLD (100)	25.7 (2.3)	20 (18)	56 (13)	50	None	12 wks (4 wks)	sICAM-1, sVCAM-1, hsCRP	SC
Fatahi et al. (2019) [57]	Iran (Gov)	9 walnuts + 150 g fatty fish/wk	300 g fatty fish/wk	RP	Overweight/Obese (100), T2D (0), HTN (0), CVD (0)	33.3 (5.6)	99 (99)	53.4 (1.6)	0	NA	12 wks (12 wks)	hsCRP, Il-6, TNFɑ	Low
Holscher et al. [58]	USA (Gov, Ind)	42 g/d walnuts	No walnuts (No active intervention)	RCO	Healthy (100)	28.8 (0.9)	18 (18)	53.1 (2.2)	55.6	1 wk	7 wks (3 wks)	sICAM-1, sVCAM-1	Low
Hwang et al. (2019) [41]	Korea (Ind)	45 g/d walnuts	Isocaloric white bread	RCO	MetS (100): Overweight/Obese (NR), HTN (NR), HLD (NR)	27.1 (2.6)	119 (84)	39.5 (14.5)	50	6 wks	38 wks (16 wks)	hsCRP, HbA1C	SC
Katz et al. (2012) [59]	USA (Ind)	56 g/d shelled, unroasted walnuts (walnut-enriched ad libitum diet)	Ad libitum diet without walnuts	RCO	MetS (100): Overweight/Obese (NR), HTN (NR), HLD (NR)	33.2 (4.4)	46 (40)	57.4 (11.9)	39.1	4 wks	24 wks (8 wks)	HOMA-IR	High
Ma et al. (2010)[42]	USA (NR)	56 g/d shelled, unroasted walnuts (walnut-enriched ad libitum diet)	Ad libitum diet without walnuts	RCO	T2D (100), Overweight/Obese (NR), HTN (NR), HLD (NR), CVD (0)	32.5 (5.0)	24 (21)	58 (9.2)	41.7	8 wks	28 wks (8 wks)	HbA1C, HOMA-IR	SC
Pieters et al. (2005) [60]; Mukuddem-Petersen et al. (2007)[61]	South Africa (Gov, Ind)	63–108 g/d walnuts (20% energy)	Nut-free diet	RP	MetS (100): Obese (91), HTN (NR), HLD (NR)	35.2 (NR)	68 (64)	45 (10)	45.3	NA	11 wks (8 wks)	HOMA-IR; hsCRP	Low
Nijke et al. [43]	USA (Ind)	56 g/d walnuts on calorie-adjusted diet	Calorie adjusted diet without walnuts	RP	Increased risk for T2D: Overweight (NR), HLD (NR), HTN (NR), T2D (0), CVD (0)	30.1 (4.1)	112 (97)	54.9 (11.4)	30.4	12 wks	15 mo (6 mo)	HbA1C	SC
Rock et al. (2016) [62]	USA (Gov, Ind)	42 g/d walnuts (higher fat [35% energy], lower CHO [45% energy] diet)	No walnuts (Higher fat [35% energy], lower CHO [45% energy] diet	RP	Overweight/Obese (100), Insulin resistant (51.4), T2D (0)	33.5 (NR)	245 (194)	50 (NR)	0	NA	1 yr (1 yr)	hsCRP, IL-6, HOMA-IR	SC
Ros et al. (2004)[63]	Spain (Gov, Ind)	40–65 g/d walnuts	Isoenergetic Mediterranean-type diet without walnuts	RCO	HLD (100)	NR	21 (20)	55 (NR)	40	None	8 wks (4 wks)	sICAM-1, sVCAM-1, hsCRP	Low
Tapsell et al. (2004) [46]	Australia (Ind)	30 g/d walnuts (low-fat/modified-fat diet)	No walnuts (low-fat/modified-fat diet)	RP	T2D (100)	29.2 (2.6)	58 (55)	59.3 (8.1)	56	NA	6 mo (6 mo)	HbA1C	SC
Tapsell et al. (2009) [45]	Australia (Ind)	30 g/d walnuts (low fat [30%] diet)	No walnuts (low fat [30%] diet)	RP	Overweight (100), T2D (100)	33.2 (4.2)	50 (35)	54 (8.7)	NR	NA	1 yr (1 yr)	HbA1C, HOMA-IR	Low
Tapsell et al. (2017) [47]	Australia (Ind, Aca)	30 g/d walnuts (+ Interdisciplinary Advice; IW)	No walnuts (Interdisciplinary Advice; I)	RP	MetS (34.9), Overweight/Obese (100), HTN (NR), HLD (NR)	32 (4.5)	377 (178)	44.3 (10.5)	26	NA	1 yr (1 yr)	HbA1C	Low
Tindall et al. (2019) [64]	USA (Gov, Aca, Ind)	57–99 g/d whole walnuts (WD)	Walnut fatty acid-matched diet (WFMD): same fatty acid composition as WD, but devoid of walnuts	RCO	Increased risk for CVD: Overweight/ Obese (100), HTN (NR), HLD (NR), T2D (0)	30 (1.0)	45 (36)	43 (10)	55.6	23 d	186 d (6 wks)	hsCRP	SC
Wu et al. (2010) [48]	China (Gov, Aca, Ind)	30 g/d walnuts (in isocaloric bread)	30 g/d flaxseed (in isocaloric bread)	RP	MetS (63.4), HTN (61.8), Overweight/Obese (NR), HLD (NR), CVD (0)	25.4 (2.5)	283 (277)	48.4 (NR)	55.8	NA	12 wks (12 wks)	HbA1C	Low
Wu et al. (2014) [44]	Germany (Ind)	43 g/d shelled walnuts	No walnuts (Isocaloric background diet)	RCO	Healthy (100), Obese (0), HLD (0), HTN (0), T2D (0)	24.9 (0.6)	57 (40)	60 (1)	25	2 wks	18 wks (8 wks)	HbA1C. HOMA-IR, sVCAM-1, sICAM-1	SC
Zhao et al. (2004, 2007) [65,66];	USA (Ind)	37 g/d walnut and 15 g/d walnut oil (ALA diet)	No walnuts (AAD diet)	RCO	HLD (100), Overweight/Obese (100), T2D (0)	28.1 (0.7)	23 (23)	49.8 (1.6)	86.9	3 wks	27 wks (18 wks)	sICAM-1, sVCAM-1, hsCRP, E-selectin; TNFɑ, IL-1β, IL-6	SC
Zibaeenezhad et al. (2016) [49]	Iran (Aca)	15 g/d walnut oil	No walnuts (No active intervention)	RP	T2D (100), Overweight/Obese (NR)	27.4 (2.4)	100 (100)	54.8 (11.1)	47.8	NA	3 mo (3 mo)	HbA1C	SC

^aAAD: Average American Diet; Aca: Academia; ALA: alpha-linoleic acid; BMI: Body Mass Index (kg/m²); CHO: carbohydrate; CVD: cardiovascular disease (includes coronary artery disease, myocardial infarction, peripheral vascular disease, congestive heart failure, or carotid stenosis); Gov: Government; HbA1C: haemoglobin A1C; HLD: hyperlipidaemia (includes hypercholesterolemia); HOMA-IR: homeostatic model assessment-insulin resistance; hsCRP: high sensitivity c-reactive protein; HTN: hypertension; IL: interleukin; Ind: Industry; MetS: metabolic syndrome (includes proportion of HTN, HLD, overweight/obese); N: number; NA: not applicable; NR: not reported; RCO: randomized cross-over; RCT: randomized-controlled trial; ROB: risk of bias; RP: randomized parallel; SC: some concerns; SD: standard deviation; sICAM-1: soluble intracellular adhesion molecule-1; sVCAM-1: soluble vascular cell adhesion molecule-1; T2D: type-II diabetes; TNFɑ: tumour necrosis factor-alpha; USA: United States of America.

^bFor studies with multiple walnut-free intervention arms, the intervention most similar to the walnut intervention was chosen as the control group.

^c‘Healthy’: no chronic illness (e.g. T2D, CVD, MetS, etc.); ‘Health condition (NR)’: unspecified proportion of population has condition.

^dROB: Overall risk of bias assessment.

Figure 4. (a) Effect of walnut intake on HbA1c, reported in ten RCTs with plausible data. (b) Effect of walnut intake on HOMA-IR, reported in 4 RCTs with plausible data. Weights are derived from random-effects analysis. Each grey box represents the individual study’s effect estimate, and the horizontal line represents the 95% CI of the effect estimate. The diamond shape represents the meta-analysis pooled effect estimate and its CI. A vertical line displays the location of the meta-analysis pooled effect estimate. n: number of participants; CI: confidence interval; ROB: risk of bias; SC: some concerns.

Figure 5. Effect of walnut intake on hsCRP, reported in nine RCTs with plausible data. Weights are derived from random-effects analysis. Each grey box represents the individual study’s effect estimate, and the horizontal line represents the 95% CI of the effect estimate. The diamond shape represents the meta-analysis pooled effect estimate and its CI. A vertical line displays the location of the meta-analysis pooled effect estimate. n: number of participants; CI: confidence interval; ROB: risk of bias; SC: some concerns.

Figure 6. (a) Effect of walnut intake on sICAM-1, reported in seven crossover RCTs with plausible data. (b) Effect of walnut intake on sVCAM-1, reported in five crossover RCTs with plausible data. Weights are derived from random-effects analysis. Each grey box represents the individual study’s effect estimate, and the horizontal line represents the 95% CI of the effect estimate. The diamond shape represents the meta-analysis pooled effect estimate and its CI. A vertical line displays the location of the meta-analysis pooled effect estimate. n: number of participants; CI: confidence interval; ROB: risk of bias; SC: some concerns.

Figure 7. (a) Effect of walnut intake on IL-6, reported in five RCTs with plausible data. (b) Effect of walnut intake on TNF-$\mathrm{\alpha },$ reported in four RCTs with plausible data. Weights are derived from random-effects analysis. Weights are derived from random-effects analysis. Each grey box represents the individual study’s effect estimate, and the horizontal line represents the 95% CI of the effect estimate. The diamond shape represents the meta-analysis pooled effect estimate and its CI. A vertical line displays the location of the meta-analysis pooled effect estimate. n: number of participants; CI: confidence interval; ROB: risk of bias; SC: some concerns.

Figure 8. Effect of walnut intake on E-selectin, reported in four RCTs with plausible data. Weights are derived from random-effects analysis. Each grey box represents the individual study’s effect estimate, and the horizontal line represents the 95% CI of the effect estimate. The diamond shape represents the meta-analysis pooled effect estimate and its CI. A vertical line displays the location of the meta-analysis pooled effect estimate. n: number of participants; CI: confidence interval; ROB: risk of bias; SC: some concerns.

Table 7. GRADE evidence profile table^a.

Quality assessment							Strength of Evidence^b	Summary & Justification
No of studies	Design (Ref.)	Limitations	Inconsistency	Indirectness	Imprecision	Dose-response
Cognition-related outcomes: Cognitive function, Mood, Stroke
5	RCTs [35–39]	Serious limitations: Overall ROB was high for 60% of trials and SC for 40% of trials. Studies were limited mostly due to concerns arising from the randomization process and deviations from intended interventions.	No serious inconsistency: No trials reported significant differences in mood and total cognitive scores between walnut and control groups. Significant differences were observed in population subgroups and/or subdomains only.	No serious indirectness: Clinical and/or validated cognition-related tests used.	Imprecise: All studies reported wide CIs or other measures of variance.	Not applicable: no within study comparisons of different walnut intake amounts.	⊕⊕OO LOW	None of the 5 RCTs found a significant effect of walnut on mood or cognitive function in complete study populations, though subanalyses and/or subdomains demonstrated a walnut effect. Due to concerns regarding risk of bias and imprecision in measures, we conclude that the SOE for the effects of walnut intake on cognition-related outcomes is low.
7	Observational [20,29–34]	Some limitations: Overall ROB was SC for 86% of trials and low for 14% of trials. Studies were limited mostly due to participant selection, self-reported walnut intake, and incomplete or selective reporting of results.	No serious inconsistency: Significant differences were reported between walnut and control groups for mood (100%) and total cognitive function scores (80%). The remaining study reported a significant effect of walnut on a subdomain of cognitive function.	No serious indirectness: Clinical and/or validated cognition-related tests used.	Some imprecision: 43% of studies reported wide CIs and/or small sample sizes.	Dose-response is present for all cognition- related outcomes.	⊕⊕OO LOW	There are insufficient data to support a hypothesis on the associations between walnut intake and mood or stroke, as ≤ 2 studies reported on these outcomes. However, 3 cross-sectional studies and 1 prospective cohort study found significant associations between walnut intake and cognitive function. Despite moderate imprecision and concerns of bias across these studies, the demonstration of a dose response effect upgraded the SOE for the association between walnut intake and cognition-related outcomes to a low rating.
Inflammation
17	RCTs [41,44, 50–58,61–66]	Some limitations: Overall ROB was low for 44%, SC for 44%, and high for 12% of trials.	No serious inconsistency: Across studies reporting inflammation markers, 60–100% did not find a significant effect of walnut compared to control.	No serious indirectness: Biomarkers of inflammation.	Some imprecision: Results were imprecise across studies, as indicated by moderate or high statistical heterogeneity of meta-analyses.	Not applicable; no within study comparisons of different walnut intake amounts.	⊕⊕⊕O MODERATE	The majority of studies (81%) reporting inflammation outcomes observed no effects of walnut. Additionally, meta-analyses found no significant effects of walnut on inflammation (hsCRP, VCAM, ICAM, TNFa, E-selection, IL-6). Due to concerns of bias and imprecision, the SOE for the effects of walnut intake on inflammation outcomes was rated as moderate.
Glucose homeostasis
13	RCTs [40–49,59, 60,62]	Some limitations: Overall ROB was low for 31%, SC for 54%, and high for 15% of the trials reporting glucose outcomes. Studies were limited mostly due to concerns arising from the randomization process, deviations from the intended intervention, missing outcome data, and selection of reported results.	No serious inconsistency: 70% of studies did not find a significant effect of walnut on HbA1c and none of the included studies found an effect of walnut on HOMA-IR.	No serious indirectness: HbA1c and HOMA-IR are validated measures of glucose homeostasis.	Imprecise: Results were imprecise across studies, as indicated by moderate to large statistical heterogeneity in meta-analyses. The 2 studies that could not be meta-analysed also raised concerns of imprecision, due to large measures of variance and lack of quantitative data.	Not applicable; no within study comparisons of different walnut intake amounts.	⊕⊕⊕O MODERATE	70% of RCTs found no significant effect of walnut on HbA1c and no effects were observed on HOMA-IR. The meta-analyses did not find an overall significant effect of walnut on HbA1c or HOMA-IR. Due to concerns of bias and imprecision, SOE for the effects of walnut intake on glucose outcomes was rated as moderate.

^aCI: confidence interval; GRADE: Grades of Recommendation; RCT: randomized-controlled trials; ROB: risk of bias; SC: some concerns; SOE: strength of evidence.

^bSymbols indicate the following strength of evidence: ⊕⊕⊕⊕, HIGH (further research is very unlikely to change our confidence in the estimate of association); ⊕⊕⊕O, MODERATE (further research is likely to have an important impact on our confidence in the estimate of association and may change the estimate); ⊕⊕OO, LOW (further research is very likely to have an important impact on our confidence in the estimate of association and is likely to change the estimate); and ⊕OOO, VERY LOW (any estimate of association is very uncertain).

Miller HC, Struyf D, Baptist P, et al. A mind cleared by walnut oil: the effects of polyunsaturated and saturated fat on extinction learning. Appetite. 2018;126:147–155.

 PubMed Web of Science ®Google Scholar

Pribis P, Bailey RN, Russell AA, et al. Effects of walnut consumption on cognitive performance in young adults. Br J Nutr. 2012;107(9):1393–1401.

 PubMed Web of Science ®Google Scholar

Pribis P. Effects of walnut consumption on mood in young adults-a randomized controlled trial. Nutrients. 2016;8(11):668.

 PubMed Web of Science ®Google Scholar

Probst A, Humpeler S, Heinzl H, et al. Short-term effect of macronutrient composition and glycemic index of a yoghurt breakfast on satiety and mood in healthy young men. Forsch Komplementmed. 2012;19(5):247–251.

 PubMedGoogle Scholar

Sala-Vila A, Valls-Pedret C, Rajaram S, et al. Effect of a 2-year diet intervention with walnuts on cognitive decline. The Walnuts and Healthy Aging (WAHA) study: a randomized controlled trial. Am J Clin Nutr. 2020;111(3):590–600.

 PubMed Web of Science ®Google Scholar

Arab L, Ang A. A cross sectional study of the association between walnut consumption and cognitive function among adult US populations represented in NHANES. J Nutr Health Aging. 2015;19(3):284–290.

 PubMed Web of Science ®Google Scholar

Arab L, Guo R, Elashoff D. Lower depression scores among walnut consumers in NHANES. Nutrients. 2019;11(2):275.

 PubMed Web of Science ®Google Scholar

Bishop NJ, Zuniga KE. Investigating walnut consumption and cognitive trajectories in a representative sample of older US adults. Public Health Nutr. 2020;24(7):1741–1752.

 PubMed Web of Science ®Google Scholar

Guasch-Ferre M, Liu X, Malik VS, et al. Nut consumption and risk of cardiovascular disease. J Am Coll Cardiol. 2017;70(20):2519–2532.

 PubMed Web of Science ®Google Scholar

Liu X, Guasch-Ferré M, Drouin-Chartier J-P, et al. Changes in nut consumption and subsequent cardiovascular disease risk among US men and women: 3 large prospective cohort studies. J Am Heart Assoc. 2020;9(7):e013877.

 PubMed Web of Science ®Google Scholar

O’Brien J, Okereke O, Devore E, et al. Long-term intake of nuts in relation to cognitive function in older women. J Nutr Health Aging. 2014;18(5):496–502.

 PubMed Web of Science ®Google Scholar

Valls-Pedret C, Lamuela-Raventós RM, Medina-Remón A, et al. Polyphenol-rich foods in the Mediterranean diet are associated with better cognitive function in elderly subjects at high cardiovascular risk. J Alzheimers Dis. 2012;29(4):773–782.

 PubMed Web of Science ®Google Scholar

Aronis KN, Vamvini MT, Chamberland JP, et al. Short-term walnut consumption increases circulating total adiponectin and apolipoprotein A concentrations, but does not affect markers of inflammation or vascular injury in obese humans with the metabolic syndrome: data from a double-blinded, randomized, placebo-controlled study. Metabolism. 2012;61(4):577–582.

 PubMed Web of Science ®Google Scholar

Bamberger C, Rossmeier A, Lechner K, et al. A walnut-enriched diet reduces lipids in healthy caucasian subjects, independent of recommended macronutrient replacement and time point of consumption: a prospective, randomized, controlled trial. Nutrients. 2017;9(10):1097.

 PubMed Web of Science ®Google Scholar

Bhardwaj R, Dod H, Sandhu MS, et al. Acute effects of diets rich in almonds and walnuts on endothelial function. Indian Heart J. 2018;70(4):497–501.

 PubMedGoogle Scholar

Burns-Whitmore B, Haddad E, Sabaté J, et al. Effects of supplementing n-3 fatty acid enriched eggs and walnuts on cardiovascular disease risk markers in healthy free-living lacto-ovo-vegetarians: a randomized, crossover, free-living intervention study. Nutr J. 2014;13(1):29.

 PubMed Web of Science ®Google Scholar

Canales A, Sánchez-Muniz FJ, Bastida S, et al. Effect of walnut-enriched meat on the relationship between VCAM, ICAM, and LTB4 levels and PON-1 activity in ApoA4 360 and PON-1 allele carriers at increased cardiovascular risk. Eur J Clin Nutr. 2011;65(6):703–710.

 PubMed Web of Science ®Google Scholar

Chiang Y-L, Haddad E, Rajaram S, et al. The effect of dietary walnuts compared to fatty fish on eicosanoids, cytokines, soluble endothelial adhesion molecules and lymphocyte subsets: a randomized, controlled crossover trial. Prostaglandins Leukot Essent Fatty Acids. 2012;87(4-5):111–117.

 PubMed Web of Science ®Google Scholar

Cortes B, Núñez I, Cofán M, et al. Acute effects of high-fat meals enriched with walnuts or olive oil on postprandial endothelial function. J Am Coll Cardiol. 2006;48(8):1666–1671.

 PubMed Web of Science ®Google Scholar

Damasceno NRT, Pérez-Heras A, Serra M, et al. Crossover study of diets enriched with virgin olive oil, walnuts or almonds. Effects on lipids and other cardiovascular risk markers. Nutr Metab Cardiovasc Dis. 2011;21(Suppl 1):S14–20.

 PubMed Web of Science ®Google Scholar

Fatahi S, Haghighatdoost F, Larijani B, et al. Effect of Weight Reduction Diets Containing Fish, Walnut or Fish plus Walnut on Cardiovascular Risk Factors in Overweight and Obese Women. Arch Iran Med. 2019;22(10):574–583.

 PubMed Web of Science ®Google Scholar

Holscher HD, Guetterman HM, Swanson KS, et al. Walnut consumption alters the gastrointestinal microbiota, microbially derived secondary bile acids, and health markers in healthy adults: a randomized controlled trial. J Nutr. 2018;148(6):861–867.

 PubMed Web of Science ®Google Scholar

Hwang H-J, Liu Y, Kim H-S, et al. Daily walnut intake improves metabolic syndrome status and increases circulating adiponectin levels: randomized controlled crossover trial. Nutr Res Pract. 2019;13(2):105–114.

 PubMed Web of Science ®Google Scholar

Katz DL, Davidhi A, Ma Y, et al. Effects of walnuts on endothelial function in overweight adults with visceral obesity: a randomized, controlled, crossover trial. J Am Coll Nutr. 2012;31(6):415–423.

 PubMed Web of Science ®Google Scholar

Ma Y, Njike VY, Millet J, et al. Effects of walnut consumption on endothelial function in type 2 diabetic subjects: a randomized controlled crossover trial. Diabetes Care. 2010;33(2):227–232.

 PubMed Web of Science ®Google Scholar

Pieters M, Oosthuizen W, Jerling JC, et al. Clustering of haemostatic variables and the effect of high cashew and walnut diets on these variables in metabolic syndrome patients. Blood Coagul Fibrinolysis. 2005;16(6):429–437.

 PubMed Web of Science ®Google Scholar

Mukuddem-Petersen J, Stonehouse (Oosthuizen) W, Jerling JC, et al. Effects of a high walnut and high cashew nut diet on selected markers of the metabolic syndrome: a controlled feeding trial. Br J Nutr. 2007;97(6):1144–1153.

 PubMed Web of Science ®Google Scholar

Njike VY, Ayettey R, Petraro P, et al. Walnut ingestion in adults at risk for diabetes: effects on body composition, diet quality, and cardiac risk measures. BMJ Open Diab Res Care. 2015;3(1):e000115.

 PubMed Web of Science ®Google Scholar

Rock CL, Flatt SW, Pakiz B, et al. Effects of diet composition on weight loss, metabolic factors and biomarkers in a 1-year weight loss intervention in obese women examined by baseline insulin resistance status. Metabolism. 2016;65(11):1605–1613.

 PubMed Web of Science ®Google Scholar

Ros E, Núñez I, Pérez-Heras A, et al. A walnut diet improves endothelial function in hypercholesterolemic subjects: a randomized crossover trial. Circulation. 2004;109(13):1609–1614.

 PubMed Web of Science ®Google Scholar

Tapsell LC, Gillen LJ, Patch CS, et al. Including walnuts in a low-fat/modified-fat diet improves HDL cholesterol-to-total cholesterol ratios in patients with type 2 diabetes. Diabetes Care. 2004;27(12):2777–2783.

 PubMed Web of Science ®Google Scholar

Tapsell LC, Batterham MJ, Teuss G, et al. Long-term effects of increased dietary polyunsaturated fat from walnuts on metabolic parameters in type II diabetes. Eur J Clin Nutr. 2009;63(8):1008–1015.

 PubMed Web of Science ®Google Scholar

Tapsell LC, Lonergan M, Batterham MJ, et al. Effect of interdisciplinary care on weight loss: a randomised controlled trial. BMJ Open. 2017;7(7):e014533.

 PubMed Web of Science ®Google Scholar

Tindall AM, Petersen KS, Skulas-Ray AC, et al. Replacing saturated fat with walnuts or vegetable oils improves central blood pressure and serum lipids in adults at risk for cardiovascular disease: a randomized controlled-feeding trial. J Am Heart Assoc. 2019;8(9):e011512.

 PubMed Web of Science ®Google Scholar

Wu H, Pan A, Yu Z, et al. Lifestyle counseling and supplementation with flaxseed or walnuts influence the management of metabolic syndrome. J Nutr. 2010;140(11):1937–1942.

 PubMed Web of Science ®Google Scholar

Wu L, Piotrowski K, Rau T, et al. Walnut-enriched diet reduces fasting non-HDL-cholesterol and apolipoprotein B in healthy Caucasian subjects: a randomized controlled cross-over clinical trial. Metabolism. 2014;63(3):382–391.

 PubMed Web of Science ®Google Scholar

Zhao G, Etherton TD, Martin KR, et al. Dietary alpha-linolenic acid reduces inflammatory and lipid cardiovascular risk factors in hypercholesterolemic men and women. J Nutr. 2004;134(11):2991–2997.

 PubMed Web of Science ®Google Scholar

Zhao G, Etherton TD, Martin KR, et al. Dietary alpha-linolenic acid inhibits proinflammatory cytokine production by peripheral blood mononuclear cells in hypercholesterolemic subjects. Am J Clin Nutr. 2007;85(2):385–391.

 PubMed Web of Science ®Google Scholar

Zibaeenezhad M, Aghasadeghi K, Hakimi H, et al. The effect of walnut oil consumption on blood sugar in patients with diabetes mellitus type 2. Int J Endocrinol Metab. 2016;14(3):e34889.

 PubMed Web of Science ®Google Scholar