Walnuts and pumpkin seeds: a few points off LDL, no blood pressure change

A 2026 meta-analysis of 49 randomized trials in 4,611 adults found that walnuts lowered LDL cholesterol by 5.68 mg/dL (95% CI -7.89 to -3.47), total cholesterol by 6.39 mg/dL and triglycerides by 6.24 mg/dL, with no overall change in HDL. A second 2026 meta-analysis, of 36 randomized trials, found that walnuts raised flow-mediated dilation, a measure of how well an artery widens, by 0.93 percentage points (95% CI 0.27 to 1.59). Blood pressure did not change overall. The reviews disagree on apolipoprotein B. An earlier meta-analysis of 25 trials in 2,155 people found a small fall, -0.06 (95% CI -0.10 to -0.01) in units its abstract does not state, while the 49-trial review saw a fall only at more than 50 g a day. The EU refused walnut sellers the general claim that walnuts support cardiovascular health. Pumpkin seeds are a different food, and we found no heart trial of them. Per 100 g raw they carry 29.9 g of protein and about 500 mg of magnesium, against 14.6 g and 142 mg in walnuts, and the US survey database counts both in 28 g ounces.

How much is actually in it

• Per 100 g, raw pumpkin seeds carry about twice the protein of walnuts and more than three times the magnesium, while walnuts carry about 49 g of polyunsaturated fat. — USDA Foundation Foods samples of raw pumpkin seeds (8) and raw English walnuts (8); CoFID pumpkin flesh (n = 8 + 8 samples) Evidence E source
  Pumpkin seeds: protein 29.9 g, magnesium 499.7 mg (468.5-511.0), iron 8.36 mg, zinc 6.34 mg (4.34-10.0); walnuts: protein 14.6 g, magnesium 142.4 mg, PUFA 49.3 g (45.55-51.49), fiber 5.21 g; pumpkin flesh (CoFID): 13 calories, carotene 450 ug

What a real portion looks like

• The US survey database counts an ounce of walnuts or shelled pumpkin seeds as 28 g, and a cup of cooked pumpkin as 230 g. — FNDDS portion weights Evidence E source
  Walnuts: 1 oz 28 g, 1 cup 120 g, 1 nut 2 g; pumpkin seeds unsalted: 1 oz without shell 28 g, 1 cup without shell 144 g; pumpkin cooked: 1 cup 230 g

What cooking and storage change

• Roasting nuts and seeds keeps all their magnesium and iron but loses about 15-20% of thiamin and folate by the USDA group retention table. — USDA retention factors for the group Nut and Seed Products, roasted Evidence E source
  Folate 80%, thiamin 85%, calcium, iron and magnesium 100% retained (Retn_Code 2203)

What it does to your health

• A meta-analysis of 25 randomized trials in 2,155 people found walnuts lowered apolipoprotein B slightly but did not change blood pressure. — 25 RCTs with 26 intervention arms, 2155 participants (n = 2155) Evidence A source
  ApoB WMD -0.06 (95% CI -0.10 to -0.01); SBP WMD -1.20 mmHg (-4.02 to 1.61), not significant
• A 2026 meta-analysis of 36 randomized trials found walnuts raised flow-mediated dilation by 0.93 percentage points without changing blood pressure overall. — adults in 36 RCTs, search to February 2026 (n = 36 RCTs) Evidence A source
  FMD +0.93% (95% CI 0.27 to 1.59); SBP -0.65 mmHg (-1.61 to 0.30) and DBP -0.68 mmHg (-1.56 to 0.19), not significant
• In a 5-week crossover trial in 25 people at high cardiovascular risk, meat products with added walnuts lowered total cholesterol by 6.8 mg/dl compared with the same products without walnuts. — 25 adults at high cardiovascular risk, meat products five times a week for five weeks (n = 25) Evidence B source
  Total cholesterol -6.8 mg/dl (95% CI -12.8 to -0.85) vs walnut-free product
• A 2026 meta-analysis of 49 randomized trials in 4,611 adults found walnuts lowered LDL cholesterol by about 5.7 mg/dL, total cholesterol by 6.4 and triglycerides by 6.2, with no overall change in HDL or apolipoprotein B. — adults in 49 RCTs (57 effect sizes), search to August 2025 (n = 4611) Evidence A source
  LDL-C -5.68 mg/dL (95% CI -7.89 to -3.47); TC -6.39 mg/dL (-8.54 to -4.23); TG -6.24 mg/dL (-9.49 to -2.99); HDL-C and Apo-B not significant overall

Safety, limits and interactions

• The FDA-approved levothyroxine label warns that walnuts may bind the drug and reduce its absorption. — people taking levothyroxine Evidence E source
  Possible reduced levothyroxine absorption

What people believe that the data does not support

• In an 8-week randomized crossover trial in 76 young adults, 1.4 oz (40 g) of walnuts a day at dinner shortened time to fall asleep and raised a urinary melatonin marker. — 76 young adults (mean 24.1 y, 85.5% women), 8 weeks each arm (n = 76) Evidence B source
  Evening urinary 6-SMT up (p = 0.029), sleep latency improved (p = 0.001)
• A meta-analysis of 27 studies in 1,121 adults found eating nuts, walnuts included, raised resting energy expenditure by 28.6 calories a day, a rise that was not significant. — 1121 adults in 27 studies of almonds, walnuts, peanuts and other nuts (n = 1121) Evidence A source
  Resting energy expenditure +28.6 calories/d (95% CI -10.7 to 67.8), not significant; energy compensation -280.5% to +176.4%
• The EU refused the general claim that walnuts support cardiovascular health. — EU Register on nutrition and health claims Evidence E source
  Non-authorized

Who this works differently for

• In 20 boys aged 2 to 7 in a Thai area with endemic bladder stones, pumpkin seed supplements lowered calcium oxalate crystals in urine compared with orthophosphate. — 20 boys aged 2-7, Ubol province, Thailand; pumpkin seeds 60 mg/kg body weight per day (n = 20) Evidence C source
  Lower calcium-oxalate crystal occurrence and urinary calcium; higher phosphorus, pyrophosphate and potassium
• Among 25 children with pumpkin seed allergy at a Turkish clinic, 64% had anaphylaxis, yet all of them tolerated pumpkin flesh. — 25 children with IgE-mediated pumpkin seed allergy, tertiary clinic, Turkey 2016-2024 (n = 25) Evidence D source
  Anaphylaxis 64%; co-allergy to tree nuts 76%; pumpkin pulp tolerated in all
• Challenge-confirmed tree nut allergy affects under 2% of people, and walnut is among the most common tree nut allergies in the USA and UK. — 36 prevalence studies, mostly children, Europe, UK and USA (n = 36 studies) Evidence C source
  Challenge-confirmed prevalence below 2%; probable 0.05-4.9%

The bottom line

In the 36-trial review, diastolic pressure fell only in trials that lasted 12 weeks or longer. In a 5-week crossover trial in 25 adults at high cardiovascular risk, meat products with walnuts mixed in lowered total cholesterol by 6.8 mg/dl (95% CI -12.8 to -0.85) against the same products without them. Those walnuts came inside restructured meat, and the diet phase was unblinded. Whether nuts raise the energy you burn is not settled. A meta-analysis of 27 studies in 1,121 adults, walnuts among the nuts tested, found resting energy use rose by 28.6 calories a day (95% CI -10.7 to 67.8), which did not reach significance. Pumpkin seeds were not among the nuts in it. An open-label crossover trial in 76 young adults, mostly women, gave 1.4 oz (40 g) of walnuts at dinner for 8 weeks. Time to fall asleep improved (p = 0.001) on self-reported scores, and the abstract gives no effect size. Pumpkin flesh is mostly water, 13 calories per 3.5 oz (100 g) in the UK table. In a 1987 study in 20 Thai boys aged 2 to 7 from an area with endemic bladder stones, pumpkin seed supplements of 60 mg per kg of body weight a day lowered calcium oxalate crystals in urine compared with orthophosphate. The periods ran in sequence with no randomization, and the abstract gives no size. Roasting nuts and seeds keeps all their magnesium and iron and about 80 to 85 percent of folate and thiamin by a USDA group factor from 1975. If you take levothyroxine, its approved label warns that walnuts may bind the drug and reduce its absorption, without saying by how much. Challenge-confirmed tree nut allergy affects under 2 percent of people, though only seven of 36 prevalence studies used a food challenge. Walnut is among the most common tree nut allergies in the USA and the UK. Pumpkin seed allergy is rare. In 25 children with it at one Turkish clinic, 64 percent had anaphylaxis and 76 percent were also allergic to tree nuts, yet every one of them tolerated pumpkin flesh.

Sources

1. pwal-r7-11 · USDA Foundation Foods and CoFID 2021, dataset
   Seeds, pumpkin seeds (pepitas), raw (Foundation, 8 samples): Protein 29.9079 G; Magnesium, Mg 499.7 MG range 468.5-511.0; Iron, Fe 8.363 MG range 7.84-9.42; Zinc, Zn 6.341 MG range 4.34-10.0; Total lipid (fat) 40.03 G. Nuts, walnuts, English, halves, raw (Foundation, 8 samples): Protein 14.5644 G; Magnesium, Mg 142.4 MG range 125.0-148.0; Fatty acids, total polyunsaturated 49.331 G range 45.55-51.49; Fiber, total dietary 5.21 G range 4.3-7.38. CoFID 13-326 Pumpkin, flesh only, raw: Energy 13 kcal; Carotene 450 µg; Potassium 130 mg
   USDA FoodData Central Foundation Foods, release 2026-04-30; CoFID 2021 code 13-326 · checked 2026-09-29
2. pwal-r7-12 · USDA FNDDS, dataset
   Walnuts, excluding honey roasted (FNDDS 42116000): 1 nut 2 g; 1 cup 120 g; 1 oz 28 g. Pumpkin seeds, unsalted (FNDDS 43101150): 1 cup, without shell 144 g; 1 oz, without shell 28 g; 1 oz, in shell, yields 15 g. Pumpkin, cooked (FNDDS 73201020): 1 cup 230 g; 1 cup, mashed 250 g
   USDA FNDDS 2021-2023, food codes 42116000, 43101150 and 73201020 · checked 2026-09-29
3. pwal-r7-13 · USDA Table of Nutrient Retention Factors, dataset
   USDA-RETN-R6:2203 NUTS,ROASTED, group Nut and Seed Products: phosphorus_mg 80% retained, vitamin_c_mg 80%, folate_total_ug 80%, thiamin_mg 85%, calcium_mg 100%, iron_mg 100%, magnesium_mg 100%
   USDA Table of Nutrient Retention Factors, Release 6, 2007, Retn_Code 2203 · checked 2026-09-29
4. pwal-r7-01 · meta-analysis of randomized trials
   Twenty-five RCTs comprising 26 intervention arms with 2155 patients were included. Walnut significantly decreased ApoB (WMD = -0.06; 95% CI: -0.10, -0.01, p = 0.002), but did not affect ApoA1 (WMD = -0.50; 95% CI: -1.34, 0.33, p = 0.249), systolic blood pressure (SBP) (WMD = -1.20; 95% CI: -4.02, 1.61, p = 0.401) and diastolic blood pressure (DBP) (WMD = -0.44; 95% CI: -2.55, 1.67, p = 0.682).
   Endocrinol Diabetes Metab, 2026 · checked 2026-09-29
5. pwal-r7-02 · meta-analysis of randomized trials
   Pooled analysis data from 36 RCTs showed no significant overall effect of walnut consumption on SBP (WMD: -0.65 mmHg; 95% CI: -1.61 to 0.30, p = 0.183) or DBP (WMD: -0.68 mmHg; 95% CI: -1.56 to 0.19, p = 0.125). ... Walnut intake significantly increased FMD (WMD: 0.93%; 95% CI: 0.27 to 1.59, p = 0.006)
   J Hum Nutr Diet, 2026 · checked 2026-09-29
6. pwal-r7-03 · randomized controlled trial
   The regular consumption of walnut-enriched meat products compared with that of the restructured meat products without added walnuts provokes a decrease in total cholesterol of 6.8 mg/dl (CI(95%): -12.8, -0.85).
   J Am Coll Nutr, 2008 · checked 2026-09-29
7. pwal-ws12-01 · meta-analysis of randomized trials
   The pooled analysis of 49 RCTs (57 effect sizes; 4611 participants) showed that Walnut consumption significantly reduced and TG (WMD: -6.24 mg/dL, 95% CI: [-9.49, -2.99], p < 0.001), TC (WMD: -6.39 mg/dL, 95% CI: [-8.54, -4.23], p < 0.001), and LDL-C (WMD: -5.68 mg/dL; 95% CI: [-7,89, -3.47], p < 0.001), while no significant effect was observed on HDL-C (WMD: 0.71 mg/dL, 95% CI: [-0.24, 1.68], p = 0.145), Apo-A1 (WMD: 0.0, 95% CI: [-0.01, 0.00], p = 0.573), or Apo-B (WMD: 0.1, 95% CI: [-0.02, 0.05], p = 0.462). However, subgroup analyses indicated significant reductions in Apo-B at higher walnut doses (> 50 g/day)
   Food Sci Nutr, 2026 · checked 2026-09-29
8. pwal-r7-09 · FDA-approved drug label
   Soybean flour, cottonseed meal, walnuts, and dietary fiber may bind and decrease the absorption of levothyroxine sodium from the gastrointestinal tract.
   DailyMed, levothyroxine sodium label, 2026 · checked 2026-09-29
9. pwal-r7-04 · randomized controlled trial
   We conducted an open-label randomized crossover trial (NCT04799821) in 76 young adults (24.1 ± 3.4 years; 85.5% women) who either ingested 40 g of walnuts daily during dinner (intervention) or refrained from eating walnuts or any other nuts (control) for 8 weeks, with a 2-week washout period. ... The 8-week walnut intervention significantly increased evening urinary 6-SMT concentrations (p = 0.029) and improved sleep latency (p = 0.001), while no differences were found between the baseline and control conditions.
   Food Funct, 2025 · checked 2026-09-29
10. pwal-r7-05 · meta-analysis of randomized trials
    Twenty-eight articles from 27 studies (16 energy intake studies, 10 EE studies, and 1 study investigating both) with 1121 participants were included in this review, with a variety of nut types addressed (almonds, Brazil nuts, cashews, chestnuts, hazelnuts, peanuts, pistachios, walnuts, and mixed nuts). ... The meta-analyses identified a nonsignificant increase in REE associated with nut consumption (weighted mean difference: 28.6 kcal/d; 95% CI: -10.7, 67.8 kcal/d).
    Adv Nutr, 2023 · checked 2026-09-29
11. pwal-r7-10 · EU health claim decision
    POL-HC-8134 Walnuts Walnuts support cardiovascular health. Non-authorised
    EU Register on nutrition and health claims, claim POL-HC-8134, snapshot 2026-09-21 · checked 2026-09-29
12. pwal-r7-06 · non-randomized controlled trial
    We investigated the effect of pumpkin-seed supplementation on oxalcrystalluria and urinary composition in 20 boys age 2-7 yr from a hyperendemic area of Ubol province in Thailand. ... Pumpkin seeds lowered calcium-oxalate crystal occurrence and calcium level but increased phosphorus, pyrophosphate, glycosaminoglycans, and potassium values in urine as compared with orthophosphate supplementation.
    Am J Clin Nutr, 1987 · checked 2026-09-29
13. pwal-r7-07 · retrospective case series
    Twenty-five children (76% boys) were identified, with a median diagnosis age of 2.98 years. Immediate reactions were universal, and anaphylaxis occurred in 64%. ... Co-allergy was common (tree nuts, 76%; other seeds, 72%); watermelon seed sensitization occurred in 83%, with six patients experiencing clinical reactions after watermelon seed ingestion. Pumpkin pulp was tolerated in all the patients.
    Allergy Asthma Proc, 2026 · checked 2026-09-29
14. pwal-r7-08 · systematic review
    Challenge-confirmed IgE-mediated tree nut allergy prevalence was less than 2 % (although only seven studies used this gold standard) while probable tree nut allergy prevalence ranged from 0.05 to 4.9 %. ... Prevalence of individual tree nut allergies varied significantly by region with hazelnut the most common tree nut allergy in Europe, walnut and cashew in the USA and Brazil nut, almond and walnut most commonly reported in the UK.
    Curr Allergy Asthma Rep, 2015 · checked 2026-09-29