BiomaLearnFoodsNutrition factsNutrientsGuidesAnswersEvidence

Pumpkin seeds: a magnesium food whose best trial is about the oil

The strongest trial evidence with pumpkin seeds in its name is about an oil capsule. A network meta-analysis of 19 randomized trials in 1,658 people with pattern hair loss found pumpkin seed oil raised hair density over placebo, a standardized difference of 0.59 (95% CI 0.09 to 1.09). Only a few trials sat behind each of the 16 supplements it compared. Nobody in those trials was eating the seeds. The seed itself is a dense source of minerals. Eight USDA samples of raw pepitas held 469 to 511 mg of magnesium per 3.5 oz (100 g), so a 28 g ounce of shelled seeds gives about 140 mg. The UK food table lists only 270 mg per 3.5 oz (100 g), and the reason for that gap is not known, so read the magnesium as high but uncertain. Selenium ran from 0 to 65 micrograms per 3.5 oz (100 g) across the same eight samples, too wide for any single figure to mean much.

Every statement here is tied to a source. The badge says what kind of evidence stands behind it: A is a meta-analysis or systematic review, E is the position of an expert body without its own analysis. How we verify →

The numbers, per 3.5 oz (100 g)

USDA entrykcalProteinFatCarbsFiber
Seeds, pumpkin seeds (pepitas), raw51529.9 g40 g18.7 g5.08 g

These are the USDA entries that are pumpkin seeds, not foods made from it; they differ in cut, part, pack or preparation. Linked names have a page with the full profile and per-portion figures.

As people eat it, per 3.5 oz (100 g)

As eatenkcalProteinFatCarbsFiber
Pumpkin seeds, NFS57429.8 g49 g14.7 g6.5 g
Pumpkin seeds, salted56729.5 g48.5 g14.5 g6.4 g
Pumpkin seeds, unsalted57429.8 g49 g14.7 g6.5 g

USDA Food and Nutrient Database for Dietary Studies (FNDDS, release 2024-10-31): pumpkin seeds as it is prepared and served, from the recipes behind the national dietary survey, rather than a single laboratory sample.

How much is actually in it

What a real portion looks like

What your body absorbs

What cooking and storage change

What it does to your health

Safety, limits and interactions

What people believe that the data does not support

Who this works differently for

What is still unknown

The bottom line

Every study of the seed as food is small. Adding 2.3 oz (65 g) of pumpkin seed to a high-carbohydrate meal cut the post-meal glucose response by about 35 percent in 15 healthy adults, in a single sitting and with more than two ounces of seed. The authors say it still needs testing over the long term, and we found no longer trial. Eight women who ate 1.1 oz (30 g) of kernels with 1.1 oz (30 g) of iron-fortified cereal every day for four weeks saw serum iron rise from 60 to 85 micrograms per dL. There was no control group, and the cereal carried 7.1 mg of the iron against 4.0 mg from the seeds, so the seed's share cannot be separated out. In 23 postmenopausal women with raised blood pressure, 3 g a day of pumpkin seed oil for six weeks lowered systolic pressure while placebo did not, and the abstract does not say by how much. In 1987, 20 Thai boys aged 2 to 7 from an area with endemic bladder stones had fewer calcium oxalate crystals in their urine on pumpkin seed than on orthophosphate. That is a urine marker in one unusual population, not a count of stones. A German case-control study of 2,884 women with breast cancer linked eating sunflower and pumpkin seeds with lower risk than eating none, an odds ratio of 0.66. The two seeds were counted together, diet was recalled from memory, and the confidence interval printed in the abstract, 0.77 to 0.97, does not contain its own estimate. The popular idea of seed cycling for hormones rests on one case report of a woman with polycystic ovary syndrome who also took myo-inositol and folic acid and conceived through ICSI, so the seeds cannot be credited with anything there. Roasting is gentle on the minerals. By the USDA factors for roasted nuts and seeds as a group, magnesium, iron and zinc stay at 100 percent, while thiamin falls to 85 percent and folate to 80. If you buy them in the shell, an ounce yields only 15 g of seed. The caution that matters most is allergy. In a Turkish clinic series of 25 children with pumpkin seed allergy, 64 percent had anaphylaxis, and every one of them still ate pumpkin flesh without trouble. Three in four of them were also allergic to tree nuts. A separate case describes a 2-year-old, already allergic to milk and egg, who reacted the same way. Tolerating pumpkin tells you nothing about the seed, and which seed proteins cause these reactions is still not worked out. In a UK survey of 3,735 retail dried seeds from 2007 and 2008, E. coli turned up in 9 percent and at unsatisfactory levels in 1.5 percent, pumpkin seeds among them, without a separate share given for pumpkin. One 62-year-old man needed manual disimpaction and colonoscopy for a rectal blockage made of pumpkin seeds. We found no study of how well the minerals in the seed are absorbed and none in people with kidney disease.

Compared with other foods

All food comparisons

Questions about pumpkin seeds, answered from the trials

Pumpkin seeds compared

Related foods

More from the same food group (nut and seed products)

Guides that discuss this food

Sources

  1. ev-pspg-16 · food composition dataset
    Seeds, pumpkin seeds (pepitas), raw | Zinc 6.341 mg per 100 g (range 4.34-10.0, n=8) | Magnesium 499.7 mg per 100 g (range 468.5-511.0, n=8) | Iron 8.363 mg per 100 g (range 7.84-9.42, n=8)
    USDA FoodData Central, Foundation Foods (release 2026-04-30), FDC 2515380, Seeds, pumpkin seeds (pepitas), raw · checked 2026-10-07
  2. pks-r5-01 · USDA FoodData Central Foundation Foods, dataset
    Seeds, pumpkin seeds (pepitas), raw (FDC 2515380, Foundation): Magnesium 499.7 MG range 468.5-511.0 n=8; Phosphorus 1150.0 MG range 1120.0-1190.0 n=8; Iron 8.363 MG range 7.84-9.42 n=8; Copper 1.225 MG range 1.1-1.32 n=8; Manganese 4.121 MG range 3.65-4.94 n=8; Selenium 20.52 UG range 0.0-64.7 n=8
    USDA FoodData Central, Foundation Foods, FDC 2515380, release 2026-04-30 · checked 2026-09-29
  3. pks-r5-02 · CoFID 2021, dataset
    CoFID 14-842 Pumpkin seeds: Energy (kcal) 565; Fat (g) 45.6; Protein (g) 24.4; Copper (mg) 1.57; Iron (mg) 10.00; Magnesium (mg) 270; Phosphorus (mg) 850; Zinc (mg) 6.6
    CoFID 2021 (Public Health England), code 14-842, OGL v3.0 · checked 2026-09-29
  4. pks-r5-04 · USDA FNDDS, dataset
    Pumpkin seeds, NFS (FNDDS 43101050, fdc_id 2707579): 1 cup, without shell 144 g; 1 cup, in shell, yields 46 g; 1 oz, without shell 28 g; 1 oz, in shell, yields 15 g; 1 package 50 g; Quantity not specified 28 g
    USDA FNDDS, FDC 2707579 (food code 43101050) · checked 2026-09-29
  5. ev-pspg-14 · agency guideline
    In addition, routine iron supplementation may prevent women from meeting their zinc requirements by competing with zinc for absorption (3). This may also occur with calcium supplements and foods fortified with inorganic calcium salts (1).
    World Health Organization, WHO antenatal care recommendations for a positive pregnancy experience. Nutritional interventions update: zinc supplements during pregnancy, 2021 · checked 2026-10-07
  6. pks-r5-03 · USDA Table of Nutrient Retention Factors, dataset
    USDA-RETN-R6:2203 NUTS,ROASTED, group Nut and Seed Products: phosphorus_mg 80% retained, folate_total_ug 80%, thiamin_mg 85%, riboflavin_mg 95%, niacin_mg 95%, vitamin_b6_mg 95%, magnesium_mg 100%, iron_mg 100%, zinc_mg 100%, copper_mg 100%
    USDA Table of Nutrient Retention Factors, Release 6, 2007, Retn_Code 2203 · checked 2026-09-29
  7. ev-pspg-01 · meta-analysis
    The evidence suggests that zinc supplementation may result in little or no difference in reducing preterm births (risk ratio (RR) 0.87, 95% confidence interval (CI) 0.74 to 1.03; 21 studies, 9851 participants; low-certainty evidence).
    Cochrane Database Syst Rev, 2021 · checked 2026-10-07
  8. ev-pspg-02 · meta-analysis
    Finally, for other birth outcomes, zinc supplementation may make little or no difference to mean birthweight (MD 13.83, 95% CI -15.81 to 43.46; 22 studies, 7977 participants; low-certainty evidence), and probably makes little or no difference in reducing the risk of low birthweight (RR 0.94, 95% CI 0.79 to 1.13; 17 studies, 7399 participants; moderate-certainty evidence) and small-for-gestational age babies when compared to placebo or no zinc supplementation (RR 1.02, 95% CI 0.92 to 1.12; 9 studies, 5330 participants; moderate-certainty evidence).
    Cochrane Database Syst Rev, 2021 · checked 2026-10-07
  9. ev-pspg-04 · meta-analysis
    In the analysis of all trials, oral magnesium supplementation compared to no magnesium was associated with no significant difference in perinatal mortality (stillbirth and neonatal death prior to discharge) (risk ratio (RR) 1.10; 95% confidence interval (CI) 0.72 to 1.67; five trials, 5903 infants), small-for-gestational age (RR 0.76; 95% CI 0.54 to 1.07; three trials, 1291 infants), or pre-eclampsia (RR 0.87; 95% CI 0.58 to 1.32; three trials, 1042 women).
    Cochrane Database Syst Rev, 2014 · checked 2026-10-07
  10. pks-r5-05 · randomized crossover trial
    Glucose iAUC was significantly lower in P compared with C (reduction of ~35%, P = .025).
    Nutr Res, 2018 · checked 2026-09-29
  11. pks-r5-09 · case-control study
    High and low consumption of soybeans as well as of sunflower and pumpkin seeds were associated with significantly reduced breast cancer risk compared to no consumption (OR = 0.83, 95% CI = 0.70-0.97; and OR = 0.66, 95% CI = 0.77-0.97, respectively).
    Nutr Cancer, 2012 · checked 2026-09-29
  12. ev-pspg-10 · 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%.
    Allergy Asthma Proc, 2026 · checked 2026-10-07
  13. ev-pspg-11 · prevalence survey
    E. coli was detected in 9% of samples, with 1.5% containing unsatisfactory levels (> or = 10(2)/g). These included melon, pumpkin, sesame, hemp, poppy, linseed, sunflower and mixed seeds.
    Food Microbiol, 2009 · checked 2026-10-07
  14. ev-pspg-15 · agency position
    The Tolerable Upper Intake Level for supplemental magnesium is 350 mg for adults, and it ranges from 65 to 350 mg for children and adolescents, depending on age. Too much magnesium from food does not pose a health risk in healthy individuals because the kidneys eliminate excess amounts in the urine [29].
    NIH Office of Dietary Supplements, Magnesium Fact Sheet for Health Professionals · checked 2026-10-07
  15. pks-r5-10 · 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%. ... Pumpkin pulp was tolerated in all the patients.
    Allergy Asthma Proc, 2026 · checked 2026-09-29
  16. pks-r5-11 · case report
    CONCLUSIONS: Pumpkin seed anaphylaxis can develop in a child with a tolerance to pumpkin pulp.
    Nutrition, 2021 · checked 2026-09-29
  17. pks-r5-12 · cross-sectional survey
    E. coli was detected in 9% of samples, with 1.5% containing unsatisfactory levels (> or = 10(2)/g). These included melon, pumpkin, sesame, hemp, poppy, linseed, sunflower and mixed seeds.
    Food Microbiol, 2009 · checked 2026-09-29
  18. pks-r5-13 · case report
    We report a case of a man, aged 62 years, with a rectal bezoar composed of pumpkin seeds with complications necessitating extensive treatment, including manual disimpaction and colonoscopy.
    Clin Med Res, 2012 · checked 2026-09-29
  19. ev-pspg-03 · meta-analysis
    Compared with no zinc, zinc monotherapy among healthy pregnant women resulted in higher serum zinc level (standard mean difference (SMD) the second trimester = 0.32, 95% confidence interval (CI) 0.20 to 0.44; SMDthe third trimester = 0.51, 95% CI 0.27 to 0.76), lower fetal intrauterine retardation rate (risk ratio = 0.23, 95% CI 0.16 to 0.35), longer neonatal birth length (SMD = 0.66, 95% CI 0.21 to 1.12), bigger birth head circumference (SMD = 0.58, 95% CI 0.08 to 1.09), higher 1-min Apgar score (SMD = 0.28, 95% CI 0.06 to 0.49) and cord blood zinc level (SMD = 0.36, 95% CI 0.17 to 0.56).
    J Evid Based Med, 2025 · checked 2026-10-07
  20. ev-pspg-05 · meta-analysis
    The frequency of leg cramps after treatment was not decreased in the treatment group compared to the control group (WMD = -0.47, 95% CI: -1.14-0.20, P = 0.167).
    Taiwan J Obstet Gynecol, 2021 · checked 2026-10-07
  21. ev-pspg-09 · systematic review
    The most commonly cited MP species were Zingiber officinale Roscoe, Allium sativum L. and Cucurbita pepo L. and these were used for relief of NVP, motion sickness and as a nutritional supplement.
    J Ethnopharmacol, 2018 · checked 2026-10-07
  22. ev-pspg-12 · agency position
    During pregnancy, the amount of zinc needed increases to accommodate fetal growth, and the FNB therefore recommends that pregnant women consume 3 mg/day more zinc than nonpregnant women in the same age group [1,3]. Similarly, the zinc requirement increases by 4 mg/day during lactation. NHANES data from 2001 to 2014 show that 11% of pregnant women in the United States have total zinc intakes from foods and supplements that are below the EAR [25].
    NIH Office of Dietary Supplements, Zinc Fact Sheet for Health Professionals · checked 2026-10-07
  23. ev-pspg-13 · agency guideline
    WHO does not recommend zinc supplementation as part of routine ANC. Zinc supplementation is recommended only in the context of rigorous research to improve our knowledge of its effect in pregnant women.
    World Health Organization, WHO antenatal care recommendations for a positive pregnancy experience. Nutritional interventions update: zinc supplements during pregnancy, 2021 · checked 2026-10-07
  24. pks-r5-14 · network meta-analysis
    pumpkin seed oil [PSO; SMD = 0.59 hairs/cm2, 95% CI (0.09, 1.09)]
    Front Nutr, 2025 · checked 2026-09-29
  25. pks-r5-15 · case report
    Following the initial evaluations, they were put on a comprehensive fertility treatment plan that incorporated Myo-inositol and folic acid supplementation, and seed cycling as a dietary intervention.
    J Pharm Bioallied Sci, 2025 · checked 2026-09-29
  26. ev-pspg-06 · meta-analysis
    Overall, compared with control intervention for gestational diabetes, magnesium supplementation was able to significantly decrease FPG (MD = -7.33 mg/dL; 95 % CI = -7.64 to -7.02 mg/dL; P < 0.00001) and HOMA-IR (MD = -0.99; 95 % CI = -1.76 to -0.22; P = 0.01), but resulted in no obvious impact on serum insulin (MD = -4.17 μIU/mL; 95 % CI = -8.49 to 0.14 μIU/mL; P = 0.06), preterm delivery (OR = 0.42; 95 % CI = 0.06 to 2.95; P = 0.38), macrosomia (OR = 0.34; 95 % CI = 0.08 to 1.35; P = 0.13) or BMI change (MD = -0.01 kg/m2; 95 % CI = -0.06 to 0.04 kg/m2; P = 0.63).
    Eur J Obstet Gynecol Reprod Biol, 2024 · checked 2026-10-07
  27. ev-pspg-07 · uncontrolled trial
    The statistical analysis showed a significant difference between the pre and post consumption phase for higher serum iron (60 +/- 22 vs. 85 +/- 23 ug/dl), higher transferrin saturation percent (16.8 +/- 8.0 vs. 25.6 +/- 9.0%), and lower TIBC (367 +/- 31 vs. 339 +/- 31 ug/dl).
    Biofactors, 2007 · checked 2026-10-07
  28. pks-r5-06 · uncontrolled before and after trial
    The statistical analysis showed a significant difference between the pre and post consumption phase for higher serum iron (60 +/- 22 vs. 85 +/- 23 ug/dl), higher transferrin saturation percent (16.8 +/- 8.0 vs. 25.6 +/- 9.0%), and lower TIBC (367 +/- 31 vs. 339 +/- 31 ug/dl).
    Biofactors, 2007 · checked 2026-09-29
  29. pks-r5-07 · randomized controlled trial
    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
  30. pks-r5-08 · randomized controlled trial
    RESULTS: AIx, brachial and central systolic BP significantly (P < 0.05) decreased following PSO but not after placebo.
    Complement Ther Clin Pract, 2019 · checked 2026-09-29
  31. ev-pspg-08 · pilot study
    14 prenatal mothers from Rural Amalapuram in Andhra Pradesh's East Godavari district made up the study sample.
    Andugula SK et al., Bioinformation, 2022, Impact of pumpkin seeds supplementation on anemia in antenatal mothers at Andhra Pradesh, India · checked 2026-10-07
  32. pks-r5-16 · case report
    Pumpkin seed allergens have not been well characterized. Homology between amino-acid sequences in storage proteins may indicate cross reactivity between different edible seeds and nuts.
    Nutrition, 2021 · checked 2026-09-29
  33. pks-r5-17 · randomized crossover trial
    Pumpkin seed has potential as a hypoglycemic food, which now deserves to be confirmed in long-term studies.
    Nutr Res, 2018 · checked 2026-09-29

Review. Reviewed on 2026-09-29 for evidence level, dose and population context, and claims a reader could misread. Bioma Learn has no human medical reviewer at this time, and we say so rather than invent one. Methodology. This is information, not medical advice.