Hemp seed: shifts blood fatty acids, not cholesterol, in few trials

The strongest hemp number in the literature is a 43 percent higher complete response in functional constipation, pooled from 17 east Asian trials with 1,681 patients. It belongs to Hemp Seed Pill, a Chinese formula of several herbs, and the authors of that analysis note the effect cannot be put on hemp seed as a food. For the seed itself, a 2017 review of 113 controlled trials of foods sold as superfoods found none that tested hemp. The small trials since then say one consistent thing. In 30 adults with overweight, 4 weeks of hemp products raised blood ALA, GLA and DGLA, and left cholesterol, glucose, insulin and blood pressure where they were.

The numbers, per 3.5 oz (100 g)

What your body absorbs

• In a 4-week crossover trial in 30 adults with overweight, eating hemp products raised blood ALA, GLA and DGLA but left lipids, glucose, insulin and blood pressure unchanged. — 30 normoglycemic adults aged 18 to 65, BMI 25 to 35, 4 weeks per arm (n = 30) Evidence B source
  Plasma and red-cell ALA, GLA, DGLA and EPA higher after hemp than control; no difference in serum lipids, glucose, insulin, blood pressure or body composition
• Taking 2 tbsp (30 ml) of hempseed oil a day for 4 weeks raised linoleic and gamma-linolenic acid in serum lipids of 14 healthy adults, with no change in fasting cholesterol, glucose or insulin versus flaxseed oil. — 14 healthy volunteers, 2 tbsp (30 ml) oil per day, 4 weeks per arm with 4-week washout (n = 14) Evidence B source
  Higher LA and GLA in serum cholesteryl esters and triglycerides after hempseed oil (P < 0.001); total-to-HDL ratio lower, P = 0.065

What it does to your health

• Before a meal, 20 or 40 g of hemp protein lowered the following blood glucose rise in a dose-dependent way, just as soy protein did, in two crossover trials of 27 and 16 adults. — healthy adults, 27 in experiment 1 and 16 in experiment 2, single-visit crossovers (n = 43) Evidence B source
  0-60 min glucose mean and AUC lower than carbohydrate control (p < 0.05), dose-dependent; 40 g lowered pre-meal insulin; no difference hemp versus soy

Safety, limits and interactions

• In a 1997 test, volunteers who ate two hemp seed bars or three hemp flour cookies produced urine that screened positive for marijuana, though none passed GC-MS reporting criteria. — volunteers eating commercial hemp seed bars or cookies, urine for 24 h (n = -) Evidence C source
  5 specimens positive at 20 ng/mL after two bars; several positive at 50 and 20 ng/mL after three cookies; GC-MS 0.6 to 3.1 ng/mL, below reporting criteria
• In 35 adults, oil with 0.046 or 0.092 mg THC per dose gave no positive THC test in saliva, blood or urine. — 35 adults aged 22 to 52, 1 tsp (5 mL) oil doses over a three-week protocol (n = 35) Evidence C source
  No THC detected at any time point; only trace THC acid (0.0001 to 0.0004 mg/L) at 4 h
• A high-THC hemp oil sold as a supplement (14.8 mg THC a day) gave plasma THC levels comparable to the prescription drug dronabinol in 6 volunteers. — 6 volunteers, inpatient, hemp oils with 0.39 to 14.8 mg THC a day versus dronabinol 7.5 mg (n = 6) Evidence B source
  THC and 11-OH-THC never above 6.1 ng/mL; THCCOOH up to 43.0 ng/mL, detectable 39.5 h after dosing
• A 2010 German Salmonella Montevideo outbreak was tied to a supplement made mainly of hemp seed flour, with case odds 27.5 times higher among women who took it. — case-control study of 55 adult women, Germany, 2010 (n = 55) Evidence C source
  OR 27.5 (95% CI 3.1 to infinity), 43.5 at the outbreak peak

What people believe that the data does not support

• Two 1 g capsules of hempseed oil a day for 12 weeks did not change any plasma fatty acid or blood lipid in 86 healthy adults. — 86 healthy men and women, 2 g oil per day in capsules, 12 weeks (n = 86) Evidence B source
  No significant change in plasma fatty acids, TC, HDL, LDL, TG, platelet aggregation or inflammatory markers
• Across 17 trials with 1,681 patients in east Asia, the Chinese herbal formula called Hemp Seed Pill raised complete response in functional constipation by 43% versus placebo. — 1,681 constipation patients in 17 east Asian trials, moderate to high risk of bias (n = 1681) Evidence A source
  RR 1.43 (95% CI 1.20 to 1.71) for complete response and +0.95 weekly CSBM (0.56 to 1.35) in two high-quality placebo trials; no rise in adverse events

Who this works differently for

• In 36 children aged 6 to 16 with primary hyperlipidemia, 3 g of hempseed oil a day for 8 weeks raised the red-cell omega-3 index by 1.18 points but did not improve blood lipids. — 36 hyperlipidemic children and adolescents aged 6 to 16 on diet therapy, 8 weeks (n = 36) Evidence B source
  Omega-3 index +1.18 +/- 1.42%, total n-3 PUFA +1.57%, no change in serum lipid profile versus control

What is still unknown

• A systematic review of foods sold as superfoods found 113 controlled trials on metabolic syndrome markers and none of them tested hemp seed. — controlled human intervention trials up to April 2017, 17 superfoods (n = 0 hemp trials) Evidence C source
  Hemp seed 0 trials, versus flaxseed 22 and chia 5

The bottom line

The oil tells the same story. In 14 healthy adults, 2 tbsp (30 ml) of hempseed oil a day for 4 weeks raised linoleic and gamma-linolenic acid in serum lipids with no change in fasting cholesterol, glucose or insulin compared with flaxseed oil. Two 1 g capsules a day for 12 weeks changed no plasma fatty acid or blood lipid in 86 healthy adults, and that dose is far below what a spoonful of oil gives. In a pilot in 36 children aged 6 to 16 with primary hyperlipidemia, 3 g of oil a day for 8 weeks raised the red-cell omega-3 index by 1.18 points and did not improve their blood lipids, with no placebo oil for comparison. Hemp protein behaves like other proteins before a meal. In two crossover trials of 27 and 16 healthy adults, 20 or 40 g of hemp protein concentrate taken before eating lowered the following rise in blood glucose in step with the dose, just as soy protein did. That was a single-visit effect, and the abstract does not say by how much. The cautions are about what travels with the seed. In a 1997 test, two hemp seed bars or three hemp flour cookies gave urine that screened positive for marijuana, though no sample passed the confirmatory GC-MS criteria, and those were 1997 products, not seed sold today. In 35 adults, sesame oil spiked with 0.046 or 0.092 mg THC per dose gave no positive THC test in saliva, blood or urine. A high-THC hemp oil sold as a supplement in 2006, at 14.8 mg THC a day, gave 6 volunteers plasma THC comparable to the prescription drug dronabinol. If you face workplace drug testing, check the THC content on anything labeled hemp oil. Raw hemp seed flour in a supplement was tied to a 2010 Salmonella outbreak in Germany, with case odds 27.5 times higher among women who took it and a confidence interval running from 3.1 to infinity. We found no study of hemp seed in pregnancy, in infants or in kidney disease.

Sources

1. hsd-r6-01 · RCT crossover
   Hemp seeds are high in polyunsaturated fatty acids (PUFAs) including gamma linolenic acid (GLA), stearidonic acid (SDA), alpha linolenic acid (ALA) and linoleic acid (LA). ... No differences in serum lipid levels, glucose and insulin concentrations, blood pressure, or body composition were observed between treatments.
   Food Funct, 2026 · checked 2026-09-29
2. hsd-r6-02 · RCT crossover
   The volunteers consumed HO and FO (30 ml/day) for 4 weeks each. ... The HO period resulted in higher proportions of both LA and gamma-linolenic acid in serum cholesteryl esters (CE) and triglycerides (TG) as compared with the FO period (P < 0.001)
   Eur J Nutr, 2006 · checked 2026-09-29
3. hsd-r6-05 · RCT crossover
   Results suggest that hemp protein, like soybean, dose-dependently lowers postprandial BG and insulin concentrations compared with a carbohydrate control.
   Appl Physiol Nutr Metab, 2021 · checked 2026-09-29
4. hsd-r6-07 · controlled provocation
   In this study, ingestion of hemp seed food products resulted in urine specimens that screened positive for marijuana. No specimens gave a GC-MS quantitative value above the limit of detection for marijuana.
   J Anal Toxicol, 1997 · checked 2026-09-29
5. hsd-r6-08 · controlled crossover
   No THC was detected in oral fluid, blood or urine samples at any time-point following consumption of the low or high THC dose.
   Forensic Sci Int, 2018 · checked 2026-09-29
6. hsd-r6-09 · RCT crossover
   This study demonstrated that subjects who used high THC content hemp oil (347 mug/mL) as a dietary supplement had THC and metabolites in plasma in quantities comparable to those of patients using dronabinol for appetite stimulation.
   Ther Drug Monit, 2006 · checked 2026-09-29
7. hsd-r6-10 · case-control
   Trace-back of the supplement's main ingredient, hemp seed flour, and subsequent microbiological testing by pulsed-field gel electrophoresis supported its likely role in transmission.
   Euro Surveill, 2011 · checked 2026-09-29
8. hsd-r6-03 · RCT
   However, supplementation with hempseed oil did not significantly alter the concentration of any plasma fatty acid. The lipid parameters (TC, HDL-C, LDL-C and TG) did not show any significant differences among the four groups.
   J Am Coll Nutr, 2008 · checked 2026-09-29
9. hsd-r6-06 · meta-analysis
   Two high-quality trials showed that HSP compared with placebo significantly increased weekly CSBM (mean difference, 0.95; 95% CI: 0.56, 1.35) and had a higher complete response rate (risk ratio [RR], 1.43; 95% CI: 1.20, 1.71) in patients with functional constipation.
   Phytomedicine, 2021 · checked 2026-09-29
10. hsd-r6-04 · RCT pilot
    Eight weeks of supplementation with HSO significantly (p < .01) reduced the RBC content of total saturated and monounsaturated FAs (-5.02 ± 7.94% and - 2.12 ± 2.23%, respectively), increased the levels of total n-3 and n-6 PUFAs (+1.57 ± 1.96% and + 5.39 ± 7.18%, respectively) and the omega-3 index (+1.18 ± 1.42%), but failed to affect the serum lipid profile compared to the control group.
    Food Res Int, 2019 · checked 2026-09-29
11. hsd-r6-11 · systematic review
    Seventeen superfoods were identified, including a total of 113 intervention trials: blueberries (8 studies), cranberries (8), goji berries (3), strawberries (7), chili peppers (3), garlic (21), ginger (10), chia seed (5), flaxseed (22), quinoa (1), cocoa (16), maca (1), spirulina (7), wheatgrass (1), acai berries (0), hemp seed (0) and bee pollen (0).
    Food Funct, 2018 · checked 2026-09-29