Almonds: a real effect, about three percent of your cholesterol

Thirty-six publications and 2485 people put a number on almonds and cholesterol: LDL down 0.13 mmol/L, total cholesterol down 0.16, apolipoprotein B down 4.6 mg/dL. That is roughly 3 percent of a typical LDL, which is real and small. Europe refused the claim that says exactly this, that a 28 gram handful a day helps maintain healthy cholesterol, because a refusal records what a dossier proved to assessors at the time rather than what later trials show. Both facts belong on the same page. The weight story is weaker: half a kilogram across 37 trials, with body mass index, body fat percentage and visceral fat all unchanged, and half the outcomes graded very low certainty.

The numbers, per 100 g

What a real portion looks like

• The antioxidant markers moved only above sixty grams a day, which is more than twice a standard handful. — 8 studies (5 randomised trials and 3 crossover trials) with 424 adults (n = 424) Evidence A source
  At doses above 60 g a day, malondialdehyde -0.46 (p = 0.002), 8-hydroxy-2-deoxyguanosine -5.83 (p < 0.001) and uric acid -0.64 (p = 0.009) fell while superoxide dismutase rose 2.02 (p = 0.008); glutathione peroxidase unchanged; heterogeneity 92 to 96%

What your body absorbs

• Whole raw almonds give up the fewest calories, and grinding or roasting them hands over more. — 21 studies, both in vitro and in adults over 18 (n = 21 studies) Evidence C source
  Metabolizable energy was 18.5 kJ/g for raw almonds against 22.6 kJ/g for pistachios, and rose with physical form (flour above chopped above whole) and with heat processing (nut butter above roasted above raw)

What it does to your health

• Almonds lowered LDL cholesterol by 0.13 millimoles per litre across thirty-six trials. — 36 publications giving 48 almond-control datasets with 2485 participants (n = 2485) Evidence A source
  LDL cholesterol -0.132 mmol/L (95% CI -0.190 to -0.075), total cholesterol -0.160 (-0.218 to -0.101), non-HDL cholesterol -0.204 (-0.281 to -0.127), total to HDL ratio -0.154, LDL to HDL ratio -0.112, apolipoprotein B -4.552 mg/dL (-6.460 to -2.645) and ApoB to ApoA ratio -0.027, all statistically significant

What people believe that the data does not support

• Europe refused the almond cholesterol claim that the trials support, because the dossier and the evidence are judged separately. — — Evidence E source
  Three register entries name almonds and all three carry the status Non-authorised; the refused wording is that a handful of 28 g of almonds a day within a diet low in saturated fat helps maintain healthy blood cholesterol and promotes a healthy heart

What is still unknown

• Almonds shaved off about half a kilogram and left body mass index, body fat percentage and visceral fat unchanged. — 37 randomised controlled trials with 43 treatment arms in adults (n = 37 trials) Evidence A source
  Body weight -0.45 kg (95% CI -0.85 to -0.05), waist circumference -0.66 cm, fat mass -0.66 kg and hunger score -1.15 mm all significant; BMI, body fat percentage, fat-free mass, waist-to-hip ratio, visceral adipose tissue, fullness and desire to eat all unchanged; GRADE certainty very low for appetite, body fat percentage, fat-free mass, visceral fat and waist-to-hip ratio, moderate for the rest
• Almonds moved one inflammation marker and not the one doctors actually measure. — 11 studies in adults (n = 11 studies) Evidence A source
  C-reactive protein unchanged (-0.28 mg/L, 95% CI -0.81 to 0.25, p = 0.29); interleukin-6 lower by 0.10 pg/mL (-0.15 to -0.05, p < 0.001)

The bottom line

The calorie arithmetic is the part worth keeping. Raw almonds release about 18.5 kilojoules per gram, well under what their fat, protein and carbohydrate predict, because some fat never escapes the cell walls. Grinding them into flour or roasting them hands over more. So whole raw almonds are the version that costs you least. The antioxidant marker findings needed more than 60 grams a day, which is about 45 almonds and 350 calories, and the studies behind them disagree with each other so strongly that the pooled figures are hard to trust. One thing we could not check: almond research is heavily funded by the industry that sells almonds, and none of the four syntheses we read breaks its results down by who paid.

Sources

1. alm-dr-04 · systematic review and meta-analysis of randomised controlled trials
   Almond doses of > 60 g/day significantly reduced MDA (WMD = -0.46, p = 0.002), 8-OHdG (WMD = -5.83, p < 0.001), and UA (WMD = -0.64, p = 0.009), while increasing SOD (WMD = 2.02, p = 0.008).
   Sci Rep, 2025 · checked 2026-09-22
2. alm-dr-05 · systematic review with narrative synthesis
   ME was influenced by nut type (ME = 22.6 kJ/g for pistachios; ME = 18.5 kJ/g for raw almonds), physical form (flour > chopped > whole nuts), heat processing (butter > roasted > raw) and dose of consumption.
   Adv Nutr, 2023 · checked 2026-09-22
3. alm-dr-01 · systematic review and meta-analysis of randomised controlled trials
   Almond consumption significantly reduced LDL-C (-0.132 mmol/L; 95% CI: -0.190, -0.075 mmol/L; p < 0.001), TC (-0.160 mmol/L; 95% CI: -0.218, -0.101 mmol/L; p < 0.001), non-HDL-C (-0.204 mmol/L; 95% CI: -0.281, -0.127 mmol/L; p < 0.001)
   Nutrients, 2025 · checked 2026-09-22
4. alm-dr-06 · regulatory register entry
   POL-HC-6583 Almonds Consuming a handful (28 grams) of almonds a day as part of a diet low in saturated fat, helps to maintain healthy blood cholesterol and promotes a healthy heart Non-authorised
   EU Register on nutrition and health claims, заява POL-HC-6583, знімок 2026-09-21 · checked 2026-09-22
5. alm-dr-02 · systematic review and dose-response meta-analysis of randomised controlled trials
   Pooled effect sizes indicated a significant reducing effect of almond consumption on body weight (WMD: -0.45 kg, 95% CI: -0.85, -0.05, p = 0.026), WC (WMD: -0.66 cm, 95% CI: -1.27, -0.04, p = 0.037), FM (WMD: -0.66 kg, 95% CI: -1.16, -0.17, p = 0.009), and hunger score (WMD: -1.15 mm, 95% CI: -1.98, -0.32, p = 0.006) compared with the control group.
   Obes Rev, 2024 · checked 2026-09-22
6. alm-dr-03 · systematic review and meta-analysis
   The overall estimate indicated that the almond consumption had no significant effect on serum CRP level (weighted mean difference (WMD) = -0.28 mg/l, 95 % confidence interval (CI): - 0.81, 0.25; p = 0.29). Regarding IL-6, almond consumption significantly decreased serum IL-6 level (WMD = -0.1 pg/ml, 95 % CI: -0.15, -0.05; p < 0.001).
   Complement Ther Med, 2023 · checked 2026-09-22