It chews up fibrin, the protein mesh holding a clot together 1. It doesn't stop clots from forming the way aspirin does; it breaks down what's already there. The blood-pressure effect runs through a second, milder path: small peptides that nudge down the same system ACE-inhibitor drugs target.
Two things are measured well, one never at all. Blood pressure is the clean signal: at 2,000 units a day for eight weeks, the top number falls about 3 to 5 points, the bottom 2 to 3 2. Enough to move a borderline reading; not enough to replace a pill someone needs.
The clotting markers move too: fibrinogen down roughly a tenth, two clotting factors down about a sixth 3.
The plaque claim is what marketers lean on hardest, and it's the shakiest. High-dose trials in people who already have plaque show big shrinkage 4, but had no placebo group or weren't randomized. At the standard dose, over three years, in healthy adults, plaque didn't budge 5. Whether any of this prevents a heart attack or stroke has never been tested.
Three groups, three answers. A good fit: untreated mildly high blood pressure (say 130–145 over 85–95), no blood thinner, willing to try the gentle option and recheck. A speculative fit: diagnosed carotid plaque, with a cardiologist on board for the high-dose version. A bad fit, most readers: healthy, normal pressure, no plaque, the exact group the null trial studied 5.
Before buying anything, settle whether you're the right person for it.
None of this is felt. No energy lift, no mood shift; the reward is on paper. By week eight, if you started high, the home cuff reads a few points lower: a 142/88 person becomes roughly 137/85. A follow-up panel shows fibrinogen and two clotting factors a notch down. Beyond that the data runs out. Nobody has shown it prevents a heart attack or stroke, or adds years.
The fine print — when to skip it, and what people get wrong
Do not take it with warfarin, apixaban, rivaroxaban, dabigatran, edoxaban, preventive aspirin, or clopidogrel; nor with a recent stroke, active ulcer, clotting disorder, pregnancy, or breastfeeding 6. Never swap it for a prescribed blood thinner. Soy allergy is a flag.
Not the same as eating natto: the food carries vitamin K2 the supplement strips out. "Natural blood thinner" is half-true; it breaks down fibrin but doesn't prevent clots, and no cardiologist backs it as a swap. "Take it just in case" is the exact use that was tested and failed 5.
Common misfires: the capsule was never truly 2,000 units (no enforced assay standard, so buy third-party tested); expecting the plaque effect at the blood-pressure dose 7; quitting before eight weeks; hidden bleeding risk from not telling the doctor you take it.
- 1Sumi H, Hamada H, Tsushima H, Mihara H, Muraki H (1987). A novel fibrinolytic enzyme (nattokinase) in the vegetable cheese Natto; a typical and popular soybean food in the Japanese diet. Experientia. link
- 2Yuan L, Liangpunsakul S, Yuan J, Yu Y, Zhao F, Lu Y, Sun J, Lu Y, Zhou Y, Zhang L, Wang Y (2023). Nattokinase supplementation and cardiovascular risk factors: a systematic review and meta-analysis of randomized controlled trials. Reviews in Cardiovascular Medicine. link
- 3Hsia CH, Shen MC, Lin JS, Wen YK, Hwang KL, Cham TM, Yang NC (2009). Nattokinase decreases plasma levels of fibrinogen, factor VII, and factor VIII in human subjects. Nutrition Research. link
- 4Ren NN, Chen HJ, Li Y, McGowan GW, Lin YG (2017). A clinical study on the effect of nattokinase on carotid artery atherosclerosis and hyperlipidaemia. Zhonghua Yi Xue Za Zhi. link
- 5Hodis HN, Mack WJ, Meiselman HJ, Kalra V, Liebman H, Hwang-Levine J, Dustin L, Kono N, Mert M, Wenby RB, Huesca E, Rochanda L, Li Y, Yan M, St John JA, Whitfield L (2021). Nattokinase atherothrombotic prevention study: a randomized controlled trial. Clinical Hemorheology and Microcirculation. link
- 6Chang YY, Liu JS, Lai SL, Wu HS, Lan MY (2008). Cerebellar hemorrhage provoked by combined use of nattokinase and aspirin in a patient with cerebral microbleeds. Internal Medicine. link
- 7Chen H, McGowan EM, Ren N, Lal S, Nassif N, Shad-Kaneez F, Qu X, Lin Y (2022). Effective management of atherosclerosis progress and hyperlipidemia with nattokinase: a clinical study with 1,062 participants. Frontiers in Cardiovascular Medicine. link
Nattokinase
$100-200/year at the standard 2,000 FU/day from a reputable brand; $300-600/year at the higher 6,500-10,800 FU/day plaque-modifying doses. Not covered by insurance.
One to three capsules daily; no lifestyle change required. The only friction is sourcing a reputable brand with FU activity labeled and storing it out of heat.
Six small RCTs and a 2023 meta-analysis (n=546) confirm a modest, replicable blood-pressure effect (Yuan 2023; Kim 2008); biochemical mechanism is well-characterized (Sumi 1987; Kurosawa 2015). Atherosclerosis evidence is mixed: positive at high dose in affected populations (Ren 2017; Chen 2022), null in the most rigorous low-risk primary-prevention RCT (Hodis 2021). No hard-endpoint trial. Plausible mechanism, small studies, monitor for updates.
Measurable 3-5 mmHg systolic / 2-3 mmHg diastolic blood pressure reduction at 8 weeks (Kim 2008; Yuan 2023 meta-analysis), plus fibrinogen and factor VII/VIII reductions visible on a coagulation panel (Hsia 2009). Real but small; the reader does not feel hypertension lifting — the wins show up on a cuff and a lab printout.
Surrogate-endpoint reductions in BP and coagulation factors and a high-dose carotid plaque signal in already-affected populations (Ren 2017; Chen 2022) suggest plausible mortality benefit, but the most rigorous primary-prevention RCT (Hodis 2021 NAPS) was null on IMT progression in low-risk adults, and no trial has measured hard endpoints. Small additive effect at best.