Direct answer
Nattokinase has small, manufacturer-linked blood-pressure trials, but the review’s pooled precision is questionable because original standard errors were entered as standard deviations. A laboratory fibrinaloid assay and human blood-marker study do not demonstrate patient microclot clearance or Long COVID recovery. Bleeding/allergy cases and theoretical medicine interactions warrant prescriber review; nattokinase should not replace prescribed antithrombotic care. Whole natto’s vitamin K2 does not determine every supplement’s content, and some funding and product-supply details remain unreported.
- Nattokinase is a protease associated with fermented soybean food, natto. FU describes laboratory fibrinolytic activity, not a validated clot-treatment dose or an interchangeable mass amount. Trial preparations do not establish every product’s coating or equivalence (review Methods; original BP trial).
- The cited recombinant-enzyme experiment degraded laboratory fibrinaloid clots; it did not test oral treatment in patients. Human blood-marker observations are separate and do not establish Long COVID symptom relief or patient microclot clearance (published experiment; oral biomarker study).
- Small trials report BP signals, with manufacturer involvement. The review’s reported pooled precision is questionable because its figure labels Kim’s standard errors as standard deviations. These data do not justify replacing prescribed treatment (Li Figure 6; Kim Figure 1).
- Bleeding and allergy warrant caution: cases have been reported, while medicine-class interaction risks are mostly theoretical rather than quantified. Discuss blood thinners and planned procedures with the prescriber; do not stop prescribed medicines yourself (clinical monograph; aspirin case; hemoperitoneum case).
- Whole natto is rich in vitamin K2, which matters for warfarin. That does not establish the content of every enzyme supplement; the NSK-SD trial described K2 removal. Assess the exact product and keep vitamin K intake consistent under prescriber advice (NIH food/warfarin guidance; trial preparation).
- FDA does not preapprove dietary supplements for safety or effectiveness. Selling nattokinase as a supplement is not approval of a clot-treatment claim; disease claims are subject to drug requirements (official FDA explanation).
Table of contents
- Evidence summary
- What nattokinase is
- Preparations and activity units
- How it works
- Claims and evidence
- Benefits by claim
- Claims and limits
- Safety findings and gaps
- Medicines and procedural precautions
- Who should seek advice first
- Studied doses and limits
- How laboratory evidence is used
- Funding, conflicts and source roles
- Frequently asked questions
- Sources and funding notes
Evidence summary
| Claim | Evidence | Source | Funding/conflict | Evidence boundary |
|---|---|---|---|---|
| Blood pressure | Li Figure 6 reports SBP: two trials/149 people, −3.45 mmHg; DBP: three/223, −2.32 mmHg | original Figure 6; Kim Figure 1 | Review: no external funding/no conflicts declared; company-linked primary trials | Kim’s SEM values appear under SD in the pool; published precision requires correction |
| Blood pressure, individual trial | 86 randomized; 73 completers; 2,000 FU/day for eight weeks; change-score signals | Kim original | JBSL support/supplied capsules, two company-affiliated authors and Korean research grants | Small selected completer sample; not clinical event or medicine-combination proof |
| Lipid and glucose markers | Small pools; no consistent favourable lipid pattern, small glucose increase reported | review Results | Review declarations do not clear each primary trial’s financing | Subgroup association is not randomized dose-response; no diabetes benefit |
| Clot/microclot treatment in people | Purified-protein assay; small oral study measured normal-range blood-marker changes | Grixti original; Kurosawa original | Grixti patent/grants; Kurosawa KAKENHI support, procurement terms unknown | Neither experiment demonstrated clearance of patients’ clots or clinical recovery |
| Long COVID symptom relief | The cited assay did not treat Long COVID patients; the clinic guide is narrative | published assay; clinic guide | Clinic/product-adjacent content; not a controlled efficacy trial | Not demonstrated by the cited sources |
| Stroke/MI/DVT prevention | Reported risk markers and a small adjunct stroke-rehabilitation trial are different outcomes | review; stroke trial | Stroke trial: Asia Euro Pharmaceutical support; other trial finances need separate review | Does not establish general event prevention or replacement of prescribed care |
What nattokinase is
Nattokinase is a fibrinolytic serine protease associated with Bacillus subtilis fermentation of soybeans. Whole natto, specified enzyme extracts and recombinant laboratory material are different preparations (background monograph; experimental production). A clinic guide promotes microclot-related benefits, but that narrative is not evidence of efficacy or a measure of why public interest changed (clinic article).
FU is an assay of fibrin-degrading activity. Conversion to milligrams requires the preparation’s specific activity: Jensen’s trial used 100 mg NSK-SD standardized to at least 2,000 FU, while Kim reported one 2,000 FU capsule daily. Neither protocol validates every coating or retail product (review Methods; original preparation; Kim original).
Preparations and activity units
| Preparation | Description | Activity/composition | Evidence boundary |
|---|---|---|---|
| Trial capsules | Defined enzyme material versus matched placebo | FU measures in-vitro fibrinolytic activity; coating details cannot be assumed | Protocols and preparation-specific results (review Methods; Kim) |
| NSK-SD trial preparation | Jensen: 100 mg/day; Kurosawa: single 2,000 FU dose | Jensen describes K2 removal; does not determine every retail batch’s content | Jensen original; Kurosawa original |
| Whole natto food | Fermented soybeans with multiple constituents | High menaquinone content; not equivalent to isolated enzyme dosing | NIH food guidance; review exclusion criteria |
| Other powders/extracts | Composition and activity require actual product specifications | Do not infer activity or equivalence from mass alone | The reviewed trials do not establish current market prevalence (trial table) |
How it works
Laboratory proteolysis and effects on fibrinolysis pathways provide hypotheses, not proof of safe clot treatment (mechanism overview; purified-protein assay). Kim’s trial reported a net renin-activity change, but did not establish it as the cause of lower BP or test combinations with BP medicines (original Methods/Results).
Grixti’s published paper formed clots with commercial fibrinogen, bacterial LPS and thrombin, then applied recombinant nattokinase made in the laboratory. It had no human or animal subjects and did not test patient blood, oral absorption or symptom improvement (original Methods/ethics).
Claims and evidence
The RTHM guide links nattokinase to Long COVID microclots and promotes a branded extract. Its claim that the experiment used Long COVID plasma conflicts with Grixti’s stated purified-protein Methods. The guide is evidence of that publisher’s narrative, not a controlled trial or a survey of social-media claims (clinic guide; original Methods).
Li’s review included seven trials qualitatively and six quantitatively. Figure 6 reports two systolic trials (149 people) and three diastolic trials (223); the prose gives a different systolic count and confidence bound. More seriously, the figure puts Kim’s standard errors (SEM) in columns labelled standard deviations (SD). That can overstate pooled precision, so the reported confidence intervals should not be treated as a reliable treatment estimate without corrected inputs and reanalysis (Li Figure 6; Kim Figure 1).
Grixti’s published laboratory assay cannot establish that swallowing nattokinase clears microclots or improves Long COVID (original experiment). Human evidence is broader than test-tube work: Kurosawa measured acute normal-range blood-marker changes, and Pham studied adjunct rehabilitation after stroke. Those designs did not demonstrate patient microclot clearance, Long COVID efficacy or general prevention of stroke, heart attack or DVT (biomarker trial; adjunct stroke trial).
Reported results and limitations
| Outcome | Effect size | Duration/dose | Source |
|---|---|---|---|
| Systolic BP, Li Figure 6 | Reported −3.45 mmHg (95% CI −4.73 to −2.18) | Two trials/149 people; I² 78%; SEM/SD input concern | original figure; primary uncertainty labels |
| Diastolic BP, Li Figure 6 | Reported −2.32 mmHg (−2.72 to −1.92) | Three trials/223 people; I² 45%; SEM/SD input concern | original figure; primary uncertainty labels |
| Systolic BP, Kim trial | Net change −5.55 mmHg (−10.5 to −0.57); p=0.029 | 2,000 FU/day, eight weeks; 73 completers | original Results |
| Diastolic BP, Kim trial | Net change −2.84 mmHg (−5.33 to −0.33); p=0.027 | 2,000 FU/day, eight weeks; 73 completers | original Results |
| Fibrinaloid degradation | Laboratory count/intensity changes; not a human clinical effect | Recombinant enzyme, purified protein mixture, technical replicates | published Methods/Results |
Benefits by claim
Blood pressure
Kim’s small eight-week completer analysis reported lower BP, with manufacturer support. Jensen’s distributor-sponsored trial found no significant between-group systolic difference, while its diastolic result favoured treatment. Those distinct results and Li’s SEM/SD input problem limit confidence in a generic BP-treatment promise (Kim Results; Jensen Results; review Figure 6).
Cholesterol and lipid profile
Li’s small pools did not show a consistent favourable lipid pattern, and reported a small glucose increase. Its subgroups compare cumulative exposure across different protocols; they do not establish a causal dose-response or justify increasing a dose. These are risk markers, not proof of fewer clinical events (original Results/limitations).
"Dissolving" clots and microclots
Grixti demonstrated degradation in purified laboratory mixtures (published experiment). Kurosawa’s 12-person oral study reported blood-marker changes within normal ranges, but did not treat diagnosed clots or assess clinical recovery (original trial). Neither establishes an effective oral dose for clearing a patient’s clots.
Long-COVID and "microclot" symptom relief
The published Grixti experiment did not treat Long COVID patients or measure their symptom recovery. The RTHM guide’s therapeutic narrative cannot supply that missing clinical evidence. These sources therefore do not establish efficacy (published study; clinic guide).
General thrombosis/stroke prevention
Risk-marker changes are different from preventing stroke, heart attack or DVT. Pham’s 61-person, single-blind trial reported functional rehabilitation signals alongside IV Naatrapyl and electroacupuncture after stroke; it did not test first/recurrent stroke prevention or standalone clot treatment (original trial). The discussed sources do not justify substituting nattokinase for prescribed antithrombotic treatment (warfarin-substitution case).
Claims and limits
| Claim | Verdict | Why |
|---|---|---|
| BP lowering | Limited, protocol-specific signals | Kim change scores positive; Jensen systolic comparison null; Li pooled precision has an SEM/SD concern (Kim; Jensen; Li) |
| Lipid improvement | No consistent favourable pattern in this review | Exposure subgroups are not randomized dose-response evidence (Results) |
| Clears patients’ clots | Not established by the discussed experiments | Laboratory degradation and acute blood markers are different outcomes (Grixti; Kurosawa) |
| Long COVID recovery | Not established by these sources | Laboratory assay and clinic narrative did not provide controlled symptom outcomes (experiment; guide) |
| Freely combining with blood thinners | Do not self-combine or substitute | Case warnings and theoretical interactions warrant prescriber review; magnitude is unquantified (monograph; aspirin case) |
Safety findings and gaps
| Side effect | Frequency/severity | Notes |
|---|---|---|
| Selected trial tolerability | No notable events reported in Li’s reviewed trials | Small selected samples and incomplete reporting do not exclude rare bleeding/allergy (review; monograph) |
| Serious bleeding | Case reports; causality and frequency uncertain | Cerebral hemorrhage with aspirin in a susceptible patient; fatal hemoperitoneum with fall/ruptured cyst and uncertain dose (original case; original case) |
| Gastrointestinal symptoms | Reliable rate not established here | The discussed trial reports do not provide a dependable GI incidence estimate (review safety section) |
| Allergy | Cases summarized; frequency uncertain | MSK discusses natto-related allergic reactions. Exact product and prior allergy matter; no universal soy-cross-reactivity rate is established (secondary monograph) |
| Vitamin K exposure | Food- and preparation-specific | Whole natto is rich in K2; Jensen describes its extract as K2-removed. Generic enzyme origin does not determine every product’s content (NIH; original trial) |
Medicines and procedural precautions
| Drug/substance class | Interaction mechanism | Direction of effect | Severity/guidance |
|---|---|---|---|
| Anticoagulants, including warfarin/DOACs | Potential fibrinolysis-related interaction | Magnitude not quantified in these studies | Do not self-combine or replace anticoagulation; seek prescriber review (monograph; substitution case) |
| Antiplatelets, including aspirin/clopidogrel | Possible additive effect; class-wide magnitude unmeasured | Aspirin case is a warning, not proof for every drug/user | MSK advises avoiding daily aspirin combinations; review medicines with the prescriber (monograph; case) |
| Fibrinolytic/thrombolytic medicines | Theoretical overlapping action | Clinical combination risk not quantified | Do not self-combine; disclose supplements to the treating team (professional monograph) |
| Planned surgery/dental procedure | Supplement-related bleeding concerns require review | No universal interruption interval established here | Tell the procedural team the product/dose and follow their plan; do not stop prescribed medicines yourself (FDA supplement precautions; bleeding warnings) |
| BP medicines | Possible additive BP lowering | Combination magnitude not measured; Kim excluded recent antihypertensive use | Discuss concurrent use and monitoring (eligibility; secondary caution) |
| Vitamin K with warfarin | Changing vitamin K intake alters warfarin effect | Natto food and isolated enzyme preparations differ | Maintain consistent intake under prescriber advice; assess actual product composition (NIH guidance; K2-removed trial preparation) |
| Other supplements | Combination effects depend on actual ingredients/exposure | Risk or safety not established by this article | Disclose all products; the cited trials do not clear a proposed stack (FDA precautions) |
MSK labels the main medicine-class interactions as theoretical. A cerebral hemorrhage case involved aspirin and major underlying susceptibility, with the authors explicitly noting that interaction data were lacking. Those reports justify caution, but do not quantify each drug combination or establish causal additivity (monograph; original Discussion).
Who should seek advice first
- People taking anticoagulant, antiplatelet or fibrinolytic medicines: do not add or substitute nattokinase without the treating prescriber’s advice (monograph; substitution case).
- People planning a procedure: disclose the exact supplement and follow the procedural team’s advice on use and timing (FDA precautions).
- People with coagulation disorders or prior cerebral bleeding: MSK warns against use with coagulation disorders; the aspirin case involved bleeding-prone cerebral microangiopathy (monograph; case).
- People with active bleeding or other serious illness should seek medical care and discuss supplements; trials in selected populations do not clear individual high-risk use (FDA advice; trial eligibility).
- Pregnant or breastfeeding people: the cited BP protocols excluded pregnancy/nursing, so their results cannot establish safety in those populations (Kim eligibility; Jensen eligibility).
- People with a previous reaction to soy, natto or a supplement should discuss its actual ingredients with a clinician. MSK summarizes natto-related allergy cases; origin alone does not establish every extract’s allergen content (monograph).
- Warfarin users should discuss changes in vitamin K intake and the exact supplement composition with the prescriber; food-derived origin alone does not establish the product’s K2 level (NIH; trial preparation).
Studied doses and limits
| Preparation | Studied amount | Protocol duration | Limit |
|---|---|---|---|
| Kim test capsule | 2,000 FU once daily | Eight weeks; 73 completers | JBSL supplied; coating/mass not interchangeable across products (original protocol) |
| NSK-SD in defined trials | Jensen 100 mg/at least 2,000 FU daily; Kurosawa single 2,000 FU | Eight weeks / eight-hour biomarker assessment | Different designs; not an optimal retail or Long COVID dose (Jensen; Kurosawa) |
The discussed laboratory and oral biomarker studies do not establish a dose for patient microclot clearance or Long COVID recovery. BP is not the only human endpoint studied: the small adjunct stroke trial used a different product and background treatments. None of those protocols validates self-treatment or substitution for prescribed medicines (laboratory study; oral markers; stroke trial).
How laboratory evidence is used
| Evidence | Type | Reason for exclusion |
|---|---|---|
| Fibrinaloid degradation | Purified proteins/recombinant enzyme in vitro | Mechanistic evidence, not oral clinical efficacy. Published paper and earlier version are the same research (published study; preprint identity) |
The review and monograph discuss cell/animal work and human observations; laboratory findings are useful background but cannot establish patient clot clearance. The recombinant fibrinaloid assay and the acute oral blood-marker study measure distinct outcomes. Human-derived ingredients in a test tube are not a treatment trial (review discussion; mechanism overview; assay; oral study).
Funding, conflicts and source roles
Research funding at a glance
7 disclosure entries. The counts below summarize independence tiers explicitly assigned in this article. They count disclosures, not studies, funding amounts or evidence quality.
Consult this article’s source and funding notes for named funders, countries, relationships and exceptions where available. Institutional backing, researcher interests and trial sponsorship are separate questions. Public funding alone does not establish independence; commercial ties alone do not prove a claim false. This overview is not a new financial audit.
| Source | Disclosed support/relationships | Assessment limit |
|---|---|---|
| Li 2023 review | No external funding or conflicts declared | Review declarations do not establish every primary trial’s finances; input/precision concern remains (original) |
| Kim 2008 / Jensen 2016 | Kim: JBSL support/supplied capsules, two company-affiliated authors and Korean grants. Jensen: JBSL-USA distributor sponsorship | Commercial support despite academic/contract-lab affiliations; full amounts/procurement terms unknown (Kim original; Jensen declarations) |
| Grixti 2024 published assay | Balvi, Novo Nordisk Foundation, South African NRF and MRC grants; Pretorius microclot-detection patent application | Funder-no-role statement reported; ultimate backing, patent terms and allocations not fully traced (original declarations) |
| Earlier bioRxiv version | Same research and named patent interest | Landing-page identity/declaration read; full published main paper used for methods/results, not independent replication (preprint record) |
| RTHM guide / Rising Trends | Clinic/product-adjacent narrative / commercial trend report | Not efficacy studies; page-specific financing and reproducible market-data methods not fully established (guide; trend report) |
| MSK monograph | Secondary clinical information service | Page-specific financial/editorial relationships not itemised; nonprofit identity is not proof of independence (monograph) |
| WebMD / additional original studies | WebMD: secondary commercial publisher. Kurosawa: KAKENHI support. Pham: Asia Euro Pharmaceutical funding, IMC active/placebo | Page-specific and product-supply terms not fully reported; case-report finances also incompletely itemised (WebMD; Kurosawa; Pham) |
No-conflict and no-external-funding declarations concern the authors or synthesis that made them. They cannot be transferred to every included trial, product or publisher. Kim and Jensen report commercial support; Grixti reports named grants and a patent interest. Unknown amounts, procurement terms or ultimate backers remain unknown (review; Kim; Jensen; Grixti).
Frequently asked questions
Can nattokinase actually dissolve blood clots or microclots in my body?
The discussed experiments do not demonstrate clearing patients’ clots after swallowing a supplement. Grixti used recombinant enzyme with purified proteins; Kurosawa measured acute blood markers in 12 healthy men, all changes within normal ranges. Neither measured clinical clot clearance or recovery (published assay; oral study).
Does nattokinase help with Long-COVID?
The cited Grixti assay did not administer treatment to Long COVID patients or measure symptoms. The RTHM narrative does not supply controlled efficacy evidence. These sources do not establish benefit (experiment; guide).
Is nattokinase's blood-pressure benefit real?
Kim reported modest change-score reductions in a small, manufacturer-supported completer sample; Jensen’s distributor-sponsored trial found a null systolic comparison and a favourable diastolic result. Li’s figure has two systolic and three diastolic trials, but labels Kim’s SEMs as SDs, so the published pooled precision needs correction. This is limited evidence, not a validated substitute for BP care (Kim; Jensen; Li Figure 6).
Is it safe to take nattokinase with aspirin or warfarin?
Do not self-combine or substitute it for prescribed blood thinners. MSK’s class-wide interaction concerns are theoretical; an aspirin-associated cerebral hemorrhage case had important underlying susceptibility and explicitly lacked interaction data. The case supports caution but cannot quantify warfarin, DOAC or clopidogrel risks (monograph; original case; failed substitution case).
Does nattokinase contain vitamin K2?
Whole natto is rich in K2; that does not determine every nattokinase supplement’s content. Jensen’s trial described its NSK-SD preparation as K2-removed. Warfarin users should keep vitamin K intake consistent and discuss the exact product with their prescriber (NIH guidance; original preparation).
Is nattokinase FDA-approved?
FDA does not preapprove dietary supplements for safety or effectiveness. A supplement’s availability is not FDA approval of a clot-treatment claim; products intended to treat or prevent disease are subject to drug requirements. This regulatory distinction is separate from whether an ingredient has clinical benefit (official FDA explanation).
Sources and funding notes
- Li 2023 cardiovascular review — Seven trials discussed, six pooled; outcome-specific samples and an SEM/SD input problem limit interpretation. No external review funding/no conflicts declared; not complete primary-trial financial clearance.
- Kim 2008 original trial — Eight-week, 73-completer BP change-score analysis; JBSL support/supplied capsules, company-affiliated authors and Korean grants. The same paper is indexed as PMID18971533; that record is not a second trial.
- Grixti 2024 published laboratory paper — Purified proteins/recombinant enzyme, no patient blood or clinical treatment; named grants and microclot-detection patent interest. Main methods/results/declarations read, supplemental movies not reviewed.
- Earlier bioRxiv record — Same research; landing-page abstract/identity/patent declaration read. Published original main paper supplies the methodological basis, not separate replication.
- MSK nattokinase monograph — Secondary clinical warnings and bibliography; class interactions mainly theoretical. Page-specific financial declarations not itemised, so nonprofit identity does not certify independence.
- WebMD monograph — Available secondary medicine cautions; some current overview/safety fields are empty. Not primary clinical efficacy or complete financial evidence.
- RTHM guide — Clinic/product-adjacent microclot narrative, not controlled treatment evidence. Rising Trends provides commercial search-interest reporting, not causal clinical evidence.
- Additional original/official sources: Chang 2008 aspirin/hemorrhage case; Ramachandran 2021 hemoperitoneum case; Elahi 2015 warfarin-substitution case; Jensen 2016 NSK-SD trial; Kurosawa 2015 oral biomarkers; Pham 2020 adjunct rehabilitation; NIH vitamin K guidance; FDA supplement regulation. Case-level causality and unreported financial/product terms remain uncertain.
Have a question — or want us to cover something?
Ask about anything on this page, or request the next deep dive: an ingredient, a supplement, or a health concern. We use published research, evidence syntheses, and regulatory guidance, with clear source links.
One daily research roundup
Get the topics, key findings and links from our new articles in one email. At most one digest a day, only when there is something new.