Lumbrokinase
A fibrinolytic enzyme complex from earthworms. Distinct from nattokinase and considerably more potent on fibrin in vitro, which cuts both ways.
Lumbrokinase quick facts
| Suggested dose | 20–40 mg (roughly 300,000–600,000 IU) daily on an empty stomach. |
| How often | daily |
| Who it's for | Discussed in post-COVID and hypercoagulability circles. Should be a physician-supervised decision, not a self-experiment. |
Higher potency means the bleeding interaction is correspondingly more serious, and the human evidence base is thinner than nattokinase's — mostly Chinese studies in stroke. It is used clinically in some settings, which is more than most supplements can claim, but this belongs in a monitored context rather than casual self-experimentation. Same absolute rules on anticoagulants and surgery.
How Lumbrokinase actually works
An enzyme complex from earthworms, more potent than nattokinase in fibrin degradation assays and acting through both direct fibrinolysis and plasminogen activation. It appears to have greater specificity for fibrin over other proteins, which is the theoretical argument for a better safety margin.
Where to get Lumbrokinase
Find Lumbrokinase on iHerb →The evidence for Lumbrokinase
Graded by what exists behind each claim.
✅ Clinically validated
- Most trials are Chinese-language, in ischemic stroke and unstable angina, reporting improved fibrinolytic markers and clinical scores. Methodology and reporting are generally below Western trial standards.
- No large independent Western RCT exists.
📊 Correlative data
- Derived from earthworms, which have long use in traditional Chinese medicine, and it has been used clinically in China for ischemic stroke. Almost no independent Western observational data.
🧪 Theoretical / extrapolated benefits
- Directly degrades fibrin and activates plasminogen — a genuine dual mechanism, unlike agents that only do one.
How to read these tiers: they say how much human evidence exists, not how well something works — and ✗ flags harm, never a disappointing trial. How the evidence tiers work →
What Lumbrokinase actually does
Lumbrokinase is not a molecule. It is a fraction. Nattokinase is one 27.7 kDa subtilisin-family serine protease and you can write its sequence down. Lumbrokinase is a group of serine proteases co-purified from the gut and body wall of Lumbricus rubellus, and a careful purification of that worm separates five distinct fibrinogenolytic protein bands with molecular weights of 42.6, 27.03 and 14 kDa among them Stephani 2023. Every downstream problem on this page — the units, the brand-to-brand comparison, the missing pharmacokinetics — follows from the fact that the product is a mixture whose composition is set by an extraction process rather than by a specification.
Two different catalytic behaviors, sold as one thing. Some components of this fraction are direct fibrinolytics: they hydrolyze fibrin themselves, the way plasmin does, without asking permission. Others behave as plasminogen activators, converting the body's own plasminogen to plasmin the way tissue plasminogen activator does Stephani 2023Hazare 2024. That distinction is not academic. A plasminogen activator borrows a regulated system that has its own brakes — plasminogen activator inhibitor-1, alpha-2-antiplasmin, thrombin-activatable fibrinolysis inhibitor. A direct fibrinolytic has no brake at all. A product that contains both has one mechanism the body can modulate and one it cannot.
The specific activities are measurable, and they are not equal. In the same purification, the F25 fractions showed fibrinogenolytic activity of 974.85 U/mg while the F85 fractions reached 1,484.11 U/mg Stephani 2023. That is a 1.5-fold difference in potency per milligram between two fractions of the same animal, prepared in the same laboratory, on the same day. Hold that number: it is the reason milligrams on a lumbrokinase label carry no information.
Where the non-fibrin mechanism sits. Beyond fibrin, lumbrokinase has been shown in a fibroblast model to reduce migration and adhesion by attenuating AP-1 driven ICAM-1 expression Nguyen 2023, and the broader earthworm-extract literature reports antiplatelet and blood-viscosity effects alongside the proteolytic one Zhu 2024Pinzon 2021. Note what that means for safety: the antiplatelet arm and the fibrinolytic arm are additive with each other, inside the same capsule.
And the question nobody has answered. This is a 14–43 kDa protein being swallowed. Enteric coating protects it from gastric acid; nothing protects it from pancreatic trypsin and chymotrypsin in the duodenum, and nothing explains how a protein of that size crosses the enterocyte intact. No study has measured lumbrokinase, or a signature peptide of it, in human plasma after an oral dose. The only evidence that anything systemic happens is that plasma fibrinogen falls Cao 2013 — which is an inference, not a measurement of the drug.
Cell, rodent, human — and where it stops
In cells. Fibrin and fibrinogen plate assays, where activity is reported in units per milligram of protein rather than as a concentration, because a mixture of five proteases has no single molecular weight to compute a molarity from Stephani 2023. In a fibroblast monolayer, reduced migration and adhesion with lower AP-1 and ICAM-1 signaling Nguyen 2023.
In rodents. An intra-abdominal adhesion model, in which lumbrokinase reduced post-surgical adhesion formation Nguyen 2023. Note the route and the target: this is peritoneal fibrin, adjacent to where the enzyme was placed. It is not evidence that a swallowed capsule reaches a coronary artery.
In people, twice, and both times without a placebo. Three hundred and ten patients after ischemic stroke were allocated 192 to standard treatment plus enteric-coated lumbrokinase and 118 to standard treatment alone, for 12 months, with fibrinogen, carotid intima-media thickness, plaque volume and NIH Stroke Scale scores recorded; the incidence of total vascular endpoints was 2.08% against 6.78%, an absolute reduction of 4.7 percentage points Cao 2013. Separately, 180 acute ischemic stroke patients were randomized open-label to standard therapy (aspirin 100 mg, atorvastatin 20 mg) with or without DLBS1033, a protein fraction from the same worm: NIH Stroke Scale improvement was −5.57 ± 2.16 against −3.64 ± 2.65 at discharge (P < 0.001) and −6.62 ± 2.64 against −5.14 ± 2.41 at day 30 (P = 0.001) Pinzon 2021.
The specific obstacle is not absorption. It is that the outcome hardest to bias is the one that did not move. In the DLBS1033 trial the ordinal scales — NIH Stroke Scale, Barthel Index, both scored by unblinded assessors in an open-label design — separated with P values below 0.01. The dichotomized functional outcome, modified Rankin Scale under 2 at day 30, was 86.7% against 80.0% and did not reach significance (P = 0.302) Pinzon 2021. That is the classic signature of assessor expectation rather than drug effect, and it is exactly why placebos exist. Neither human trial of an earthworm fibrinolytic used one.
The second obstacle is allocation. The stroke-prevention trial randomized 192 to 118, not 1:1 Cao 2013, and reported a 4.7 percentage-point difference in an endpoint that occurred in fewer than a dozen people in the smaller arm. An English-language reader also cannot audit the trial registration, the protocol or the analysis plan. This is real evidence and it is the best that exists; it is not evidence of the kind that would license the confidence with which this product is sold.
Lumbrokinase — which form, and does it matter
Four incompatible unit systems are on sale in the United States right now, and the label database shows all of them. Allergy Research Group's Lumbrokinase is listed as delayed-release vegicaps declaring 600,000 IU NIH Office of Dietary Supplements. Doctor's Best sells Lumbrokinase 20 mg declaring 720,000 units NIH Office of Dietary Supplements. Double Wood sells Lumbrokinase 150 mg declaring 10,000 FU per gram NIH Office of Dietary Supplements. Other listings in the same database declare 750,000 FU, 375 FU, 20,000 IU per unit mass, or no activity at all — only a milligram weight of powder.
Do the arithmetic and it stops being a labeling quibble. Doctor's Best puts 720,000 units into a 20 mg capsule, which is 36,000 units per milligram NIH Office of Dietary Supplements. Double Wood's 10,000 FU per gram is 10 FU per milligram NIH Office of Dietary Supplements. Those two numbers differ by three and a half orders of magnitude, and neither brand is lying: an IU here is not an International Unit in the pharmacopoeial sense — there is no reference standard for lumbrokinase to calibrate one against — and an FU is a fibrinolytic unit defined by whichever fibrin-plate assay the manufacturer chose, whose value depends on the clot preparation, the incubation temperature and the reference enzyme.
The published chemistry says the same thing from the other direction. Two fractions from one purification of one worm differed 1.5-fold in fibrinogenolytic activity per milligram Stephani 2023. If the same laboratory cannot get the same specific activity from two fractions, two companies buying from two suppliers cannot be assumed to have got the same enzyme.
What that means for the site's own card, precisely. The card gives 20–40 mg, roughly 300,000–600,000 IU, which implies about 15,000 IU per milligram. Doctor's Best's own label implies 36,000 units per milligram NIH Office of Dietary Supplements. The two figures cannot both describe the same enzyme at the same purity, so the milligram range and the IU range on any single label should be treated as two unrelated statements until the manufacturer names the assay.
Enteric coating is not a feature, it is the product. Almost every listing above says delayed-release or enteric-coated, and the one human prevention trial specifies enteric-coated capsules Cao 2013. A plain capsule delivers a protein to gastric acid, and a protein delivered to gastric acid is a peptide meal.
And DLBS1033 is not lumbrokinase. It is a separately characterized bioactive protein fraction from the same species, developed and trialed as a pharmaceutical in Indonesia Pinzon 2021. Quoting its stroke data on a lumbrokinase bottle is the same error as quoting a micronized fraction's data on a plain flavonoid capsule.
What would have to be true, and how you would know it was not
1. Fibrinogen, Activity (Clauss), mg/dL — down, and only if it started high. The mechanism is fibrinogen depletion and the one prevention trial reports fibrinogen as its proximal marker Cao 2013. Predict a fall of 30–60 mg/dL over 8 to 12 weeks from a baseline above 400 mg/dL. Draw it before the first capsule and at the same lab afterwards, fasting, because fibrinogen is an acute-phase protein and a cold or a dental abscess moves it more than any capsule will.
2. Prothrombin Time / INR — unchanged, and that is the trap. This enzyme acts on fibrin and fibrinogen, not on the vitamin K-dependent factors and not on factor Xa. So a normal INR and a normal anti-Xa level are not reassurance that you are not bleeding more easily on this product; they are measurements of pathways it does not touch. Predict INR at 1.0 ± 0.1 throughout, and treat that result as uninformative rather than as a green light.
3. The prediction that cuts against the product. In someone whose fibrinogen is already normal — say 250 mg/dL — predict no change in fibrinogen, no change in D-dimer beyond assay noise, and no change in any symptom, at any dose, at 12 weeks. A fibrinogen-depleting agent needs something to deplete, and the trial that supports this product enrolled stroke patients selected for elevated fibrinogen as a risk factor Cao 2013. Most people buying this are not that population.
4. hs-CRP, mg/L — unchanged. The mechanism is proteolysis of a clotting substrate, not suppression of interleukin-6 driven hepatic acute-phase synthesis. If hs-CRP falls and fibrinogen does not, something else changed in those twelve weeks — weight, alcohol, a resolved infection — and the capsule has been credited for it.
What will fool you: feeling better after a stroke or a viral illness. Both of the human trials measured recovery trajectories in people who were going to recover Cao 2013Pinzon 2021, and neither had a placebo arm to say how much of that recovery was the capsule.
What nobody has tested yet
Nobody has ever measured this protein in human blood. That is the missing experiment and it is now trivially runnable: the tryptic signature peptides of the purified fibrinolytic bands are published Stephani 2023, so a targeted LC-MS/MS assay can be built for them. Give ten fasted volunteers one enteric-coated 600,000 IU capsule, draw plasma at 0, 1, 2, 3, 4, 6, 8 and 12 hours, and look. If nothing appears, the entire oral rationale collapses and the fibrinogen changes need another explanation. If something appears, this becomes the first orally bioavailable therapeutic protease and that is a much bigger finding than a supplement.
Nobody has run a placebo. Both human trials are open-label Cao 2013Pinzon 2021, and in one of them the blinded-style dichotomous endpoint did not separate while the unblinded ordinal ones did Pinzon 2021. A 200-person, double-blind, placebo-controlled trial with fibrinogen at 12 weeks as the primary endpoint would cost very little and has never been attempted.
Nobody has cross-assayed the brands. Buy ten retail bottles, run one fibrin-plate assay against one reference clot, and publish activity per capsule beside the label claim. Given a 1.5-fold spread within a single purification Stephani 2023 and four incompatible unit systems on the shelf NIH Office of Dietary SupplementsNIH Office of Dietary SupplementsNIH Office of Dietary Supplements, the retail spread is likely to be far larger, and one afternoon of laboratory time would settle it.
And nobody has run it on top of an anticoagulant under measurement. The population most interested in this product is the population most likely to already be taking apixaban or clopidogrel, and there is not one controlled dataset on that combination. The experiment is a small pharmacodynamic study with thromboelastography or a fibrinolysis-sensitive assay, because the standard coagulation panel is blind to this mechanism by construction.
Lumbrokinase — its own safety story, not its category's
The bleeding risk here is real, and its specific feature is that routine monitoring cannot see it. Warfarin is monitored by INR. Direct oral anticoagulants are monitored, when they are monitored at all, by drug-specific anti-Xa. Antiplatelet drugs are monitored by platelet function assays. A fibrinolytic acts downstream of all three: it degrades the clot that has already formed. Add it to any of them and the additive risk is invisible on every test the prescribing doctor will order. That is a different and worse situation from most supplement interactions, where the interaction at least shows up somewhere.
The antiplatelet arm is inside the same capsule. Earthworm protein fractions are reported to inhibit platelet aggregation and reduce blood viscosity alongside their fibrinolytic activity Zhu 2024Pinzon 2021. So this is not a single-mechanism product being stacked on an antiplatelet drug; it is a two-mechanism product, and the two mechanisms compound.
Surgery, dental work and the arithmetic of a washout. Fibrinogen has a plasma half-life of roughly four days, so recovery from a fibrinogen-lowering agent is a synthesis problem rather than a clearance problem. Two weeks is about three and a half fibrinogen half-lives, which is why two weeks is the right answer and one weekend is not.
It is an animal protein extract, and that carries an allergy risk nothing else in this cohort has. Earthworm proteins are documented occupational and environmental allergens in people who handle bait and vermicompost. An enteric-coated capsule delivers intact foreign protein to the small intestine by design Cao 2013. Anyone with a history of anaphylaxis to an invertebrate protein should not be experimenting here, and a first dose is not a thing to take alone at night.
The sourcing question nobody asks. Earthworms concentrate cadmium and lead from the soil they live in — that is precisely why they are used as sentinel organisms in soil contamination monitoring. A product made by extracting protein from farmed worms inherits whatever their substrate contained. This is a reasoned concern rather than a published finding for these specific products, and the way to resolve it is to ask the manufacturer for a heavy-metals certificate of analysis by lot. Most will not have one.
And the boundary. There is no controlled human data at all for the indication this is most discussed for — persistent microclots after viral illness. The two trials that exist are in ischemic stroke Cao 2013Pinzon 2021. Using a potent fibrinolytic for an indication with no trial, no dose-finding and no monitoring, in a person who may be on an anticoagulant, is a physician's decision and it is not a close call.
Sources read for this page
- Cao YJ, et al. Oral fibrinogen-depleting agent lumbrokinase for secondary ischemic stroke prevention: results from a multicenter, randomized, parallel-group and controlled clinical trial.. Chinese Medical Journal 2013 · PMID 24229674
- Pinzon RT, et al. Effect of DLBS1033 on Functional Outcomes for Patients with Acute Ischemic Stroke: A Randomized Controlled Trial.. Stroke Research and Treatment 2021 · PMID 33927846
- Stephani L, et al. Purification and proteomic analysis of potent fibrinolytic enzymes extracted from Lumbricus rubellus.. Proteome Science 2023 · PMID 37158880
- Nguyen QTT, et al. Lumbrokinase, a Fibrinolytic Enzyme, Prevents Intra-Abdominal Adhesion by Inhibiting the Migrative and Adhesive Activities of Fibroblast via Attenuation of the AP-1/ICAM-1 Signaling Pathway.. BioMed Research International 2023 · PMID 36685669
- Hazare C, et al. Diverse origins of fibrinolytic enzymes: A comprehensive review.. Heliyon 2024 · PMID 38434287
- Zhu Z, et al. Pharmacological effects of bioactive agents in earthworm extract: A comprehensive review.. Animal Models and Experimental Medicine 2024 · PMID 38957072
- NIH Office of Dietary Supplements. Dietary Supplement Label Database record 183657 - Lumbrokinase (Allergy Research Group), 30 delayed-release vegicaps, 600,000 IU.. NIH DSLD label 183657
- NIH Office of Dietary Supplements. Dietary Supplement Label Database record 63230 - Lumbrokinase 20 mg (Doctor's Best), 60 veggie caps, 720,000 units.. NIH DSLD label 63230
- NIH Office of Dietary Supplements. Dietary Supplement Label Database record 315659 - Lumbrokinase 150 mg (Double Wood Supplements), 120 capsules, 10,000 FU/gram.. NIH DSLD label 315659
How you would know if it worked
A fibrinolytic enzyme has a real read-out, and it is the one the microclot conversation tends to skip. D-dimer is a fibrin degradation product, so systemic fibrin breakdown is exactly where it appears; fibrinogen is the substrate being consumed. Both have to be drawn before the first dose, because a single D-dimer taken afterwards answers nothing at all. Ask for a CBC alongside them, because the hemoglobin and platelet count are the numbers that would show a bleeding problem starting: this is considerably more potent on fibrin in vitro than nattokinase, the trial base is almost entirely Chinese-language with reporting below Western standards, and no large independent Western randomized trial exists. Stop 2 weeks before any surgery or dental extraction, and if you take an anticoagulant or antiplatelet, this is a physician's decision rather than a purchase.
- D-Dimer Retest: Only as directed clinically.
- Fibrinogen Activity Retest: Every 6–12 months if elevated.
The cheapest panel carrying D-Dimer and at least one other of these is Clotting & Advanced Cardiac Risk, at $504 — the panel is named for a different question, and the marker is the same marker. That is the whole cost of finding out.
Draw before you start, not after. A result with nothing to compare it to answers nothing.
Lumbrokinase — safety & side effects
- Well tolerated; GI upset and headache are the usual reports.
- Fibrinolytic, like nattokinase, and generally considered more potent. Additive bleeding risk with anticoagulants and antiplatelets — warfarin, apixaban, rivaroxaban, clopidogrel, and aspirin at any dose. The interaction is pharmacodynamic rather than metabolic, so it does not show up as a changed drug level; it shows up as bruising, nosebleeds or a raised INR.
- Stop at least two weeks before any surgery or dental extraction. Surgeons ask about prescription blood thinners and rarely about supplements, so this one is on you to volunteer. Avoid with any bleeding disorder or active bleeding.
- Long-term safety has not been characterized — the trials run weeks to months, not years. That is a real limit on what anyone can tell you about daily use for a decade. The human trial base is smaller than nattokinase's.
Not medical advice. If you take prescription medication or have a diagnosed condition, check this against it with a pharmacist or doctor — pharmacists are underused and free.
- When to take it, and what to take it with
- Which form actually absorbs
- Who it's worth it for
- Coach Cam's stacks and notes
- Fasted or with food, and when in the day
- Morning or night, and why that window
- Around training, or deliberately away from it
- What it must not share a window with
Everything above is free and stays free. Skool is where it becomes a plan — Lumbrokinase in an order, with the rest of what you're running.
Unlock in Skool — $10/mo →Bloodwork to run alongside Lumbrokinase
Baseline first, then again at 8–12 weeks.
| Marker | What it’s watching for |
|---|---|
| ApoB (Apolipoprotein B) | Counts the particles that cause plaque — a normal LDL-C can hide risk |
| Lipoprotein(a) — Lp(a) | Genetic, largely unmodifiable, and worth knowing once in your life |
| hs-CRP (High-Sensitivity C-Reactive Protein) | The inflammatory half of cardiovascular risk |
| Lipid Panel (Cholesterol, HDL, LDL, Triglycerides) | The standard baseline these are usually aimed at |
The Real Cardiovascular Risk panel covers these in one order — 9 markers, $187.60 with the discount applied.
Check results you already have → · All 103 markers A–Z
Lumbrokinase — frequently asked questions
What is Lumbrokinase?
A fibrinolytic enzyme complex from earthworms. Distinct from nattokinase and considerably more potent on fibrin in vitro, which cuts both ways.
What is the suggested dose of Lumbrokinase?
20–40 mg (roughly 300,000–600,000 IU) daily on an empty stomach. This is a general reference for education only — statements have not been evaluated by the FDA and this is not medical advice.
Where can I find Lumbrokinase dosing and the full breakdown?
The suggested dose and the full evidence — clinical, correlative and theoretical — are on this page. What's inside Skool is when to take it, which form actually absorbs, the brand worth buying and Coach Cam's stacks.
Where can I buy Lumbrokinase?
Coach Cam sources Lumbrokinase from vetted, top-rated brands on iHerb — use the buy link on this page.
Lumbrokinase inside a finished plan
One arm of 1 Protocol Blueprint, free to read in full.
What Lumbrokinase is used for
Lumbrokinase appears under 1 goal in the goal router.
Related Cardiovascular supplements
Where this goes next
Lumbrokinase is the residual risk arm of this plan. The page above is the free breakdown of one compound; the plan it belongs to — the dosing, the order to correct things in, the week-by-week schedule and what to retest — is a lesson inside Skool.