Medicinal Mushrooms and Blood Sugar: What the Research Actually Tested — and What Came Out

In brief5 points · 1-minute read

• Reishi was tested in 84 type 2 diabetics across three arms for 16 weeks, and neither HbA1c nor fasting glucose changed.
• A 2025 meta-analysis of 17 trials and 971 participants found no effect of reishi on fasting glucose, with very low certainty on every outcome.
• Lion's mane, turkey tail and chaga have never been measured for glucose, insulin or HbA1c in humans; the link exists only in rodents.
• The study cited as maitake in prediabetes used Grifola gargal, a different species, and only triglycerides fell in its 10 human participants.
• The single positive human finding belongs to Agaricus blazei, a mushroom we do not sell, with HOMA-IR 3.6 versus 6.6 in a small unreplicated trial.

Questions about this article? Ask the Triterra bot ›

Medicinal mushrooms — reishi (Ganoderma lucidum, “lingzhi” in traditional Chinese medicine), lion’s mane (Hericium erinaceus, “yamabushitake” in Japan), cordyceps (“dong chong xia cao”), turkey tail (Trametes versicolor) and maitake (Grifola frondosa) — are sold as dietary supplements and quoted online, again and again, as “blood-sugar lowering”. In humans, for the four mushrooms we grow, there is not one randomized controlled trial that showed a significant drop in HbA1c or in fasting glucose.

The strongest evidence that exists — a Cochrane review, a double-blind trial and a GRADE-rated meta-analysis — points to zero.

What does exist: on reishi — measured evidence of no effect (Cochrane 2015, a double-blind trial in 2016, a meta-analysis in 2025). On cordyceps — one arm of that same trial, and a fermented drink in healthy people. On lion’s mane, turkey tail and chaga — no human blood-sugar measurement at all. On maitake — a case letter with no abstract, and a study on a different species that gets cited as maitake. And the only positive finding belongs to a mushroom we do not sell.

Do medicinal mushrooms lower blood sugar?

Not according to what was measured in humans. Reishi was tested in randomized controlled trials in type 2 diabetics — and lowered neither HbA1c nor fasting glucose. Cordyceps was tested once, in a combined arm, with the same result. Lion’s mane, turkey tail and chaga were never measured for blood sugar. The only positive finding belongs to Agaricus blazei.

It matters to separate two kinds of “no”. For reishi and cordyceps there is evidence of no effect: it was measured — and came out zero. For lion’s mane, turkey tail and chaga there is absence of measurement: nobody checked. The table sums it up mushroom by mushroom.

Why are beta-glucans linked to blood sugar at all?

Because beta-glucans from oats and barley have been studied for post-meal blood sugar — and from there the generalization leaked onto mushrooms. But fungal beta-glucan is a different molecule, and in humans it has not been measured against glucose: there is not one meta-analysis on mushroom beta-glucan and glycemic control. The direct evidence comes from mice.

The most direct test of fungal beta-glucan in humans was done in 2021 with shiitake: 52 participants with mildly elevated cholesterol consumed 3.5 grams of fungal beta-D-glucan a day versus placebo for 8 weeks — and no significant difference was found on any lipid measure, nor any inflammatory or immune response. The study measured lipids, not glucose, but it demonstrates that what oat beta-glucan does, mushroom beta-glucan does not necessarily do — even at a high dose.

The real sources of the claim are rodents: in 2011, reishi polysaccharides lowered fasting glucose in diabetic mice; in 2013, an alpha-glucan from maitake lowered glucose and glycated hemoglobin in mice. In humans it did not repeat. On the difference between beta-glucan and “total polysaccharides”: beta-glucan versus polysaccharides.

What did the Cochrane review find?

The 2015 Cochrane review pooled 5 randomized trials with 398 participants; 3 of them contributed data to the analysis, and all of their participants were type 2 diabetics taking 1.4–3 grams of reishi a day for 12–16 weeks. HbA1c: WMD -0.10% (95%CI -1.05 to 0.85; 130 participants) — not significant, statistically or clinically. Fasting glucose: 0.30 mmol/L (95%CI -0.95 to 1.55).

A precision most people who cite it miss: this is a review of reishi for the treatment of cardiovascular risk factors, not “on diabetes” — all the participants analyzed simply happened to be type 2 diabetics. And it holds a trap: post-meal blood sugar came out in two directions — “two hours after a meal” in favor of placebo (WMD 0.7 mmol/L; 95%CI 0.29 to 1.11), and “glucose area under the curve at the fourth hour” in favor of reishi (WMD -49.4 mg/dL/h). Whoever quotes only the second is distorting; the authors themselves wrote that the meaning of the effect is unclear.

What did the largest double-blind trial find?

A 2016 randomized, double-blind, placebo-controlled trial: 84 participants with type 2 diabetes and metabolic syndrome in three arms — reishi, reishi + cordyceps, placebo — 3 grams a day, 16 weeks. The two intervention arms were pooled. Result: HbA1c 0.13% (95%CI -0.35 to 0.60; p=0.60), fasting glucose 0.03 mmol/L (95%CI -0.90 to 0.96; p=0.95). Zero.

Note the detail almost everyone who cites this study gets wrong: not all 84 participants received reishi. They were split into three groups, and only two of them received extract. If you see someone write “84 participants took 3 grams of reishi” — they did not read the abstract. The secondary outcomes were unchanged too, with no increase in the risk of adverse events. The authors’ conclusion: the evidence does not support the use of reishi for cardiovascular risk factors in diabetes or metabolic syndrome.

And what does the 2025 meta-analysis say?

A GRADE-rated meta-analysis pooled 17 randomized controlled trials of reishi, 971 participants, doses of 200–11,200 mg a day, durations of 1–24 weeks. On fasting glucose — no significant effect. Nor on body-fat percentage, waist circumference, blood pressure, lipid profile, inflammatory markers or liver enzymes. Certainty of evidence: “very low” on every outcome.

And yet it must not be written that it “found nothing” — because it did find something. A significant drop in BMI (WMD -0.43; 95%CI -0.77 to -0.10; p=0.011), in creatinine (WMD -0.14; 95%CI -0.27 to -0.02; p=0.028) and in heart rate (WMD -3.92; 95%CI -7.45 to -0.40; p=0.029), and a rise in the antioxidant enzyme GPx (WMD 2.29; 95%CI 1.67 to 2.92; p<0.001).

But whoever quotes the BMI drop must quote the rating too: “very low” on every single one of the outcomes, the positive ones included. And on a page about blood sugar, the number that matters is fasting glucose — no effect. A 2021 pilot in women with fibromyalgia (6 grams a day of reishi powder, 6 weeks) — no significant difference in fasting glucose.

“Maybe the dose was too low?”

No. The double-blind trial tested 3 grams a day for 16 weeks. The meta-analysis covered doses from 200 mg up to 11,200 mg a day, and durations from one week to 24 weeks. The fibromyalgia pilot tested 6 grams a day. Across that whole range, in humans, fasting glucose did not move. That is not an excuse the data allow.

And what about the “positive” finding on reishi and insulin sensitivity? One trial from 2011: 26 patients with borderline-high blood pressure or cholesterol — not diabetics — 1.44 grams of reishi a day versus placebo, 12 weeks, crossover; 23 evaluable. Insulin and HOMA-IR were lower after reishi; BMI and blood pressure did not change; a significant carry-over effect prevented a full analysis. The authors’ conclusion: “reishi may have mild antidiabetic effects” — and that is where the phrase that recurs in the literature and at MSKCC comes from. A single study, 23 participants, not in diabetics.

What has been measured on lion’s mane and blood sugar?

In humans — nothing. PubMed holds 7 clinical trials of lion’s mane, all on cognition, mood and depression, and not one of them measured glucose, insulin or HbA1c. A direct search for “lion’s mane + fasting glucose or HbA1c” returns 2 results — both in rodents. That is an absence, not negative evidence.

What was measured: in 2009, 30 adults with mild cognitive impairment took 3 grams a day for 16 weeks — the cognitive score rose, and “laboratory tests showed no adverse effect”. In 2023, 41 healthy adults received 1.8 grams a day — a faster Stroop after a single dose (p=0.005). Neither has a metabolic endpoint.

The link to blood sugar exists only in the laboratory: in 2024, erinacines lowered fasting glucose and insulin in transgenic mice; in 2018, fermented lion’s mane juice improved insulin in diabetic rats. Mice, rats and fermented juice — not an extract, and not humans.

What has been measured on cordyceps?

The only randomized controlled evidence of cordyceps on a human glycemic endpoint is the reishi + cordyceps arm of the 2016 trial: 3 grams a day, 16 weeks, type 2 diabetics — HbA1c 0.13% (p=0.60), fasting glucose 0.03 mmol/L (p=0.95). Zero. A fermented cordyceps drink tested in 2024 in healthy people: “blood sugar parameters were not different between the groups”.

In the 2024 trial, healthy adults (10 of each sex in each group) drank, for 8 weeks, a fermented beverage containing 2.85 mg of cordycepin versus placebo: NK-cell activity rose — and the abstract states it explicitly: “blood sugar, lipid and safety parameters were not different between the groups”.

From here, two citation traps. “Cordyceps lowers blood sugar in kidney patients” — the 2014 Cochrane review (22 studies, 1,746 participants) measured creatinine and urinary protein, and not a single glycemic endpoint. “Cordyceps reduced hyperglycemia in kidney-transplant recipients” — there it was given as an adjunct to immunosuppressive therapy and compared with azathioprine; immunosuppressants are known to raise blood sugar, and the comparison is with a drug, not with a placebo. The only finding in diabetics themselves concerns a complication: in 2015, in 120 type 2 diabetics with renal insufficiency undergoing angiography, contrast-induced nephropathy was less frequent in the cordyceps groups (P<0.05). More: mushrooms and kidney health.

And what about turkey tail and chaga?

Here there is not even “weak evidence” — there is a complete absence of measurement. Chaga (Inonotus obliquus) has no clinical trial tagged on PubMed — not on blood sugar, not on anything else. Turkey tail has 4 randomized controlled trials in all of PubMed, and in none of them was glucose, insulin or HbA1c measured.

Here are all four, so you can check for yourself. 2026: 90 participants, polypore mushroom mycelium (including Trametes versicolor) as an adjunct to COVID-19 vaccination — no adverse events, kidney and liver function unchanged. 2014: 24 volunteers, PSP from turkey tail versus amoxicillin — the only endpoint: the bacterial composition of stool. 2021: 91 women carrying HPV, a vaginal gel — a normal smear in 78.0% versus 54.8%. 2014: oral HPV, turkey tail + reishi — clearance in 88% versus 5%. Vaccine, microbiome, HPV, HPV. Blood sugar — zero times.

A targeted search for “turkey tail or PSK or PSP + glucose or diabetes + clinical trial” returned one result — a false positive. And on chaga — zero. That is one of the reasons we do not grow it: chaga — why we do not grow it.

What is actually known about maitake?

Less than it seems. The study cited online as “maitake in prediabetes” was done on Grifola gargal — a different species in the same genus, not Grifola frondosa. Even there, the only effect in its 10 human participants was a drop in triglycerides; the glucose reductions were measured in mice. On maitake itself: a case letter from 2001 — and mice.

In 2020 a paper was published in which 10 participants with prediabetes consumed 9.2 grams a day of a hot-water extract for 4 weeks. The paper’s own title declares the species: Grifola gargal; the genus is shared, and so in blogs it turns into “maitake”. Wrong twice over: it is not Grifola frondosa, and in the humans the only significant effect was a drop in triglycerides — glucose fell in KK-Ay and ob/ob mice. Whoever cites this study as proof that maitake lowers blood sugar — did not read it.

And on the real Grifola frondosa? A 2001 letter to the editor in Diabetic Medicine, one page, whose abstract is not available on PubMed; what is documented is the publication type — a case report — and keywords that include glibenclamide and hypoglycemia. Any number quoted “from the Konno study” cannot be verified.

The third trap: “maitake improved insulin resistance in PCOS”. In 2010, 72 patients were randomized to a maitake fraction or to clomiphene citrate; ovulation by cycles was 41.7% versus 69.9% (p=0.0006) in favor of the drug. The outcome measure was ovulation on ultrasound; insulin resistance appeared in the rationale only. What remains is mice — and from there, “maitake protects beta cells”.

The only positive finding — and whom does it belong to?

To Agaricus blazei — a mushroom we neither grow nor sell. In 2007, 72 type 2 diabetics on metformin and gliclazide for over six months received, on top of their drugs, 1,500 mg a day of Agaricus blazei extract or placebo for 12 weeks. HOMA-IR at the end: 3.6 (SD 2.5) versus 6.6 (SD 7.4), p=0.04. Adiponectin: +20.0% versus -12.0%, p<0.001.

We write this out in full even though it sells us nothing — precisely because of that: every “no evidence” on this page rests on our not having filtered for convenience. The finding: a randomized, double-blind, placebo-controlled (cellulose) trial, drawn from 536 registered diabetics, 29 versus 31 in the analysis.

And now the weaknesses, with the same honesty: the standard deviation in the placebo group is 7.4 on a mean of 6.6. A small sample, a single center, 12 weeks, never replicated — a search for “Agaricus blazei + insulin resistance + clinical trial” returns 2 results: this one, and one in calves. Real positive evidence, and fragile. And what it teaches: when something is measured positive in humans, this is what it looks like — with a p, with an SD, with participants in each arm. When that is missing for reishi or maitake, it is not because nobody looked.

What about safety and drug interactions?

In the randomized trials of reishi and cordyceps, no hypoglycemia was observed. Per Memorial Sloan Kettering, reishi’s documented interactions are with anticoagulants (bleeding risk), immunosuppressants, and in vitro with CYP450 enzymes — with no section on diabetes medication. Even so: if you take diabetes medication, talk to your physician before any supplement, and never change, reduce or stop drug treatment on your own.

What was measured: in the Cochrane review, reishi takers had a 1.67-fold risk of an adverse event (95%CI 0.86 to 3.24) — not significant, and not serious. With lion’s mane and turkey tail, laboratory tests and kidney and liver function did not change. But “safe” is not “completely safe”: a 2007 case report described fatal fulminant hepatitis attributed to reishi powder; in both documented cases the patients had previously taken boiled lingzhi without toxicity, and were harmed 1–2 months after switching to powder — alongside other medications.

What the “Reishi Mushroom” page at Memorial Sloan Kettering says (updated 5.5.2026), exactly and no more. Reported adverse effects: nausea, insomnia, dry mouth, constipation, itching and dizziness. Case reports: hepatotoxicity (with one fatal case), chronic diarrhea, hypereosinophilia with liver nodules, and acute neurotoxicity in a child. Interactions: anticoagulants and antiplatelets — reishi may increase bleeding risk (an explicit recommendation to consult if you are on blood thinners such as warfarin); immunosuppressants — reishi may enhance the immune response and interfere with treatment; CYP450 — in vitro, reishi polysaccharides inhibited CYP2E1, CYP1A2 and CYP3A, and the clinical significance is unknown; and a possible interaction with chemotherapy that depends on free radicals. The page has no section on diabetes medication, and we are not inventing one. More on do medicinal mushrooms interact with medications.

Why this is the most sensitive topic on the page: diabetes medication prevents real damage — to the kidneys, the eyes, the nerves, the heart — and a supplement that lowered blood sugar in no trial is not a substitute for it. Do not reduce a dose “because you are taking a mushroom”, and if you chose to try a supplement — ours included — tell your physician and keep measuring.

What we actually measure in our bottle

After all these chapters, it is fair to ask: so what do you actually know about what you sell? Here is the answer — and it is about the quality of the extract only, because that is the one thing we have laboratory numbers for.

Everything starts with fresh fruiting body from our farm in the Galilee — not mycelium grown on grain, and not imported powder that someone else dried somewhere else. Every bottle starts in our greenhouse, not in a shipping container. The mushroom goes from harvest to extraction with no drying step; as far as we know, we are among the very few in the world who work this way. The reason is simple: every drying step is a step where something is lost, and we do not want to guess how much. From there — a triple extraction, whose alcohol stage alone runs 7 weeks. The extraction ratios, on a fresh-mushroom basis: reishi 1:3, cordyceps 1:2.5, lion’s mane 1:2, and turkey tail + reishi 1:3. Alcohol in the finished extract: 32%.

And then the one number we are willing to sell on: beta-glucan, tested at TÜV Austria on the finished extract — not on the raw material — on a dry basis. And beside it a number no less important: alpha-glucan, which is starch, was not detected in any of the extracts. That is the chemical proof there is no grain in the bottle — and the number was measured on what you drink, not on what we harvested. The market has a ladder of three rungs: mycelium-on-grain (usually below 7% beta-glucan, with starch hiding under the heading “total polysaccharides”), imported dried fruiting body, and fresh fruiting body from the farm. We are on the third rung, and this is what the laboratory measured:

All our tests are public, and you can read them yourself on the lab results page, report after report. That is what we know how to measure and prove about what is in the bottle. What it will do for blood sugar was not measured — so it is not written.

What this page does not say — and what we do not claim

We do not claim that any mushroom extract — ours or anyone else’s — lowers blood sugar, “balances” blood sugar, improves insulin sensitivity, prevents diabetes or treats it. We do not claim our extracts are “suitable for people with diabetes” — the bottle chapter speaks about extract quality only. Nor do we claim the opposite: for lion’s mane, turkey tail and chaga, “no evidence” is not “proven not to help” — there, nobody measured. For reishi and cordyceps they did measure, and it came out zero. And we do not end with “future research may show” — because that is not what the sources say.

If your blood sugar is high — what do you actually do?

See a physician. Blood sugar levels are a matter for medical monitoring and treatment, and diabetes has treatments that were measured and work. No supplement — ours included — is a substitute for them. If you are already being treated: do not change, reduce or stop drug treatment on your own, and tell your physician about any supplement you are considering.

And there is a difference between a capsule and a plate. In 2023, 60 adults without cardiometabolic disease ate, for 8 weeks, a full Mediterranean-style diet with or without 84 grams of mushrooms a day: fasting glucose -2.9±1.18 mg/dL versus +0.6±1.10 (p=0.034); cholesterol fell in both groups — thanks to the diet. Small, in healthy people, and food — not a capsule and not diabetics. But it is the only example on this page where “mushrooms” and “fasting glucose” meet in humans with a significant p.

The bottom line

For the four mushrooms we grow, in humans, there is not one randomized controlled trial that showed a drop in HbA1c or in fasting glucose. Reishi was tested properly — Cochrane, a double-blind trial in 84 participants across three arms, a meta-analysis of 971 — and came out zero. Cordyceps was tested once, and came out zero. Lion’s mane, turkey tail and chaga were not measured at all. Maitake rests on a case letter with no abstract and on a study of a different species. The only positive finding belongs to Agaricus blazei, in a small sample never replicated. In our bottle, beta-glucan was measured; what it will do for blood sugar — was not measured, and so is not written.

Frequently asked questions

Does reishi lower blood sugar?

Not according to what was measured. A Cochrane review pooled three trials in type 2 diabetics (1.4–3 grams a day, 12–16 weeks) and showed an HbA1c change of just -0.10%, with a confidence interval that crosses zero comfortably. A randomized double-blind trial from 2016, 84 participants in three arms, found a difference of 0.13% in HbA1c and 0.03 mmol/L in fasting glucose — that is, zero.

Is there any research at all on lion’s mane and blood sugar in humans?

No. The seven existing clinical trials of lion’s mane deal with cognition, mood and depression, and not one of them measured glucose, insulin or HbA1c. The only two studies that link it to glucose were done in transgenic mice and in diabetic rats — and in the rats it was a fermented juice, not an extract. That is an absence of measurement, not evidence either way.

I heard maitake helps with blood sugar. Is that true?

What exists on maitake is a 2001 case letter whose abstract is not available, and studies in mice. The study most often cited as “maitake in prediabetes” was actually done on Grifola gargal — a different species — and in its ten human participants only triglycerides fell; glucose fell in mice. The 2010 PCOS trial measured ovulation, not insulin resistance.

If I am on metformin, will a mushroom help me on top of it?

The only evidence of that kind is on Agaricus blazei — not on any of our four mushrooms: 1,500 mg a day on top of gliclazide and metformin, 12 weeks, HOMA-IR 3.6 versus 6.6 on placebo, p=0.04. A sample of 29 versus 31, a single center, never replicated. And in any case: do not change your drug treatment without your physician.

Is there a risk of blood sugar dropping too low (hypoglycemia)?

As far as what was measured — no hypoglycemia was observed in the randomized trials of reishi or cordyceps, and in the 2016 trial there was no overall increase in adverse events. The only report linking a mushroom to hypoglycemia is a 2001 case letter on maitake with glibenclamide, whose content is not accessible. Anyone taking diabetes medication should talk to their physician before a supplement, and keep measuring.

Maybe the dose in the studies was too low?

No. The double-blind trial tested 3 grams a day for 16 weeks, and the fibromyalgia pilot 6 grams a day. The 2025 meta-analysis pooled 17 trials with doses from 200 mg up to 11,200 mg a day, for one week to 24 weeks — and fasting glucose was unaffected across the whole range. The certainty of evidence was rated “very low” on every outcome.

Does reishi get along with my medications?

According to Memorial Sloan Kettering: reishi may increase bleeding risk with anticoagulants and antiplatelets such as warfarin; may enhance the immune response and interfere with immunosuppressant treatment; and in vitro its polysaccharides inhibited CYP2E1, CYP1A2 and CYP3A, with the clinical significance unknown. There is no section there on diabetes medication. Bring your list of medications to your physician before you start.

So does eating mushrooms help at all?

That is an entirely different question from an extract. In a 2023 randomized trial in 60 healthy adults, adding 84 grams of mushrooms a day (white button and oyster) to a full Mediterranean-style diet changed fasting glucose by -2.9 mg/dL versus +0.6 in the control (p=0.034). Small, in participants without diabetes, and food within a healthy diet — not a capsule. Cholesterol fell in both groups, thanks to the diet.

Scientific sources (peer-reviewed)

• Cochrane review of reishi and cardiovascular risk factors: 5 trials, 398 participants; 3 analyzed, all type 2 diabetics, 1.4–3 grams a day, 12–16 weeks. HbA1c WMD -0.10% (95%CI -1.05 to 0.85; 130 participants); fasting glucose WMD 0.30 mmol/L (95%CI -0.95 to 1.55; 84 participants); post-meal glucose — conflicting results; adverse events RR 1.67 (95%CI 0.86 to 3.24), none serious — Klupp NL, et al. The Cochrane Database of Systematic Reviews, 2015. View on PubMed
• Randomized double-blind trial: 84 participants with type 2 diabetes and metabolic syndrome in three arms (reishi / reishi + cordyceps / placebo), 3 grams a day, 16 weeks; the two intervention arms pooled. HbA1c 0.13% (95%CI -0.35 to 0.60; p=0.60); fasting glucose 0.03 mmol/L (95%CI -0.90 to 0.96; p=0.95); no increase in adverse-event risk — Klupp NL, et al. Scientific Reports, 2016. View on PubMed
• GRADE meta-analysis: 17 randomized trials, 971 participants, 200–11,200 mg a day, 1–24 weeks. No effect on fasting glucose, lipids, blood pressure, inflammation or liver enzymes; drops in BMI (WMD -0.43; p=0.011), creatinine (-0.14; p=0.028) and heart rate (-3.92; p=0.029), a rise in GPx (2.29; p<0.001); “very low” certainty of evidence on every outcome — Jafari A, et al. Food Science & Nutrition, 2025. View on PubMed
• 26 patients with borderline-high blood pressure or cholesterol, 1.44 grams of reishi a day versus placebo, 12 weeks, double-blind crossover; 23 evaluable. Insulin and HOMA-IR lower after reishi; BMI and blood pressure unchanged; a significant carry-over effect prevented a full analysis of blood lipids — Chu TTW, et al. The British Journal of Nutrition, 2011. View on PubMed
• Randomized double-blind pilot in women with fibromyalgia: 6 grams a day of reishi fruiting-body powder versus carob flour, 6 weeks — no significant difference in fasting glucose, cholesterol, triglycerides, weight, fat mass, waist-to-hip ratio or blood pressure — Pazzi F, et al. Nutricion Hospitalaria, 2021. View on PubMed
• Case report: fatal fulminant hepatitis attributed to reishi powder (2005), after a first case in Hong Kong (2004); both patients had previously taken boiled lingzhi without toxicity, and the event occurred 1–2 months after switching to powder, alongside other medications — Wanmuang H, et al. Journal of the Medical Association of Thailand, 2007. View on PubMed
• Diabetic (STZ) mice, 8 per group, reishi polysaccharides at 50 or 150 mg/kg, 28 days: fasting glucose fell, insulin and body weight rose, blood lipids improved — in rodents — Li F, et al. International Journal of Molecular Sciences, 2011. View on PubMed
• Fermented cordyceps beverage with 2.85 mg of cordycepin versus placebo, 8 weeks, healthy adults (10 of each sex in each group): NK activity rose, IL-1β fell in men (p=0.049), IL-6 fell in women (p=0.047); “blood sugar, lipid and safety parameters were not different between the groups” — Ontawong A, et al. Scientific Reports, 2024. View on PubMed
• Cochrane review of cordyceps in chronic kidney disease: 22 studies, 1,746 participants; serum creatinine MD -60.76 µmol/L (14 studies, 987), creatinine clearance +9.22 mL/min, urinary protein -0.15 g per 24 hours; high risk of bias in 4 and unclear in 18; no glycemic endpoint — Zhang HW, et al. The Cochrane Database of Systematic Reviews, 2014. View on PubMed
• 120 type 2 diabetics with renal insufficiency undergoing coronary angiography, three groups (41 / 39 / 40): contrast-induced nephropathy in 11 of 120 (9.17%), less frequent in the cordyceps groups (P<0.05), and especially in the intensive group (P<0.01) — a complication, not blood sugar — Kai Z, et al. Journal of Traditional Chinese Medicine, 2015. View on PubMed
• Systematic review of 9 trials of cordyceps as an adjunct in kidney-transplant recipients: versus azathioprine — similar rejection and survival, better kidney function and a lower incidence of hyperuricemia, hyperlipidemia, hyperglycemia and liver injury; adverse events not properly monitored in most trials — Ong BY, Aziz Z. Complementary Therapies in Medicine, 2016. View on PubMed
• 30 adults aged 50–80 with mild cognitive impairment, 3 grams a day of lion’s mane powder (4×250 mg ×3 a day) versus placebo, 16 weeks: cognitive score rose; “laboratory tests showed no adverse effect”; no glycemic endpoint — Mori K, et al. Phytotherapy Research, 2009. View on PubMed
• 41 healthy adults aged 18–45, 1.8 grams of lion’s mane a day versus placebo, double-blind: faster Stroop 60 minutes after a single dose (p=0.005), a trend toward lower subjective stress after 28 days (p=0.051); no metabolic endpoint — Docherty S, et al. Nutrients, 2023. View on PubMed
• APP/PS1 transgenic mice under metabolic stress: erinacines from lion’s mane lowered fasting glucose and insulin and reduced amyloid plaque load — in rodents — Bui VT, et al. CNS Neuroscience & Therapeutics, 2024. View on PubMed
• Diabetic (STZ) Wistar rats given lion’s mane juice fermented with Lactobacillus fermentum HP3: weight, insulin and recovery from hyperglycemia improved; HbA1c “different” (not quantified) — fermented juice, in rodents — Chaiyasut C, et al. Journal of Evidence-Based Integrative Medicine, 2018. View on PubMed
• 90 participants, polypore mushroom mycelium (including Trametes versicolor) as an adjunct to COVID-19 vaccination, 8 capsules of 500 mg ×3 a day, 4 days (52 versus 38): no adverse events, kidney and liver function unchanged, 100% completed; no glycemic endpoint — Saxe G, et al. BMC Immunology, 2026. View on PubMed
• 24 healthy volunteers, PSP from turkey tail versus amoxicillin versus control, 8 weeks (22 completed): consistent microbiome changes in keeping with prebiotic activity; the only endpoint — the bacterial composition of stool — Pallav K, et al. Gut Microbes, 2014. View on PubMed
• 91 women carrying HPV with minor smear changes, a turkey tail-based vaginal gel versus watchful waiting: normal smear in 78.0% and 84.9% (3 and 6 months) versus 54.8% and 64.5%; 7 possible adverse events, 6 mild or moderate; no glycemic endpoint — Serrano L, et al. Journal of Lower Genital Tract Disease, 2021. View on PubMed
• 61 oral HPV16/18 carriers out of 472 screened: 20 in group 1 and 41 in group 2 (turkey tail + reishi), two months; viral clearance 5% versus 88% (P<0.001); no glycemic endpoint — Donatini B. International Journal of Medicinal Mushrooms, 2014. View on PubMed
• Letter to the editor / case report: “A possible hypoglycaemic effect of maitake mushroom on Type 2 diabetic patients” — abstract not available on PubMed; MeSH terms include Glyburide and Hypoglycemia; no verifiable number comes from it — Konno S, et al. Diabetic Medicine, 2001. View on PubMed
• 72 PCOS patients randomized to the SX fraction of maitake or to clomiphene citrate, up to 12 weeks: ovulation per patient 76.9% (20/26) versus 93.5% (29/31), not significant; per cycle 41.7% versus 69.9%, p=0.0006. Outcome measure — ovulation on ultrasound; insulin resistance not measured — Chen JT, et al. Journal of Alternative and Complementary Medicine, 2010. View on PubMed
• Grifola gargal — not maitake: 10 participants with prediabetes, 9.2 grams a day of a low-molecular-weight hot-water extract, 4 weeks — the only significant effect in humans: a drop in triglycerides; glucose reduction observed in KK-Ay and ob/ob mice (2% of the feed, 42 days) — Harada E, et al. International Journal of Medicinal Mushrooms, 2020. View on PubMed
• Mice with type 2 diabetes: an alpha-glucan from the maitake fruiting body lowered weight, fasting glucose, glycated hemoglobin and blood lipids, raised insulin and liver glycogen, and improved pancreatic histology — in rodents — Lei H, et al. International Journal of Medicinal Mushrooms, 2013. View on PubMed
• Agaricus blazei: out of 536 registered diabetics, 72 participants on gliclazide and metformin for over 6 months randomized to 1,500 mg a day of extract or placebo (cellulose), 12 weeks. HOMA-IR 3.6 (SD 2.5; n=29) versus 6.6 (SD 7.4; n=31), p=0.04; adiponectin +20.0% versus -12.0%, p<0.001 — Hsu CH, et al. Journal of Alternative and Complementary Medicine, 2007. View on PubMed
• 52 participants aged 18–65 with mild hypercholesterolemia, a shiitake beta-D-glucan-enriched blend at 10.4 grams a day (3.5 grams of fungal beta-glucan) versus placebo, 8 weeks: no significant difference on any lipid measure; IL-1β, IL-6, TNF-α, oxLDL unchanged; the microbiome changed — Morales D, et al. European Journal of Nutrition, 2021. View on PubMed
• 60 adults without cardiometabolic disease (age 46±12; BMI 28.3±2.84), a full Mediterranean-style diet with or without 84 grams of mushrooms a day, 8 weeks: fasting glucose -2.9±1.18 mg/dL versus +0.6±1.10 (p=0.034); total cholesterol -10.2±3.77 mg/dL in both groups (the diet, not the mushrooms) — Uffelman CN, et al. The Journal of Nutrition, 2023. View on PubMed

An additional safety source that is not a paper: the “Reishi Mushroom” page in the About Herbs database of Memorial Sloan Kettering Cancer Center (mskcc.org), updated 5.5.2026 — read directly.

Read next

• Medicinal mushrooms and autoimmune disease — what the research actually tested
• Medicinal mushrooms and the liver — what was measured in humans
• Medicinal mushrooms and hair loss — what was tested and what was not
• Medicinal mushrooms by goal — the hub
• Research on medicinal mushrooms — the research hub
• Lab results — beta-glucan in every extract
• How to read a test report (COA)
• Beta-glucan versus polysaccharides
• Reishi science — what was measured
• Cordyceps science — what was measured
• Do medicinal mushrooms have side effects?
• Do medicinal mushrooms interact with medications?
• Regulation of mushroom supplements in Israel
• Mushrooms and kidney health — what was measured
• Chaga — why we do not grow it
• Fresh or dried mushroom — what is the difference
• Extraction ratio — what the number means

This page is educational and does not constitute medical advice. It describes what published trials and case reports recorded — and what was not recorded; it is not a prediction of what any individual will experience, and it is not a recommendation to start, stop or combine any supplement with any medication — those decisions belong with your physician. Medicinal mushroom extracts are dietary supplements, not drugs, and are not intended to treat, prevent or balance diabetes or blood sugar levels; do not change, reduce or stop drug treatment on your own, and consult a physician before any supplement — especially if you take diabetes medication, anticoagulants or immunosuppressants, have an existing medical condition, or are pregnant or breastfeeding. *These statements have not been evaluated by the Food and Drug Administration. This product is not intended to diagnose, treat, cure, or prevent any disease.*