Pygeum

Best-in-class: Pygeum

An African plum bark extract used for BPH, with a Cochrane review that is cautiously positive.

How Pygeum actually works

Prunus africana bark contains phytosterols and pentacyclic triterpenes with anti-inflammatory activity at the prostate and effects on bladder contractility. The mechanism is inflammatory rather than anti-androgenic, which distinguishes it from the 5-AR-focused options and explains why it improves symptoms without changing hormones.

Where to get Pygeum

The evidence for Pygeum

Graded by what exists behind each claim.

✅ Clinically validated

• A Cochrane review concluded pygeum modestly improves urinary symptoms and flow, while noting the trials were small and methodologically limited.

📊 Correlative data

• Traditional African use of *Prunus africana* bark for urinary complaints, and prescribed in France and Italy for BPH for decades. The observational story now includes a conservation problem: harvesting pressure led to CITES listing, which is a genuine reason to check sourcing.

🧪 Theoretical / extrapolated benefits

• Anti-inflammatory and anti-proliferative effects on prostatic tissue are proposed.

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 Pygeum actually does

Pygeum is a lipid-soluble extract of the bark of Prunus africana, and the analytical work that has actually been done on that bark names six things worth measuring: atranorin, atraric acid, beta-sitosterol and its esters, ferulic acid and its esters, and N-butylbenzene sulfonamide Thompson 2019. Three different mechanisms have been pinned on three different members of that list, which is the first thing to understand about this product: it is not one molecule with one target.

Target one is 5-lipoxygenase, and it is measured in human cells. Human polymorphonuclear leukocytes stimulated with the calcium ionophore A23187 pour out 5-lipoxygenase products — leukotriene B4 and 5-HETE. Pygeum africanum extract suppresses that output Paubert-Braquet 1994. This is the arachidonic-acid branch that non-steroidal anti-inflammatories do not touch: ibuprofen blocks cyclooxygenase and leaves 5-LOX running. In a prostate whose stroma is chronically infiltrated, an agent that turns down leukotriene production is acting on the inflammatory component rather than on the androgen component, which is a different disease model from the one finasteride treats.

Target two is the androgen receptor itself. N-butylbenzenesulfonamide, isolated from Pygeum africanum bark, antagonizes the human androgen receptor: it blocks AR nuclear translocation and inhibits prostate cancer cell growth Papaioannou 2010. Note the step being blocked. This is not 5-alpha-reductase — it does not stop testosterone becoming dihydrotestosterone. It stops the liganded receptor getting into the nucleus to switch genes on, which is where bicalutamide works, one station further down the same line.

Target three is the sterol, and it is the one the label sells. Beta-sitosterol is the dominant phytosterol of the commercial extract and the number most products standardize to Thompson 2019. Its best-established action anywhere in the body is competition with cholesterol for incorporation into mixed micelles in the gut lumen, which is why phytosterols lower LDL. What it is proposed to do in a prostate — membrane effects, growth-factor signaling, an anti-proliferative action on stromal fibroblasts — is far less settled, and worth flagging as the weakest leg of a three-legged mechanism.

And the mechanism the outcome data actually point at is the bladder, not the prostate. Nothing in the pooled trials measured prostate volume. What moved was nocturia, residual urine and peak flow Ishani 2000 — storage and emptying, which are detrusor and outlet-resistance phenomena. A drug that shrank the gland would show up as a volume change first and a symptom change second. This one shows up the other way round, and that ordering is a clue about where it is working that nobody selling it has ever drawn attention to.

Cell, rodent, human — and where it stops

Human cells first, and unusually the human trials came before the mechanism. The 5-LOX experiment was published in 1994 Paubert-Braquet 1994 and the androgen-receptor work in 2010 Papaioannou 2010; the randomized trials they are meant to explain had already been pooled and published by 2000 Ishani 2000. That is the reverse of the usual supplement order and it is a point in this product's favor — the clinical signal was not reverse-engineered from a mechanism.

The pooled human evidence, with its real numbers. Eighteen randomized controlled trials, 1,562 men, mean study duration 64 days (range 30 to 122). Against control, men were more than twice as likely to report improvement (risk ratio 2.1, 95% CI 1.40 to 3.1). The pooled effect on urologic symptoms was −0.8 SD (95% CI −1.4 to −0.3). Nocturia fell by 19%, residual urine volume by 24%, and peak urine flow rose by 23% Ishani 2000. The Cochrane version of the same work reached the same conclusion and was careful about it: the trials were small, short, used varied doses and preparations, and rarely used standardized validated efficacy measures Wilt 2002.

The one trial that used a modern symptom score, and what it cannot tell you. Two hundred and nine men were randomized to 50 mg twice daily or 100 mg once daily for two months; 174 continued into a ten-month open extension. IPSS improved 38% in one arm and 35% in the other, quality of life 28%, peak flow by 1.63 mL/s (16%) and 2.02 mL/s (19%). Over twelve months IPSS fell from 16 to 9, a 46% drop, with peak flow up 1.65 mL/s Chatelain 1999. Read the design before the numbers: there was no placebo arm. Both groups got pygeum. A 46% fall in a symptom score over a year in men who know they are being treated is exactly the size of the placebo response that BPH trials routinely produce.

So the specific obstacle is a structural one, and it is unusual enough to name precisely. The placebo-controlled evidence mostly predates the International Prostate Symptom Score and did not use it Wilt 2002; the trial that used the IPSS properly had no placebo Chatelain 1999. The two halves of what you would need for a confident answer exist in different studies and have never been put in the same one. That is a different failure from ‘it does not work’ and a different failure from ‘it works’.

The second obstacle is time. Sixty-four days on average Ishani 2000. Benign prostatic hyperplasia is a twenty-year condition and the outcomes that matter — acute retention, surgery — are counted over years. Nothing in this file speaks to them. A two-month symptom trial answers “did the man feel better” and is silent on “did the disease change”, and those are not the same purchase.

Pygeum — which form, and does it matter

Somebody measured what is in the bottles against what is in the bark, and the result should change how you buy this. Four bark samples and seven commercial pygeum products were run by LC-APCI-MS for all six marker compounds Thompson 2019. Four findings, and each one matters.

One: the extract is a sterol concentrate. Total beta-sitosterol in bark was remarkably steady at about 680 micrograms per gram. Some commercial products exceeded 10,000 micrograms per gram Thompson 2019. That is more than a fourteen-fold enrichment. Whatever pygeum extract is, it is not ground bark, and the trials were done on the extract.

Two: the chemistry of the sterol is different, not just the amount. In bark only 33% of the beta-sitosterol was free; in the products nearly all of it was Thompson 2019. Free sterol and sterol ester behave differently in a gut lumen, so this is a change in the molecule delivered and not only in the dose.

Three: the ferulic esters are largely gone. Total ferulic acid averaged 450 micrograms per gram in bark and about a quarter of that in the products, with more than 90% present as esters in both Thompson 2019. If the traditional preparation owes anything to that fraction, the modern extract has thrown most of it away.

Four, and this is the one that should be printed on every pygeum page: no N-butylbenzene sulfonamide was found in any of the bark samples or any of the products Thompson 2019. The androgen-receptor mechanism — the most specific and most quoted thing anybody says about this plant Papaioannou 2010 — is attributed to a compound that a validated assay could not detect in the material at all. That does not make the receptor work wrong. It makes it a fact about a purified compound rather than a fact about the product, and those get conflated on every retail page.

What to buy, then. The trials used 100 mg per day of the lipidosterolic extract, given as 50 mg twice daily or 100 mg once Chatelain 1999, and that is the dose the pooled effect belongs to. Buy an extract that states a phytosterol percentage, because measured total beta-sitosterol generally did follow the labeled phytosterol content Thompson 2019 — the one label claim in that study that survived being measured. A bark powder at the same milligram number is not the same product and is not what was tested.

What would have to be true, and how you would know it was not

1. ipss, with a number that already subtracts the placebo response. From a baseline around 16, predict a fall of 3 to 6 points by twelve weeks. The uncontrolled twelve-month fall was 16 to 9 Chatelain 1999, and roughly half of a symptom-score fall in an open BPH study belongs to being in a study. Score the IPSS twice before starting, a fortnight apart, because a single baseline taken on a bad night sets you up to see improvement that is regression to the mean.

2. post-void residual, because a machine measures it and you do not. The pooled fall was 24% Ishani 2000. Predict a fall of that order on bladder ultrasound at twelve weeks. This is the single most informative endpoint on the page precisely because your expectations cannot influence it, and if the symptom score moves while residual volume does not, you have learned that you feel better rather than that you empty better.

3. Nocturia, counted honestly. Predict a 19% reduction Ishani 2000 — three trips a night becoming about 2.4, not one. Count a fortnight of nights before the first capsule. That prediction is deliberately unimpressive, and a product delivering exactly it would still be worth its price to some men.

4. uroflow, with the caveat that makes it nearly useless singly. Predict peak flow up about 1.5 to 2.0 mL/s Chatelain 1999, roughly the pooled 23% Ishani 2000. Within-person peak flow varies with voided volume and time of day by more than that, so a single before-and-after pair cannot show it; three flows on each occasion, with voided volumes recorded, or do not bother.

5. The prediction that cuts against the product. Predict total-testosterone, free-testosterone and dht are all unchanged at twelve weeks, and predict prostate volume on ultrasound is unchanged at twelve months. Nothing in the pooled trials measured a hormone or a gland volume Ishani 2000, and the receptor-level mechanism is attributed to a compound not detectable in the product Thompson 2019. If that prediction holds, pygeum is a symptom treatment and not a disease treatment — which is a real thing to be, but it means it will not prevent the retention or the operation.

6. psa, drawn for the right reason. Get a baseline before the first capsule. Not because a PSA-lowering effect is established here — the pooled trials report no PSA endpoint at all Ishani 2000, so it is neither shown nor excluded — but because a PSA value with nothing before it is uninterpretable, and men who start a supplement for urinary symptoms are exactly the men whose next PSA will need interpreting.

What nobody has tested yet

Nobody has run the trial the evidence is missing. Twelve months, placebo-controlled, IPSS as the primary endpoint. The placebo-controlled file averages 64 days and largely predates the score Ishani 2000 Wilt 2002; the twelve-month IPSS data have no placebo arm Chatelain 1999. This is not an exotic study. It is the standard BPH design, it has been standard since the 1990s, and for a compound prescribed in Europe for decades it has never been done.

Nobody knows which molecule does it. Three candidate mechanisms, three different constituents, and the most specific one is attributed to a compound absent from every sample assayed Thompson 2019 Papaioannou 2010. Fractionating the extract and taking the active fraction into a trial is the experiment. It would also settle whether the sterol concentration that products compete on is the thing that matters or a convenient number to print on a box.

Nobody has checked the 5-LOX mechanism in a living person. The leukotriene finding is in isolated human neutrophils Paubert-Braquet 1994. Urinary leukotriene E4 is a routine, non-invasive measure of whole-body 5-lipoxygenase output, it is used in asthma research every week, and it has never been drawn in a pygeum trial. One urine sample would say whether a 100 mg dose does anything to that pathway systemically.

And nobody has compared it with the drugs. There is no head-to-head trial against tamsulosin or finasteride. Men choosing between a bark extract and an alpha-blocker are choosing without a single direct comparison, and the two act on different parts of the problem — which means the honest answer may well be that the comparison is the wrong question and the combination has never been tested either.

Pygeum — its own safety story, not its category's

The tolerability number here is better than almost anything else on this shelf, and it deserves to be quoted exactly. Across 18 randomized trials the overall dropout rate was 12%: 13% on P. africanum, 11% on placebo and 8% on other controls, P = 0.4 versus placebo and P = 0.5 versus other controls Ishani 2000. Withdrawal for adverse effects was indistinguishable from placebo in 1,562 men. Mild digestive upset is the complaint that shows up, and it is the complaint placebo produces too.

The specific contraindication nobody prints is sitosterolemia. This is a phytosterol concentrate — some products carry over 10,000 micrograms of beta-sitosterol per gram Thompson 2019. People with sitosterolemia, a recessive defect of the ABCG5/ABCG8 sterol transporters, absorb plant sterols that everyone else excretes and accumulate them in tendon, skin and artery wall. It is rare and it is under-diagnosed, and a deliberately sterol-enriched supplement is precisely the wrong product for it. If there are xanthomas, premature atherosclerosis or unexplained hemolysis in the family, that is a conversation to have before starting rather than after.

The real risk on this page is not toxicity, it is delay. Lower urinary tract symptoms are the presenting complaint of benign prostatic hyperplasia and also of bladder cancer, prostate cancer, urinary infection, bladder stones and neurogenic bladder. The pooled trials ran a mean of 64 days Ishani 2000; two months of self-treatment is two months a cause is not looked for. Visible blood in the urine, an inability to pass urine, fever with flank pain, or symptoms getting rapidly worse are same-week medical problems and no bark extract changes that.

Two smaller ones. A phytosterol load competes with cholesterol absorption, so on ezetimibe or a plant-sterol spread the lipid-panel may drift and it is worth knowing why rather than attributing it to the prostate. And the sourcing question is real: Prunus africana is harvested from wild African populations under enough pressure to have put it into international trade controls, so certified cultivated material is the version to buy — a conservation argument rather than a safety one, but the only place on this page where your purchase affects somebody other than you.

Sources read for this page

• Ishani A, et al. Pygeum africanum for the treatment of patients with benign prostatic hyperplasia: a systematic review and quantitative meta-analysis. American Journal of Medicine 2000 · PMID 11099686
• Wilt T, et al. Pygeum africanum for benign prostatic hyperplasia. Cochrane Database of Systematic Reviews 2002 · PMID 11869585
• Chatelain C, et al. Comparison of once and twice daily dosage forms of Pygeum africanum extract in patients with benign prostatic hyperplasia: a randomized, double-blind study, with long-term open label extension. Urology 1999 · PMID 10475357
• Thompson RQ, et al. Chemical comparison of Prunus africana bark and pygeum products marketed for prostate health. Journal of Pharmaceutical and Biomedical Analysis 2019 · PMID 30316061
• Papaioannou M, et al. NBBS isolated from Pygeum africanum bark exhibits androgen antagonistic activity, inhibits AR nuclear translocation and prostate cancer cell growth. Investigational New Drugs 2010 · PMID 19771394
• Paubert-Braquet M, et al. Effect of Pygeum africanum extract on A23187-stimulated production of lipoxygenase metabolites from human polymorphonuclear cells. Journal of Lipid Mediators and Cell Signalling 1994 · PMID 7921787

How you would know if it worked

Pygeum — safety & side effects

Bloodwork to run alongside Pygeum

Baseline first, then again at 8–12 weeks.

The Basics — Start Here panel covers these in one order — 4 markers, $32.40 with the discount applied.

Check results you already have → · All 103 markers A–Z

Pygeum — frequently asked questions

What Pygeum is used for

Pygeum appears under 1 goal in the goal router.

Related Hormonal & Wellness supplements

← Browse the full Supplement Vault

↑ Back to on this page