Does Saw Palmetto Work? What the Evidence Really Shows (Mostly Null)
Educational summary — not medical advice. Urinary symptoms, an enlarged prostate, or hair loss should be evaluated by a clinician — saw palmetto is not a treatment. This statement has not been evaluated by the FDA. This product is not intended to diagnose, treat, cure, or prevent any disease.
The honest answer
For an enlarged prostate (BPH), the strongest evidence says saw palmetto is no better than placebo. The placebo-controlled STEP (Bent 2006) and CAMUS (Barry 2011) trials found no benefit for urinary symptoms — even at up to 3× the standard dose, and the 32-trial Cochrane review agreed (Tacklind 2012). It doesn't lower PSA (Andriole 2013) and isn't linked to prostate-cancer risk (Bonnar-Pizzorno 2006). The European Permixon trials matched finasteride/tamsulosin but had no placebo arm — a nuance, not proof. Hair loss is preliminary only. It is well tolerated (Avins 2008).
BPH: the strong, independent trials are null
Start with the best evidence, because it points one way. The STEP trial randomized men with moderate-to-severe lower-urinary-tract symptoms to a standardized saw palmetto extract or placebo and found essentially no difference in symptom scores (the AUASI difference was a trivial 0.04) (Bent 2006). The larger CAMUS trial then did the key dose test: it escalated saw palmetto to double and then triple the standard 320 mg/day and still found no benefit over placebo (Barry 2011). The 2012 Cochrane review then pooled 32 randomized trials in more than 5,600 men and concluded saw palmetto did not improve urinary flow or symptom scores beyond placebo (Tacklind 2012). Three independent, high-quality lines of evidence — a rigorous RCT, a dose-ranging RCT, and a large systematic review — all land on the same null.
The trials, side by side
Read the null RCTs and the no-placebo comparisons as what they are — different questions with different strengths:
| Study | Design | Finding | The honest read |
|---|---|---|---|
| Bent 2006 (STEP) PMID 16467543 | Placebo-controlled RCT, BPH symptoms | No difference vs placebo (AUASI diff 0.04) | A clean null at the standard standardized dose |
| Barry 2011 (CAMUS) PMID 21954478 | Placebo-controlled dose-escalation RCT | No benefit even at up to 3× dose | Kills the "just needs a higher dose" argument |
| Tacklind 2012 (Cochrane) PMID 23235581 | Systematic review, 32 RCTs / 5,666 men | No improvement in flow or symptoms | The pooled, high-level verdict matches the RCTs |
| Carraro 1996 PMID 8876706 | RCT: Permixon vs finasteride (no placebo) | Permixon ≈ finasteride on symptoms | Equivalence with no placebo ≠ beating placebo |
| Debruyne 2002 PMID 12074791 | RCT: Permixon vs tamsulosin (no placebo) | Permixon ≈ tamsulosin on symptoms | Same limit — a nuance about extract type, not proof |
Why the Permixon trials don't overturn the null The European trials of Permixon (a specific hexane lipidosterolic extract) matched finasteride (Carraro 1996) and tamsulosin (Debruyne 2002) — but they had no placebo arm. Since those drugs themselves beat placebo only modestly, "as good as the drug" in a trial with no placebo cannot establish "better than placebo." It's fair to say the specific extract type might matter — a reasoned nuance worth noting — but it is not proof that saw palmetto works. And genuine Permixon isn't sold as a US supplement, so it's not what's in the bottles on Amazon anyway.
PSA and prostate cancer: no effect, no link
Two more honest data points. First, PSA: unlike finasteride, which lowers PSA (and can mask screening), the CAMUS PSA analysis found increasing doses of saw palmetto had no effect on serum PSA versus placebo (Andriole 2013). Second, prostate cancer: a prospective cohort of more than 35,000 men found no association between saw palmetto use and prostate-cancer risk in either direction (HR ≈ 0.95) (Bonnar-Pizzorno 2006). So saw palmetto neither lowers PSA nor changes cancer risk — useful to know precisely because it means it won't interfere with PSA-based screening the way a 5-alpha-reductase inhibitor can.
Hair loss: preliminary only
This is the one area with a flicker of positive signal, but it's weak and preliminary — not established. It rests on one tiny pilot in which 6 of 10 men on a saw palmetto blend showed improvement (Prager 2002), and one small open-label study comparing saw palmetto with finasteride for androgenetic alopecia in which finasteride clearly won — roughly 68% vs 38% responding (Rossi 2012). Both have tiny samples, no rigorous double-blind placebo control (Rossi was open-label, which inflates apparent benefit), and the head-to-head study had saw palmetto losing to a standard drug about 2:1. Treat saw palmetto for hair loss as unproven and preliminary, not a validated option.
So who is it reasonable for?
- Realistically, few people on the BPH evidence — the strong trials are null, so don't expect symptom or flow improvement (Tacklind 2012).
- If you try it anyway, use a disclosed, standardized 320 mg/day extract so you at least know what you're taking — most products don't disclose it (the standardization trap).
- Get evaluated for real symptoms. Urinary changes, an enlarged prostate, or hair loss deserve a clinician's assessment and treatments with actual evidence — saw palmetto doesn't lower PSA and isn't a substitute for care.
Skip or seek medical advice first if you: take anticoagulants (theoretical bleeding caution); have a hormone-sensitive condition (theoretical hormonal caution); or are relying on it instead of an evaluation for prostate or urinary symptoms.
Frequently asked questions
Does saw palmetto work for BPH?
No — the strong evidence is null. STEP (Bent 2006) and CAMUS (Barry 2011) found no benefit over placebo even at up to 3× dose, and the 32-trial Cochrane review agreed (Tacklind 2012).
What about the Permixon trials vs finasteride?
Permixon matched finasteride (Carraro 1996) and tamsulosin (Debruyne 2002) but with no placebo arm, so it can't show it beats placebo. A reasoned nuance about extract type, not proof — and Permixon isn't a US supplement.
Does it lower PSA or affect cancer risk?
No to both. No PSA effect vs placebo (Andriole 2013), and no association with prostate-cancer risk in a 35,000-man cohort (HR ~0.95, Bonnar-Pizzorno 2006).
Is it proven for hair loss?
No — preliminary only. A tiny pilot (Prager 2002) and a small open-label study where finasteride won ~68% vs 38% (Rossi 2012). Unproven, not established.
Is it safe?
Well tolerated, side effects like placebo (Avins 2008). Prudence cautions (not trial harms): theoretical bleeding/anticoagulant and hormone-sensitive concerns — ask a clinician.
Related
- Saw palmetto: the honest evidence & label trap
- Saw palmetto dosage guide — the 320 mg standardized dose (and why it's still null)
- Standardized extract vs berry powder — why you can't verify most products
- Best saw palmetto — ranked by disclosed standardization
Sources
- Bent S, et al. "Saw palmetto for benign prostatic hyperplasia." N Engl J Med. 2006. PMID: 16467543 (STEP RCT — null).
- Barry MJ, et al. "Effect of increasing doses of saw palmetto extract on lower urinary tract symptoms: a randomized trial." JAMA. 2011. PMID: 21954478 (CAMUS — null at up to 3× dose).
- Tacklind J, et al. "Serenoa repens for benign prostatic hyperplasia." Cochrane Database Syst Rev. 2012. PMID: 23235581 (32 RCTs / 5,666 men).
- Carraro JC, et al. "Comparison of phytotherapy (Permixon) with finasteride in the treatment of benign prostate hyperplasia." Prostate. 1996. PMID: 8876706 (no placebo arm).
- Debruyne F, et al. "Comparison of a phytotherapeutic agent (Permixon) with an alpha-blocker (Tamsulosin) in the treatment of benign prostatic hyperplasia." Eur Urol. 2002. PMID: 12074791 (no placebo arm).
- Andriole GL, et al. "Effect of increasing doses of saw palmetto extract on serum prostate-specific antigen." J Urol. 2013. PMID: 23253958 (no PSA effect).
- Bonnar-Pizzorno RM, et al. "Saw palmetto supplement use and prostate cancer risk." Nutr Cancer. 2006. PMID: 16965237 (no association, HR ~0.95).
- Prager N, et al. "A randomized, double-blind, placebo-controlled trial to determine the effectiveness of botanically derived inhibitors of 5-alpha-reductase in the treatment of androgenetic alopecia." J Altern Complement Med. 2002. PMID: 12006122 (hair-loss pilot, tiny).
- Rossi A, et al. "Comparitive effectiveness of finasteride vs Serenoa repens in male androgenetic alopecia." Int J Immunopathol Pharmacol. 2012. PMID: 23298508 (open-label; finasteride outperformed).
- Avins AL, et al. "A detailed safety assessment of a saw palmetto extract." Complement Ther Med. 2008. PMID: 18534327 (well tolerated).