Purpose: We tested the effects of a saw palmetto herbal blend in men with s
ymptomatic benign prostatic hyperplasia (BPH) via a randomized, placebo con
trolled trial.
Materials and Methods: We randomized 44 men 45 to 80 years old with symptom
atic BPH into a trial of a saw palmetto herbal blend versus placebo. End po
ints included routine clinical measures (symptom score, uroflowmetry and po
st-void residual urine volume), blood chemistry studies (prostate specific
antigen, sex hormones and multiphasic analysis), prostate volumetrics by ma
gnetic resonance imaging, and prostate biopsy for zonal tissue morphometry
and semiquantitative histology studies.
Results: Saw palmetto herbal blend and placebo groups had improved clinical
parameters with a slight advantage in the saw palmetto group (not statisti
cally significant). Neither prostate specific antigen nor prostate volume c
hanged from baseline. Prostate epithelial contraction was noted, especially
in the transition zone, where percent epithelium decreased from 17.8% at b
aseline to 10.7% after 6 months of saw palmetto herbal blend (p <0.01). His
tological studies showed that the percent of atrophic glands increased from
25.2% to 40.9% after treatment with saw palmetto herbal blend (p <0.01). T
he mechanism of action appeared to be nonhormonal but it was not identified
by tissue studies of apoptosis, cellular proliferation, angiogenesis, grow
th factors or androgen receptor expression. We noted no adverse effects of
saw palmetto herbal blend. When the study was no longer blinded, 41 men ele
cted to continue therapy in an open label extension.
Conclusions: Saw palmetto herbal blend appears to be a safe, highly desirab
le option for men with moderately symptomatic BPH. The secondary outcome me
asures of clinical effect in our study were only slightly better for saw pa
lmetto herbal blend than placebo (not statistically significant). However,
saw palmetto herbal blend therapy was associated with epithelial contractio
n, especially in the transition zone (p <0.01), indicating a possible mecha
nism of action underlying the clinical significance detected in other studi
es.