Upcoming event

Randomised clinical trial of prostatic artery embolisation versus a sham procedure for benign prostatic hyperplasia

Publication: European Urology, December 2020


Prostatic artery embolisation (PAE) has been associated with an improvement of lower urinary tract symptoms associated with benign prostatic hyperplasia (LUTS/BPH), but conclusive evidence of efficacy from randomised controlled clinical trials has been lacking.


To assess the safety and efficacy of PAE compared with a sham procedure in the treatment of LUTS/BPH.

Design, setting, and participants

A randomised, single-blind, sham-controlled superiority clinical trial was conducted in 80 males ≥45 yr with severe LUTS/BPH refractory to medical treatment from 2014 to 2019 in a private clinic, with efficacy assessments at 6 and 12 mo after randomisation. One patient in the PAE group and three in the sham group did not complete the study.


Patients were randomised 1:1 upon successful catheterisation of a prostatic artery to either PAE or a sham PAE procedure without embolisation. After 6 mo, all 38 patients randomised to the sham group who completed the single-blind period underwent PAE, and both groups completed a 6-mo open period.

Outcome measurements and statistical analysis

An intention-to-treat analysis of all randomised patients was performed. The coprimary outcomes were the change from baseline to 6 mo in the International Prostate Symptom Score (IPSS) and the quality of life (QoL) score at 6 mo, analysed with analysis of covariance and t test, respectively.

Results and limitations

Mean age was 63.8 ± 6.0 yr, baseline IPSS 26.4 ± 3.87, and QoL score 4.43 ± 0.52. At 6 mo, patients in the PAE arm had a greater improvement in IPSS, with a difference in the change from baseline of 13.2 (95% confidence interval [CI] 10.2–16.2, p < 0.0001), and a better QoL score at 6 mo (difference: 2.13; 95% CI 1.57–2.68, p < 0.0001) than the patients in the sham arm. The improvements in IPSS and QoL in the sham group 6 mo after they performed PAE were, respectively, 13.6 ± 9.19 (p < 0.0001) and 2.05 ± 1.71 (p < 0.0001). Adverse events occurred in 14 (35.0%) patients after PAE and in 13 (32.5%) after sham, with one serious adverse event in the sham group during the open period. No treatment failures occurred. Limitations include a single-centre trial, only severe LUTS/BPH, and follow-up limited to 12 mo.


The improvements in subjective and objective variables after PAE are far superior from those due to the placebo effect.