Doxazosin versus tamsulosin for the treatment of male lower urinary tract symptoms associated with benign prostatic hyperplasia (LUTS/BPH): systematic literature review and network meta-analysis

Doxazosin and tamsulosin are the most frequently used α1-blockers for the treatment of male lower urinary tract symptoms. Clinical practice guidelines state that all α1-blockers have a similar clinical efficacy but differ in tolerability when used in appropriate doses. We compared the efficacy and tolerability of different doses/formulations of doxazosin (immediate release 2, 4 or 8mg; gastrointestinal therapeutic system [GITS] 4 or 8mg) versus tamsulosin (modified release 0.2, 0.4 or 0.8mg; oral controlled absorption system [OCAS] 0.4mg). PubMed, EMBASE and Cochrane databases were systematically searched until December 2024 to identify RCTs with doxazosin and/or tamsulosin. Bayesian random-effects models estimated efficacy and tolerability outcomes, including all International Prostate Symptom Score (IPSS) variations, QoL, overall adverse events (AE) and discontinuation. Tests for heterogeneity, similarity, and consistency ensured robust estimates. In total, 1,866 abstracts were identified, of which we included 40 publications with 12,201 participants. Doxazosin GITS 4mg had the largest mean (95% credible interval) improvement from baseline in total IPSS (-10.07 [-11.96, -8.25]), IPSS-QoL (-1.82 [-2.07, -1.60]) and IPSS storage-subscore (-4.26 [-4.29, -3.40]), and was statistically superior compared to most tamsulosin doses. No significant differences between different doses or formulations for the two drugs were seen for the IPSS voiding-subscore, maximum urine flow or post-void residuals. Only doxazosin GITS 4mg had significantly lower odds of discontinuation due to AE compared to tamsulosin OCAS 0.4mg. This network meta-analysis found a statistically significantly greater clinical efficacy (IPSS-T, IPSS-S, QoL improvement) for doxazosin GITS 4mg versus most tamsulosin formulations and a similar overall tolerability profile.