Comparative effectiveness of oral drug therapies for lower urinary tract symptoms due to benign prostatic hyperplasia: a systematic review and network meta-analysis

Abstract

Introduction: Lower urinary tract symptoms (LUTS) due to benign prostatic hyperplasia (BPH) are common in elder men and a number of drugs alone or combined are clinically used for this disorder. But available studies investigating the comparative effects of different drug therapies are limited. This study was aimed to compare the efficacy of different drug therapies for LUTS/BPH with network meta-analysis.

Materials and methods: An electronic search of PubMed, Cochrane Library and Embase was performed to identify randomized controlled trials (RCTs) comparing different drug therapies for LUTS/BPH within 24 weeks. Comparative effects were calculated using Aggregate Data Drug Information System. Consistency models of network meta-analysis were created and cumulative probability was used to rank different therapies.

Results: A total 66 RCTs covering seven different therapies with 29384 participants were included. We found that α-blockers (ABs) plus phosphodiesterase 5 inhibitors (PDE5-Is) ranked highest in the test of IPSS total score, storage subscore and voiding subscore. The combination therapy of ABs plus 5α-reductase inhibitors was the best for increasing maximum urinary flow rate (Qmax) with a mean difference (MD) of 1.98 (95% CI, 1.12 to 2.86) as compared to placebo. ABs plus muscarinic receptor antagonists (MRAs) ranked secondly on the reduction of IPSS storage subscore, although monotherapies including MRAs showed no effect on this aspect. Additionally, PDE5-Is alone showed great effectiveness for LUTS/BPH except Qmax.

Conclusions: Based on our novel findings, combination therapy, especially ABs plus PDE5-Is, is recommended for short-term treatment for LUTS/BPH. There was also evidence that PDE5-Is used alone was efficacious except on Qmax. Additionally, it should be cautious when using MRAs. However, further clinical studies are required for longer duration which considers more treatment outcomes such as disease progression, as well as basic research investigating mechanisms involving PDE5-Is and other pharmacologic agents alleviate the symptoms of LUTS/BPH.

Figure 1. PRISMA flowchart of literature searches and results.

RCT =  randomized controlled trials.

Figure 2. Comparison network of the included studies.

The line linked between two drug therapies means there are direct comparisons from original studies. Numbers on the line mean the count number of studies comparing every pair of treatments, which were also reflected by the width of the lines. The size of every node represents the number of randomized participants. ABs  =  α-blockers. 5ARIs  =  5α-reductase inhibitors. MRAs  =  muscarinic receptor antagonists. PDE5-Is  =  phosphodiesterase 5 inhibitors.

Figure 3. Forest plot for meta-analysis of the overall effect as measured by the IPSS total score.

The difference of IPSS total score between comparisons was calculated as mean difference (MD) and MD below 0 favors the drug therapy on the left header. If the 95% confidence intervals (CI) did not include 0, it means the difference is significant. ABs  =  α-blockers. 5ARIs  =  5α-reductase inhibitors. MRAs  =  muscarinic receptor antagonists. PDE5-Is  =  phosphodiesterase 5 inhibitors.

Figure 4. Forest plot for meta-analysis of the overall effect as measured by Qmax.

The difference of Qmax between comparisons was calculated as mean difference (MD) and MD above 0 favors the drug therapy on the left header. If the 95% confidence intervals (CI) did not include 0, it means the difference is significant. ABs  =  α-blockers. 5ARIs  =  5α-reductase inhibitors. MRAs  =  muscarinic receptor antagonists. PDE5-Is  =  phosphodiesterase 5 inhibitors.

Figure 5. Forest plot for meta-analysis of the overall effect as measured by IPSS storage subscore.

The difference of IPSS storage subscore between comparisons was calculated as mean difference (MD) and MD below 0 favors the drug therapy on the left header. If the 95% confidence intervals (CI) did not include 0, it means the difference is significant. ABs  =  α-blockers. 5ARIs  =  5α-reductase inhibitors. MRAs  =  muscarinic receptor antagonists. PDE5-Is  =  phosphodiesterase 5 inhibitors.

Figure 6. Forest plot for meta-analysis of the overall effect as measured by IPSS voiding subscore.

The difference of IPSS voiding subscore between comparisons was calculated as mean difference (MD) and MD below 0 favors the drug therapy on the left header. If the 95% confidence intervals (CI) did not include 0, it means the difference is significant. ABs  =  α-blockers. 5ARIs  =  5α-reductase inhibitors. MRAs  =  muscarinic receptor antagonists. PDE5-Is  =  phosphodiesterase 5 inhibitors.

Figure 7. Cumulative probabilities of different kinds of oral drug therapies as measured by the included outcomes.

The Bayesian approach could apply the rank probabilities of each drug therapy and the cumulative probability sum the rank probabilities to give an overall probability. Larger cumulative probability represents the better effect on the improvement of IPSS total score, Qmax, IPSS storage subscore and IPSS voiding score, which also represent the rank of the drug therapies. ABs  =  α-blockers. 5ARIs  =  5α-reductase inhibitors. MRAs  =  muscarinic receptor antagonists. PDE5-Is  =  phosphodiesterase 5 inhibitors.